No CrossRef data available.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 01 January 2025
Australian enterprise agreements are intended to be simple documents that are easy for workers to understand and straightforward for businesses to apply. This article argues that simplicity is undermined when agreements contain indeterminate language and ambiguous provisions, since these features make it difficult to pin down the precise meaning of agreements. It presents findings from an empirical study that examines two forms of uncertainty in enterprise agreements: first, the use of vague, open-textured terms such as ‘reasonable’ and ‘practicable’; and, second, the inclusion of particular types of poorly drafted provisions that are difficult to interpret. The findings show that while the use of indeterminate terms increased over time, there was a substantial decline in the use of particular types of ambiguous provisions. The study also suggests that these outcomes have been influenced by two factors, legislative change and union involvement in bargaining. The findings add to our understanding of the nature of legal complexity and will provide the basis for a nuanced analysis of future proposals to reform the legislative framework for agreement-making.
My thanks go to Richard Mitchell, Anthony Forsyth, Chris Arup, Richard Johnstone and Andrew Stewart for their close engagement with the ideas presented in this paper and their insightful responses to earlier material that provided the foundation for this paper. I am also grateful to the anonymous referees for their helpful insights.
1 Department of Employment, ‘Trends in Enterprise Bargaining September 2014’ (Commonwealth of Australia, 20 January 2015) 7.
2 See pt II below. See also Sutherland, Carolyn, ‘Mapping Complexity in Australian Enterprise Agreements: A Multi-Dimensional Approach’ (2013) 26 Australian Journal of Labour Law 50, 52–4.Google Scholar
3 The only scholarship which addresses the complexity of Australian industrial instruments in any depth is Bray, Mark and Waring, Peter, ‘“Complexity” and “Congruence” in Australian Labour Regulation’ (2005) 47 Journal of Industrial Relations 1CrossRefGoogle Scholar; and Fetter, Joel and Mitchell, Richard, ‘The Legal Complexity of Workplace Regulation and its Impact Upon Functional Flexibility in Australian Workplaces’ (2004) 17 Australian Journal of Labour Law 276Google Scholar. For an analysis of the complexity of Australian workplace relations regulation more broadly, see Stewart, Andrew, ‘A Simple Plan for Reform? The Problem of Complexity in Workplace Regulation’ (2005) 31 Australian Bulletin of Labour 210.Google Scholar
4 For an overview of the larger study, see Sutherland, ‘Mapping Complexity in Australian Enterprise Agreements’, above n 2; for an analysis of the findings in relation to technicality and ‘layering’ (respectively), see Sutherland, Carolyn, ‘The Elusive Quest for Simplicity: Measuring and Assessing the Readability of Enterprise Agreements, 1993 to 2011’ (2013) 35 Sydney Law Review 349Google Scholar; and Sutherland, Carolyn, ‘Enterprise Bargaining as a Tool to Reduce Regulatory Layering: A Content Analysis Study’ (2014) 42 Federal Law Review 559.CrossRefGoogle Scholar
5 For a discussion of simplicity as a goal that has been used to justify legislative reforms to the framework for enterprise bargaining in Australia, see below nn 25–27 and accompanying text.
6 Schuck, Peter H, ‘Legal Complexity: Some Causes, Consequences, and Cures’ (1992) 42 Duke Law Journal 1, 22CrossRefGoogle Scholar. For further analysis of the nature of legal uncertainty, see Hirsch, Werner Z, ‘Reducing Law's Uncertainty and Complexity’ (1974) 21 UCLA Law Review 1233Google Scholar; D’Amato, Anthony, ‘Legal Uncertainty’ (1983) 71 California Law Review 1CrossRefGoogle Scholar; Miller, John, ‘Indeterminacy, Complexity, and Fairness: Justifying Rule Simplification in the Law of Taxation’ (1993) 68 Washington Law Review 1Google Scholar; Dodson, Scott, ‘The Complexity of Jurisdictional Clarity’ (2011) 97 Virginia Law Review 1.Google Scholar
7 Schuck, above n 6, 4.
8 Ibid 6.
9 For a discussion of the relationship between complexity and non-compliance, see Schuck, above n 6, 23–4.
10 Knox, Angela, ‘Better the Devil you Know? An Analysis of Employers’ Bargaining Preferences in the Australian Hotel Industry’ (2009) 51 Journal of Industrial Relations 25, 40CrossRefGoogle Scholar. In a legislative context, Schuck identifies the reluctance to engage in legislative reform as a ‘governance cost’ associated with complexity: Schuck, above n 6, 20–1.
11 See, eg, Ehrlich, Isaac and Posner, Richard, ‘An Economic Analysis of Legal Rulemaking’ (1974) 3 Journal of Legal Studies 257CrossRefGoogle Scholar; Kaplow, Louis, ‘A Model of the Optimal Complexity of Legal Rules’ (1995) 11 Journal of Law, Economics, and Organization 150Google Scholar; Diver, Colin, ‘The Optimal Precision of Administrative Rules’ (1983) 93 Yale Law Journal 65CrossRefGoogle Scholar. The complexity or ‘complication’ arising from precisely defined rules is particularly evident in the context of taxation laws: see Paul, Deborah, ‘The Sources of Tax Complexity: How Much Simplicity can Fundamental Tax Reform Achieve?’ (1997) 76 North Carolina Law Review 151.Google Scholar
12 See Diver, above n 11, 72–3.
13 [2014] FCA 1330 (11 December 2014).
14 Ibid [13].
15 Ibid [26].
16 Ibid [46].
17 Ibid [47]–[48].
18 See above n 6 and accompanying text.
19 Fred Hilmer et al, Enterprise-Based Bargaining Units: A Better Way of Working (Business Council of Australia, Melbourne, 1989).
20 See Macdonald, Duncan, Campbell, Iain and Burgess, John, ‘Ten Years of Enterprise Bargaining in Australia: An Introduction’ (2001) 12 Labour and Industry 1CrossRefGoogle Scholar; Nick Wailes and Russell Lansbury, ‘Flexibility vs Collective bargaining?: Australia during the 1980s and 1990s’ (Working Paper No 49, Australian Centre for Industrial Relations Research and Teaching, 1997) 41.
21 Hilmer et al, above n 19, 126.
22 McCallum, Ronald ‘Enhancing Federal Enterprise Bargaining: The Industrial Relations (Legislation Amendment) Act 1992 (Cth)’ (1993) 6 Australian Journal of Labour Law 63, 63–4.Google Scholar
23 The ‘WorkChoices’ reforms are the amendments to the WR Act made by the Workplace Relations Amendment (Work Choices) Act 2005 (Cth).
24 In relation to the WR Act and ‘WorkChoices’ amendments, see Richard Mitchell et al, ‘Assessing the Impact of Employment Legislation: The Coalition Government's Labour Law Program 1996–2007’ (2010) Australian Journal of Labour Law 274, 275. In relation to the FW Act, see Sutherland, Caroly, ‘Making the “BOOT” fit: Reforms to Agreement-making from Work Choices to Fair Work’, in Forsyth, Anthony and Stewart, Andrew (eds), Fair Work: The New Workplace Laws and the Work Choices Legacy (Federation Press, 2009) 99.Google Scholar
25 Reith, Peter, ‘Real Reform – The Government's Industrial Relations Agenda’ (1996) 8(3) Sydney Papers 1.Google Scholar
26 Kevin Andrews, ‘Workchoices: A New Workplace Relations System’ (Speech to the Australian Financial Review Conference, Melbourne, 25 October 2005).
27 ALP National Platform and Constitution 2007 (adopted by the 44th national conference, 27–29 April 2007) cl 26.
28 See FW Act s 134(1)(g). Earlier versions of this provision required awards ‘to be framed so as best to express the decision of the Commission and avoid unnecessary technicalities’: see IR Act and WR Act s 144; Work Choices Act s 570. There are similar examples in state legislation: for example, section 93(1) of the Fair Work Act 1994 (SA) requires that ‘[a]n award must be expressed in plain English and must avoid unnecessary technicality and excessive detail’.
29 FW Act s 601(3). This provision was also contained in s 143(2A) of the WR Act, but there was no equivalent provision in the IR Act or the Work Choices Act.
30 Australian Public Service Commission, ‘Australian Government Public Sector Workplace Bargaining Policy’ (Commonwealth of Australia, March 2014); Department of Education, Employment and Workplace Relations, ‘Australian Government Employment Bargaining Framework – Supporting Guidance’ (Commonwealth of Australia, January 2011). Earlier guidelines imposed by the Howard Coalition Government similarly required public sector agreements to be ‘simple’: Commonwealth of Australia, Workplace Relations Policy Parameters for Agreement Making in the Australian Public Service (April 2006).
31 Australian Public Service Commission, above n 30, 29.
32 Peter Reith, ‘Flexibilities Available in Agreement-Making’ (Ministerial Discussion Paper, Department of Employment, Workplace Relations and Small Business, May 1998) 25.
33 IR Act s 3(a).
34 WR Act s 3. See also Work Choices Act s 3; FW Act s 3.
35 See IR Act s 3(e); WR Act s 3(e); Work Choices Act s 3; FW Act s 3(e).
36 See above pt II.
37 Fetter and Mitchell, above n 3.
38 Commonwealth, Parliamentary Debates, House of Representatives, 26 February 1992, 264–70 (Paul Keating). The importance of employee protection was also emphasised by the Minister for Industrial Relations in the Second Reading Speech to the Industrial Relations Reform Bill 1993: Commonwealth, Parliamentary Debates, House of Representatives, 28 October 1993, 2777–8 (Laurence Brereton). The support of the Australian Council of Trade Unions for enterprise bargaining was conditional on the maintenance of the safety net: Martin Ferguson, ‘The Keating Agenda for Enterprise Agreements: The Impact on the Safety Net and Awards’ (Speech, Sydney, 26 July 1993) 4.
39 For a discussion of the effectiveness of the ‘no disadvantage test’, and its evolution over time, see Merlo, Omar, ‘Flexibility and Stretching Rights: The No Disadvantage Test in Enterprise Bargaining’ (2000) 13 Australian Journal of Labour Law 207Google Scholar; Forsyth, Anthony and Sutherland, Carolyn, ‘Collective Labour Relations Under Siege: The Work Choices Legislation and Collective Bargaining’ (2006) 19 Australian Journal of Labour Law 183, 184–6.Google Scholar
40 Forsyth and Sutherland, above n 39, 188–91. The Coalition Government subsequently introduced a watered-down version of the ‘no disadvantage test’ known as the ‘fairness test’ which operated from July 2007 to March 2008: for further detail, see Sutherland, Carolyn, ‘All Stitched Up? The 2007 Amendments to the Safety Net’ (2007) 20 Australian Journal of Labour Law 245.Google Scholar
41 These amendments were contained in the Workplace Relations Amendment (Transition to Forward with Fairness) Act 2008 (Cth). For further detail, see Sutherland, Carolyn, ‘First Steps Forward (with Fairness): A Preliminary Examination of the Transition Legislation’ (2008) 21 Australian Journal of Labour Law 137.Google Scholar
42 See Sutherland, ‘Making the “BOOT” fit’, above n 24, 99.
43 The exception was employees who were covered by an enterprise agreement made under the WR Act. These employees were only entitled to the conditions in the Standard where the enterprise agreement did not deal with any of the conditions specified in the Standard: Work Choices Act sch 7 cl 30.
44 Work Choices Act pt 7. See further Owens, Rosemary, ‘Working Precariously: The Safety Net after the Work Choices Act’ (2006) 19 Australian Journal of Labour Law 161, 162–9Google Scholar; Fenwick, Colin, ‘How Low Can You Go? Minimum Working Conditions under Australia's New Labour Laws’ (2006) 16 The Economic and Labour Relations Review 85, 104–10.CrossRefGoogle Scholar
45 Sutherland, Carolyn, ‘Industrial Legislation in 2008’ (2009) 51 Journal of Industrial Relations 297, 303–4CrossRefGoogle Scholar; Jill Murray and Owens, Rosemary, ‘The Safety Net: Labour Standards in the New Era’ in Forsyth, Anthony and Stewart, Andrew (eds), Fair Work: The New Workplace Laws and the Work Choices Legacy (Federation Press, 2009) 40, 43–50.Google Scholar
46 Work Choices Act ss 172(2), 173.
47 The FW Act makes explicit provision for this possibility: s 55(4)–(5).
48 Work Choices Act s 226(1); FW Act s 62.
49 FW Act s 65.
50 Work Choices Act s 353(1). The model dispute resolution is not automatically included in agreements under the FW Act. However, the Fair Work Commission must ensure the agreement contains a dispute resolution term at the time the agreement is approved: FW Act ss 186(6), 737.
51 FW Act s 205.
52 Ibid ss 202–4.
53 Work Choices Act s 695; Fair Work Regulations 2009 (Cth) reg 6.01, sch 6.1.
54 Fair Work Regulations 2009 (Cth) reg 2.09, sch 2.3.
55 Ibid reg 2.08, sch 2.2.
56 These rules are set out in more detail in Sutherland, ‘Enterprise Bargaining as a Tool to Reduce Regulatory Layering’, above n 4, 565–8.
57 Ibid.
58 FW Act ss 57–8. The exception is outworker terms in an award, which are not displaced by the agreement: FW Act s 57A.
59 In the higher education sector, the Coalition government's Higher Education Workplace Relations Requirements also contributed to these outcomes between 2005 and 2008: see Sutherland, ‘Enterprise Bargaining as a Tool to Reduce Regulatory Layering’, above n 4.
60 Ibid.
61 For further detail about the content analysis methods used in the study, see Sutherland, ‘Mapping Complexity’, above n 2, 56–9.
62 ‘Reasonableness’ is used as a typical example of an indeterminate term in the article by Schuck that provides the foundation for the present study: see Schuck, above n 6, 4.
63 The software was developed at Harvard University, and is available under a public licence at <http://sourceforge.net/projects/yoshikoder/>.
64 These terms are ‘stemmed’ for the purpose of the frequency analysis so that all versions of the word are detected. For example, the term ‘*reasonabl*’ will pick up the words ‘unreasonable’, ‘reasonably’, ‘reasonableness’ and so on.
65 For further detail about the method used to develop and verify the list used for coding, see Sutherland, ‘The Elusive Quest’, above n 4, 361.
66 Stone, Philip et al, The General Inquirer: A Computer Approach to Content Analysis (Massachusetts Institute of Technology Press, 1966) 8Google Scholar; see further Neuendorf, Kimberly, The Content Analysis Guidebook (Sage Publications, 2002) 51.Google Scholar
67 Fetter and Mitchell, above n 3. Although the study examines Australian Workplace Agreements (AWAs) rather than enterprise agreements, the authors note that the problems of legal complexity that arise from the interrelationship between the agreement and other instruments are equally applicable in the context of enterprise agreements: at 282, 305.
68 Ibid 296. The potential for problems of interpretation to arise from the ‘layering’ of industrial instruments was also identified by Bray and Waring, above n 3.
69 Fetter and Mitchell, above n 3, 294–8.
70 The relevant coding rules are set out in detail in Carolyn Sutherland, ‘Complexity and Simplicity in Australian Enterprise Agreements: Coding Framework, List of Agreements and Chi Square Tables’, Workplace and Corporate Law Research Group Working Paper No 20 (September 2013).
71 See Sutherland, ‘The Elusive Quest for Simplicity’, above n 4, 361–2.
72 The first enterprise agreement made in the higher education sector commenced operating on 28 April 1993 and the sampling frame includes all agreements made by higher education institutions registered up to 30 September 2011. The first enterprise agreement made in the fast food sector commenced operating on 3 May 1994 and the sample includes agreements made up to and including 8 March 2011.
73 For further detail about the sampling process, see Sutherland, ‘The Elusive Quest for Simplicity’, above n 4, 361–4.
74 The ‘Work Choices period’ is used as short-hand to describe the period commencing with the Work Choices Act and includes the subsequent amendments made by the Stronger Safety Net Act 2007 (Cth) (commencing on 7 May 2007) and the Workplace Relations Amendment (Transition to Forward with Fairness) Act 2008 (Cth) (commencing on 28 March 2008).
75 Schuck, above n 6, 22.
76 This test evaluates whether there is a statistically significant relationship between two variables.
77 A statistically significant relationship means that the association between the variables (dates of commencement and proportion of indeterminate words) does not arise by chance alone.
78 For the purposes of performing a Chi-square test, the percentages of indeterminate words were grouped in three categories. For the Higher Education Agreements Database, the categories were: less than 0.6 per cent; 0.6 to 0.7 per cent; and 0.8 per cent or more; for Fast Food Agreements Database 1, the categories were: less than 0.47 per cent; 0.47 to 0.63 per cent; and 0.64 per cent or more.
79 For the Higher Education Agreements Database: x2(6) = 41.682, p < 0.001; for Fast Food Agreements Database 1, x2(4) = 107.973, p < 0.001. In these results, the p-value represents the probability of observing these variations in the proportions of indeterminate words in different legislative periods purely by chance. A p-value of 0.01 indicates that the deviation from the expected pattern would have occurred by chance in no more than one percent of cases. Any result below 0.05 is considered to be statistically significant.
80 Work Choices Act pt 7; FW Act pt 2–2.
81 FW Act s 205; Fair Work Regulations 2009 (Cth) reg 2.09 sch 2.3.
82 For the purposes of performing a Chi-square test, the percentages of indeterminate words were grouped in three categories: less than 0.47 per cent; 0.47 to 0.63 per cent; and 0.64 per cent or more.
83 This refers to whether or not the agreement is considered a ‘union’ or a ‘non-union’ agreement. When the enterprise bargaining system first came into effect, all enterprise agreements were made with unions, whereas in March 1994 the Industrial Relations Reform Act 1993 (Cth) introduced a non-union agreement option, which was regulated by specific rules in the legislation. The distinction between union and non-union agreements then remained in place until the enactment of the FW Act. From 1 July 2009, the FW Act abolished any formal distinction between union and non-union agreements. However, unions were permitted to be involved in the bargaining process and could apply to Fair Work Australia (now the Fair Work Commission) to be bound by an agreement. For the sake of the empirical study, agreements made under the FW Act are taken to be union agreements where one or more unions are bound by the agreement, whereas under earlier legislation, a union needed to be a party to an agreement for it to be classified as a union agreement.
84 x2(2) = 36.683, p < 0.001.
85 A further 35 per cent of agreements were based on templates for which no source could be identified, and the remaining 19 per cent of agreements did not appear to be based on a template.
86 A Chi-square test was performed to test whether there is a statistically significant relationship between the mean percentage of indeterminate words in fast food agreements and the template sources of agreements. This analysis revealed the relationship to be statistically significant: x2(8) = 434.082, p < 0.001.
87 See the discussion of the ‘no disadvantage’ test above in pt III.A.
88 Readability was assessed in the larger study using a combination of readability tests (the Flesch Reading Ease Formula and the Flesch-Kincaid Grade Level Formula) and a frequency analysis of the use of industrial relations and legal jargon: for further detail, see Sutherland, ‘The Elusive Quest for Simplicity’, above n 4.
89 See Fetter and Mitchell, above n 3, 296.
90 See Sutherland, ‘Enterprise Bargaining as a Tool to Reduce Regulatory Layering’, above n 4.
91 See Fetter and Mitchell, above n 3, 295.
92 For the Higher Education Agreements Database, x2(3) = 137.891, p < 0.001; for Fast Food Agreements Database 2, x2(2) = 10.626, p < 0.01.
93 Sutherland, ‘Enterprise Bargaining as a Tool to Reduce Regulatory Layering’, above n 4.
94 Ibid.
95 Fetter and Mitchell, above n 3, 296.
96 For the Higher Education Agreements Database, x2(3) = 57.418, p < 0.001.
97 WR Act ss 170LY(1), 170LZ(1). It should be noted that, even where this interaction rule applied, there was the potential for confusion to arise about whether the enterprise agreement was inconsistent with an otherwise applicable award.
98 For the Higher Education Agreements Database, x2(3) = 56.466, p < 0.001; for Fast Food Agreements Database 2, x2(2) = 21.754, p < 0.001.
99 For the Higher Education Agreements Database, x2(3) = 107.467, p < 0.001.
100 However, the increase in the extent to which fast food agreements displace earlier agreements (and the consequential reduction in the use of this particular ambiguous provision in those agreements) should not be attributed to legislative change, since there were too few agreements made in the IR Act period to support this conclusion. For subsequent legislative periods, the relationship was not found to be statistically significant.
101 Sutherland, ‘Enterprise Bargaining as a Tool to Reduce Regulatory Layering’, above n 4.
102 Otherwise, the relationship between union involvement in bargaining and the use of ambiguous provisions was statistically significant. In relation to clauses that create ambiguity about legal ordering between the agreement and awards, x2(1) = 34.345, p < 0.001; in relation to clauses that create ambiguity about the legal status of earlier agreements, x2(1) = 36.101, p < 0.001; and in relation to clauses that create ambiguity about legal ordering between the agreement and earlier agreements, x2(1) = 14.262, p < 0.001.
103 Sutherland, ‘Enterprise Bargaining as a Tool to Reduce Regulatory Layering’, above n 4.