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

Enterprise bargaining and economic performance

Industrial relations systems span a spectrum from centralised to decentralised and Australia is no exception, having occupied different locations on that spectrum at different times in the past. Since the early 1990s such movement has taken Australia much further towards the decentralised end, though clearly elements of centralisation—in the form of the FW Commission and the award system—still play a crucial role. When it comes to economic outcomes, the international literature suggests several conclusions. As far as wages dispersion is concerned, decentralised systems induce much greater inequality. Summarising the literature Joe Isaacs observed in 2005: ‘On income distribution there appears to be more robust evidence that centralised systems, especially when accompanied by high union density, generally deliver lower wage dispersion’.¹²¹
When it came to economic performance, Joe Isaacs suggested that the evidence was less clear-cut. An influential study in the 1980s suggested that the best outcomes were associated with either highly centralised or highly decentralised systems, and those in the middle of the spectrum faired worst, particularly in terms of wage-cost inflation (the ‘hump-shaped hypothesis’ of Calmfors and Driffill).¹²² However, as Isaacs pointed out, more recent studies have suggested that this conclusion is based on a number of assumptions, many of which do not apply in the Australian setting. For example, criticisms of the intermediate position on the spectrum—which is where Australia was sometimes placed—assumed a closed economy. If there is exposure to international competition—as is the case with the Australian economy in the 21st century—then such competition constrains the ability of firms to pass on increased wage costs. When it comes to the more centralised end of the spectrum—such as incomes policies like the Indexation (1975–81) and the Accord (1983–91)—good outcomes at this end also required other elements to be at work, such as economy-wide coordination of wage fixing and an ability to enforce outcomes.¹²³

In concluding his review, Isaacs observed that ‘Econometric studies suggest that the Australian centralised wage-fixing episodes appear to have been successful in maintaining greater wage moderation, price stability and industrial peace’.¹²⁴ The drivers behind the move towards the decentralised end of the spectrum—towards industrial relations deregulation—were not related to economic performance, but to other industrial and political factors.

In a prescient warning in 1990 Frenkel and Peetz suggested that continuing along the path of award restructuring within a centralised framework would yield better outcomes for the economy. Instead the Business Council of Australia’s agenda for enterprise bargaining largely became the basis for the new decentralised system. As Frenkel and Peetz cautioned, the BCA agenda diverted attention away from the many factors which were likely to determine the future competitiveness of Australian firms:

the focus upon industrial relations structures diverts attention from many of the crucial issues that will determine the competitiveness of Australian enterprises: inadequacies in training, research and development, export orientation, product innovation, and time horizons.¹²⁵

In a 2001 study of enterprise bargaining Tseng and Wooden used the ABS business longitudinal study to assess productivity outcomes. They concluded that there was a correlation between firms with registered agreements and increased workplace-level productivity but they concluded that it ‘proved impossible to establish a direct causal relationship between the introduction of enterprise agreements and subsequent productivity growth’.¹²⁶

The assessment by Ron Callus of the move to enterprise bargaining suggested that economic performance was not enhanced by a more decentralised system. The complexity of the award system, with many thousands in existence in the early 1990s, was not resolved by the introduction of enterprise bargaining because most Enterprise Bargaining Agreements (EBAs) were written to operate in conjunction with the award. Moreover, where employers had multi-site operations, they ended up with more EBAs, and more complexity! As for the negotiating process, large amounts of management time became absorbed in the various rounds of bargaining, particularly in the development of the early EBAs. When it came to determining wages, there was no blossoming of enterprise-driven and productivity-determined outcomes. Rather, most enterprises simply sought to stay with a narrow band of annual wage increases which matched the norm within their industry sector. Finally, those areas of workplace operations which were most likely to result in productivity gains—such as work organisation, training and performance management, work-life policies—were increasingly relegated to human resource manuals rather than to EBAs. As Callus argued, employers would have gained more by working towards a system of industry-focussed awards rather than going down the path of EBAs.¹²⁷ By extension perhaps the broader point to take from this is that a hybrid model- involving less collective agreements which each cover more entities within an industry or labour supply chain (a model for which we advocate elsewhere in this submission) - is a desirable one.

A 2013 case-study assessment of enterprise bargaining echoed Callus’s sentiments:

While the Business Council of Australia pointed to a panoply of advantages, we look at the other side of the argument. Decentralisation may be inefficient at a number of levels. First, there is the procedure of bargaining and the resources, expertise and time that is required. Second, there are the outcomes of bargaining, where despite a more individual focus, in many cases, outcomes demonstrate very little variation across enterprises. Finally, there are the collateral consequences of bargaining: conflict, reduced trust and disruption.¹²⁸

In his assessment of industrial relations and economic performance David Peetz drew the important distinction between individual workplace productivity and broader national outcomes:

There is some evidence that industrial relations policies that enhance fairness enhance economic performance. However … the effects are conditional … in any specific workplace, industrial relations can make a difference to productivity. The decisions management makes, and the relationship it has with employees and unions, will shape what happens in the workplace and can have a noticeable effect on productivity.

That is not the same as saying, though, that if IR policy is altered at the national level, it is going to have a widespread or noticeable impact on productivity. It is what happens at the workplace that matters …

And his overall conclusion echoed that of Jeff Borland’s comments on bargaining power:

Many seek a holy grail in employment or industrial relations policy that is going to give a magic boost to the economy. But there is none—certainly not to be found in policies that aim to shift the balance of power in industrial relations one way or the other. ¹²⁹