A. Introduction and Summary

57. The Employment Relations Act (ERA) was passed into law in August 2000 following a contentious national debate. The new legislation is intended to level the playing field in the area of employment relations, which the newly elected Labor-Alliance Coalition Government felt had become unbalanced since the passage of the Employment Contracts Act (ECA) in May 1991. On the other hand, the ERA is seen—particularly by the business community—as a reversal of the trend toward labor market liberalization which culminated with the passage of the ECA. The ECA effectively ended nearly a century of centralized industrial relations in New Zealand, including by stripping unions of the legislated advantages they had enjoyed under previous labor market regimes.

58. Given these strongly held views and the uncertainty surrounding the effects of the ERA going forward, it seems to be an appropriate time to take a wider look at labor market reform efforts in New Zealand with a view to identifying areas to watch as the ramifications of the ERA unfold. This paper suggests that an assessment of the ERA would depend both on the evolution of employment growth, productivity, wage dispersion and contract structure—which can be observed relatively quickly—as well as on other effects, such as the evolution of case law in respect of the notion of “fair bargaining,” which will take years to unfold. Even with the passage of time, however, it will likely be difficult, as in the case of the ECA, to make a definitive judgement on the effects of the ERA since: (i) the effects of the legislation can never be fully separated from events elsewhere in the labor market and the macroeconomy and (ii) the objectives of the two pieces of legislation differ, so that any comparison of the extent to which the objectives are achieved may not be comparable.

59. The remainder of the paper is structured as follows. Section B provides a brief history of industrial relations in New Zealand leading up the passage of the ECA in 1991. Section C looks at the ECA and attempts to evaluate the success of that legislation in terms of achieving its stated goals. Section D presents the main features of the ERA and its probable implications for industrial relations and macroeconomic outcomes in New Zealand, which suggests areas to watch as its effects unfold. Section E concludes.

B. A Brief History of Industrial Relations in New Zealand

60. Centralized industrial relations have a long history in New Zealand, beginning with the Industrial Conciliation and Arbitration Act of 1894, which was enacted following a period of major industrial disputes. The main features of the resulting system were the requirement that registered unions and employers negotiate with each other, and that compulsory arbitration by the state was required if these negotiations failed. State arbitration resulted in an “award” that prescribed employment conditions for all workers in an occupation group whether or not they were union members. Underlying this system was the supposition that employer-worker relations were adversarial, and that heavy state involvement in the labor market was necessary to minimize industrial disruption.

61. In general, this framework benefited important constituencies in both labor and capital (Bray and Walsh, 1998). Unions benefited greatly from the system in light of the statutory requirement on employers to negotiate, compulsory arbitration, the application of awards to all workers and equity considerations flowing from centralized bargaining. In return, however, unions were constrained in their activities, the most important of these being a prohibition on strikes, under threat of losing their registration. A minority of unions—particularly those that were ideologically radical and/or in particularly powerful bargaining positions (especially those in capital-intensive industries)—were generally opposed to this system. Over time, many employers began to favor the system as well, as it served to control militant unions and limit wage growth, and competition.

62. In the post-war period, the industrial relations system described above co-existed with a full employment boom, buttressed by import protection (inter alia, through tariffs and licensing requirements) and suppressed domestic product market competition. In this period, New Zealand enjoyed one of the world’s highest standards of living, based largely on pastoral farming and resource exploitation. Technological innovation—spurred by recurring labor shortages—underlay this prosperity. Importantly, this period was characterized by high world demand for foodstuffs and natural resources, and New Zealand enjoyed preferential access to the lucrative British market.

63. The system came under increasing strain beginning in the late 1960s. First, concurrent with unsettled industrial relations elsewhere in the OECD, unions began to make aggressive wage demands, which were at first rejected by the government’s Arbitration Council. Unions then began to find direct bargaining with employers to be more effective in achieving their wage aims. A period of spiraling, double-digit wage increases followed. In 1971, the government unilaterally imposed wage controls through the Stabilization of Remuneration Act, and shortened the award period, which had averaged 2–3 years, to no more than twelve months. Under successive governments; numerous attempts were made at both wage controls and free bargaining, including through the 1973 Industrial Relations Act, which attempted to shift emphasis away from collective bargaining. Between March 1971 and November 1984, wage controls of one form or another were in force for nine years (Beaumont, 1993).

64. Unsettled industrial relations in the 1970s were exacerbated by a loss of privileged access to the British market and falling prices for New Zealand’s exports, resulting in increasing macroeconomic imbalances. The loss of access to the British market—especially for agricultural products—followed Britain’s joining the European Economic Community (EEC) in 1973. Moreover, New Zealand suffered from the EEC’s dumping of agricultural surpluses in third markets. The terms of trade deteriorated by 30 percent from the mid-1960s to the mid-1970s (Kasper, 1996), and the balance of payments was put under further stress by the two OPEC oil price hikes. Government debt rose seven-fold from the mid-1970s to the mid-1980s, and New Zealand’s sovereign credit rating was downgraded.

65. Labor market reforms in the “first wave” (1984–1991) had a relatively minor contemporaenous impact on labor market outcomes. The Industrial Relations Act put in place by the Labour Party in 1984 sought to reduce relative wage rigidities. Recourse to arbitration was made voluntary for both parties, and the Arbitration Court was instructed that it was not to be bound by past precedent when settling a wage dispute. However, wage relativities remained largely unchanged. The Labour Relations Act of 1987 sought to achieve greater flexibility in the labor market through outlawing multi-tiered bargaining (i.e., it aimed to facilitate enterprise and industry based agreements). Nevertheless, centralization remained the norm in the labor market as most unions choose to keep members on awards during this period.

66. In the end, the reforms to the industrial relations system in the 1980s—while not to be minimized—did not have the intended effect of reducing labor market rigidities. The number of workers under enterprise agreements actually fell during this time. The growing demands of employers that the labor market needed to be reformed along with other areas on the economy (especially product markets), combined with the conversion of the National Party to a free-market (from a corporatist, “think big”) agenda, set the stage for the passage of the Employment Contracts Act in 1991.

C. The Employment Contracts Act

67. Pressures for labor market deregulation culminated under the newly-elected National government with the passage of the Employment Contracts Act in May 1991, which is widely seen as the most radical labor market reform in the OECD. The underlying rationale of the EC A was to make employment contracts similar to those in all other areas of economic activity, while its aim was to increase efficiency in the labor market. In a broader sense, the aim was to bring labor market reforms “up to speed” with the reforms in product markets and macroeconomic management undertaken in the mid- and late-1980s.

68. With the fundamental shift in focus under the ECA from, multi-enterprise, collective awards to single enterprise and individual employment contracts, a century of legislative protections for unions disappeared. The main elements of the ECA were:

• All bargaining was towards an employment contract. Both employers and employees had the freedom to choose with whom and within what structures they negotiated.²² Collective contracts could cover one or more employers and any agreed group of employees as long as the contract was ratified by all parties involved.

• Giving preference to union members in contracts, or unduly pressuring an employee to join (or not join) a union was made illegal.

• The right to strike and lockout were maintained (from the 1980s reforms), but were permitted only after the expiry of a contract.

Also, union registration was terminated, and the withdrawal of the government from the award negotiation process took the taxpayer out of direct funding of labor market outcomes.

69. Nevertheless, the government continued to be involved in the labor market in two areas. First, statutory employment conditions relating to a minimum wage, annual leave, holidays and sick leave, domestic and parental leave entitlements, and prohibition of discrimination, to name a few, were maintained.²³ Second, the government remained involved in mediation and arbitration of disputes—although such arbitration was no longer mandatory—through a restructured specialized court system comprising an Employment Tribunal (for quick resolution of disputes) and an Employment Court (for appeals from the Tribunal relating to any action founded in an employment contract). Private arbitration was also permitted.

Industrial Relations Outcomes

70. The effect of the ECA on industrial relations is for the most part not controversial: union density dropped; the proportion of both individual and single employer based contracts rose; and measures of labor unrest (number of disputes and days lost) fell sharply. However, as is often the case with this type of analysis, caution is needed when attributing the changes listed below to solely the ECA. In particular, some of the outcomes observed in the 1990s may reflect the increased product market competition resulting from reforms begun in the 1980s, which reduced rents and, by extension, the gains from bargaining.

71. While union density exhibited a trend decline in New Zealand dating back to at the 1980s, there was a marked acceleration following the passage of the ECA (Table III.1). In the second half of the 1980s, there was some concentration of union membership as the number of unions fell by more than half, but total membership rose slightly. The period immediately following the passage of the ECA—i.e., the second half of 1991 and 1992—saw a noticeable spike in the annualized decline of union membership (by 20 percent over these 1½ years), while the annual rate of decline has fallen to an average of about 5 percent since. Interestingly, the number of unions has risen by almost half since end-1992. Also, at least some unions have shifted their focus from traditional negotiation and organization roles to innovative providers of services, such as training.

Table III.1.

New Zealand: Trade Unions, Membership and Union Density, 1985–98 (Selected Years)

Source: New Zealand Treasury.

Table III.1.

New Zealand: Trade Unions, Membership and Union Density, 1985–98 (Selected Years)

	Number of Unions	Total Membership	Annualized Decline	Union Density
			(in percent)
December 1985	259	683,006	--	43.5
September 1989	112	648,825	1.3	44.7
May 1991	80	603,118	4.2	41.5
December 1991	66	514,325	25.2	35.4
December 1992	58	428,160	16.8	28.8
December 1993	67	409,112	4.4	26.8
December 1994	82	375,906	8.1	23.4
December 1995	82	362,200	3.6	21.7
December 1996	83	338,967	6.4	19.9
December 1997	80	327,800	3.3	18.8
December 1998	83	306,687	6.4	17.7

Source: New Zealand Treasury.

View Table

Table III.1.

New Zealand: Trade Unions, Membership and Union Density, 1985–98 (Selected Years)

	Number of Unions	Total Membership	Annualized Decline	Union Density
			(in percent)
December 1985	259	683,006	--	43.5
September 1989	112	648,825	1.3	44.7
May 1991	80	603,118	4.2	41.5
December 1991	66	514,325	25.2	35.4
December 1992	58	428,160	16.8	28.8
December 1993	67	409,112	4.4	26.8
December 1994	82	375,906	8.1	23.4
December 1995	82	362,200	3.6	21.7
December 1996	83	338,967	6.4	19.9
December 1997	80	327,800	3.3	18.8
December 1998	83	306,687	6.4	17.7

Source: New Zealand Treasury.

72. There was a strong—and almost immediate—shift from multi-employer/awards based contract coverage to individual and single employers contacts (Table III.2). Data through 1996 show that the proportion of multi-employer/award contracts fell from about 60 percent in May 1991 to less than 10 percent by end-1992, and remained constant thereafter. Conversely, individual contracts rose from 28 percent in May 1991 to around one-half of all contracts thereafter, while single employer contracts rose from 13 percent in May 1991 to around 35 percent thereafter.

Table III.2.

New Zealand: Coverage of Individual and Collective Employment Contract

(IEC and CEC)

Source: New Zealand Treasury.

Table III.2.

New Zealand: Coverage of Individual and Collective Employment Contract

(IEC and CEC)

	IEC	Multi Employer (and awards)	Single Employer	Combined IEC/CEC	Total CEC
	(as a percentage of total contracts)
May 1991	28	59	13	--	72
August 1992	52	8	35	5	48
August 1993	40	9	37	8	54
August 1996	49	11	34	4	49

Source: New Zealand Treasury.

View Table

Table III.2.

New Zealand: Coverage of Individual and Collective Employment Contract

(IEC and CEC)

	IEC	Multi Employer (and awards)	Single Employer	Combined IEC/CEC	Total CEC
	(as a percentage of total contracts)
May 1991	28	59	13	--	72
August 1992	52	8	35	5	48
August 1993	40	9	37	8	54
August 1996	49	11	34	4	49

Source: New Zealand Treasury.

73. Key indicators of labor unrest fell sharply—and, again, almost immediately—in the period following the passage of the ECA (Table III.3). Unlike union density and contract coverage, there were differing views on what would happen to strike activity under the ECA, with some commentators predicting widespread unrest (see Kasper, op. cit.). In the event, the number of annual industrial disputes fell from an average of 234 over the “first wave” of reform period spanning 1984–90 to 66 (with very little variation) over 1991–96, a drop of around 70 percent. In terms of (thousands of) working days lost, the decline was more pronounced, falling from an average of 541 during 1984–90 to 67 over 1991–96, a drop of almost 90 percent. However, this series has been relatively volatile over in the post-ECA era.

Table III.3.

New Zealand: Indicators of Labor Unrest

Source: Bray and Walsh (1998), p. 387.

Table III.3.

New Zealand: Indicators of Labor Unrest

	Number of Industrial Disputes	Total Working Days Lost (’000)
1984	364	425
1985	384	759
1986	215	1329
1987	193	366
1988	172	382
1989	171	193
1990	137	331
1991	71	99
1992	54	114
1993	58	24
1994	69	38
1995	69	53
1996	74	73
1984-90 (average)	234	541
1991-96 (average)	66	67

Source: Bray and Walsh (1998), p. 387.

View Table

Table III.3.

New Zealand: Indicators of Labor Unrest

	Number of Industrial Disputes	Total Working Days Lost (’000)
1984	364	425
1985	384	759
1986	215	1329
1987	193	366
1988	172	382
1989	171	193
1990	137	331
1991	71	99
1992	54	114
1993	58	24
1994	69	38
1995	69	53
1996	74	73
1984-90 (average)	234	541
1991-96 (average)	66	67

Source: Bray and Walsh (1998), p. 387.

Macroeconomic Outcomes

74. Isolating the impact of the ECA on macroeconomic variables is difficult given the other structural reforms ongoing at the time of its implementation as well as the influences of the economic cycle. With these provisos, however, this section will attempt to distill the available information as it relates to the effects of the ECA on macroeconomic performance.

75. An assessment of the effect of the ECA on productivity is particularly thorny and there is no agreement in the literature as to the success or failure of the ECA in this regard. First, the impact of a successful labor market reform has countervailing effects on labor productivity. If the reform is successful in, say, better aligning pay with skills, and thereby results in increased effort by workers, then labor productivity would rise. On the other hand, a more efficient labor market could spur the demand for labor, raising employment and hence lower labor productivity.²⁴ The available data suggest that labor productivity post-ECA was unchanged from the 1980s (Table III.4). Some authors (e.g., Easton, 1996) cite this as evidence that the ECA was unsuccessful, while others (e.g., Savage, 1997) note that rapid employment growth may have dampened any productivity gains. Perhaps less contentious is the predicted effect on capital productivity, which would increase regardless which of the two effects noted above dominates. Hall (1996) showed that while labor productivity was broadly unchanged across the 1979–87 and 1992–95 expansions, annual capital productivity growth increased from an average of zero in the first of these expansions to 2.7 percent per annum in the second. This sewed to raise total factor productivity growth from 1.3 percent in the 1979–87 expansion to 2.3 percent in over 1992–95.

Table III.4.

New Zealand: Business Cycle Analysis of Productivity Growth

(In percentage growth rates)

Sources: Savage (1997), Table 6.4.

Table III.4.

New Zealand: Business Cycle Analysis of Productivity Growth

(In percentage growth rates)

	1979–87	1987–92	1992–95
	Expansion	Contraction	Expansion
Labor inputs	0.4	−1.8	2.7
Capital inputs	2.5	2.3	2.1
Labor productivity	2.0	1.9	2.0
Capital productivity	0.0	−2.2	2.7
Total factor productivity	1.3	0.3	2.3
Gross domestic product	2.5	0.0	4.8

Sources: Savage (1997), Table 6.4.

View Table

Table III.4.

New Zealand: Business Cycle Analysis of Productivity Growth

(In percentage growth rates)

	1979–87	1987–92	1992–95
	Expansion	Contraction	Expansion
Labor inputs	0.4	−1.8	2.7
Capital inputs	2.5	2.3	2.1
Labor productivity	2.0	1.9	2.0
Capital productivity	0.0	−2.2	2.7
Total factor productivity	1.3	0.3	2.3
Gross domestic product	2.5	0.0	4.8

Sources: Savage (1997), Table 6.4.

76. Recent OECD studies show that New Zealand’s structural unemployment rate and NAIRU dropped relatively sharply in the 1990s (OECD, 1997 and Richardson et al., 2000) (Table III.5). While the (weighted) OECD average structural unemployment rate increased slightly from 1990 to 1996, New Zealand’s dropped by 1.3 percent of GDP (following a steep rise in the late 1980s), among the largest drops in the OECD membership. Perhaps more relevant, New Zealand was near the top of the English-speaking group over the first part of the 1990s in terms of the fall in the structural unemployment rate—only the UK registered a larger decline. However, as with changes in labor productivity, it is not possible to establish a casual link between labor market reform and drops in NAIRU-like concepts.

Table III.5.

OECD Structural Unemployment Rates/NAIRU for Selected Countries ^1/

(In percent)

Sources: OECD (1997) and Ricardson et al. (2000).

^1/Data in the first three columns are structural balance estimates from OECD (1997), based on the non-accelerating wage rate of unemployment made by the OECD Secretariat. The last column contains estimates of the NAIRU made by OECD.

^2/First quarter.

Table III.5.

OECD Structural Unemployment Rates/NAIRU for Selected Countries ^1/

(In percent)

	1986	1990	1996	1999 2/
Australia	8.1	8.2	8.5	7.6
Canada	8.3	9.0	8.5	8.1
New Zealand	4.7	7.3	6.0	5.8
United States	6.2	5.8	5.6	5.2
United Kingdom	10.2	8.4	7.0	7.2
OECD average (weighted)	7.0	6.8	7.1	…

Sources: OECD (1997) and Ricardson et al. (2000).

^1/Data in the first three columns are structural balance estimates from OECD (1997), based on the non-accelerating wage rate of unemployment made by the OECD Secretariat. The last column contains estimates of the NAIRU made by OECD.

^2/First quarter.

View Table

Table III.5.

OECD Structural Unemployment Rates/NAIRU for Selected Countries ^1/

(In percent)

	1986	1990	1996	1999 2/
Australia	8.1	8.2	8.5	7.6
Canada	8.3	9.0	8.5	8.1
New Zealand	4.7	7.3	6.0	5.8
United States	6.2	5.8	5.6	5.2
United Kingdom	10.2	8.4	7.0	7.2
OECD average (weighted)	7.0	6.8	7.1	…

Sources: OECD (1997) and Ricardson et al. (2000).

^1/Data in the first three columns are structural balance estimates from OECD (1997), based on the non-accelerating wage rate of unemployment made by the OECD Secretariat. The last column contains estimates of the NAIRU made by OECD.

^2/First quarter.

77. Real wage growth moderated substantially in the first half of the 1990s compared with the last half of the 1980s, falling from double digit rates²⁵ to around 1½ percent. (Table III.6) (Savage, op. cit.). This is in contrast to the performance of the Australian labor market over this period, which was characterized by relatively slow employment growth and increase growth of labor productivity. Under those conditions, the move away from centralized bargaining resulted in higher real wage growth in the 1990s as workers were able to cash in on higher labor productivity growth (Dawkins, 2000).

Table III.6.

New Zealand: Wage and Labor Market Trends before and after the ECA

(In percentage growth rates)

Sources: Savage (1997), Table 6.3.

Table III.6.

New Zealand: Wage and Labor Market Trends before and after the ECA

(In percentage growth rates)

	1981 (1)-	1986 (1)-	1991 (1)-
	1986 (1)-	1991 (1)-	1996 (1)-
Nominal wages	61.7	53.6	11.1
Real product wages	−1.1	15.4	1.7
Real consumption wages	−9.4	8.5	1.3
Employment	8.0	−4.7	13.0
Unemployment rate	57.1	13.5	−39.4

Sources: Savage (1997), Table 6.3.

View Table

Table III.6.

New Zealand: Wage and Labor Market Trends before and after the ECA

(In percentage growth rates)

	1981 (1)-	1986 (1)-	1991 (1)-
	1986 (1)-	1991 (1)-	1996 (1)-
Nominal wages	61.7	53.6	11.1
Real product wages	−1.1	15.4	1.7
Real consumption wages	−9.4	8.5	1.3
Employment	8.0	−4.7	13.0
Unemployment rate	57.1	13.5	−39.4

Sources: Savage (1997), Table 6.3.

78. The ECA had a marked effect on the seasonal pattern of wages and earnings (OECD, 1996). This is a clear result of wage negotiations becoming more staggered throughout the year rather than being concentrated in a single wage round. Another empirical regularity that appears to have broken down is the relation between changes in generalized real wages and local skill shortages. This would suggest that the incidence of collective wage formation had lessened. More formally, econometric work by the OECD (1996, op. cit.) on the labor market suggested that employment in the post-ECA period responded more quickly to changes in output than in the pre-ECA period. Finally, there is evidence that “penal” overtime wage rates of up to triple ordinary time rates had disappeared following the introduction of the ECA, and were replaced in most instances by time-and-a-half rates (Beaumont, op. cit).

79. Household income inequality in New Zealand increased unambiguously over the period 1993–86 to 1995–98 (O’Dea, 2000). Moreover, the increase in inequality in New Zealand appears to have exceeded that of other English-speaking countries, leaving it with one of the highest levels of inequality in the OECD. Several caveats are in order, however: (i) cross country comparisons of this type should be treated with caution due to measurement inconsistencies; (ii) these studies show that most of New Zealand’s increase in income inequality occurred in the 1980s, and can therefore not readily be linked to the ECA and (iii) most of the increase in income inequality in the comparator countries occurred in the 1970s, which are not included in this sample. Ongoing work has identified household composition (e.g., growth of sole parent and older households) and age-mix/qualifications as contributing to one-half of the deterioration in income equality. The remaining half has yet to be identified statistically, but variables such as wage dispersion are candidates.²⁶

80. Labor force participation in the 15–64 age cohort rose strongly in New Zealand over 1990–97 (Table III.7). ILO (1999) data show that the participation rate for this group increased by 2½ percentage points over this period, with the bulk of the increase reflecting female participation. Most of the industrial countries had falling or flat participation profiles over this period, including the English-speaking group, where only the US had a (1 percentage point) rise its participation rate. New Zealand was the only country in this group to have a rise in male labor force participation.

Table III.7.

New Zealand: Labor Force Participation in Selected Countries ^1/

(In percent)

Sources: ILO (1999).

^1/For the 15–64 age group.

Table III.7.

New Zealand: Labor Force Participation in Selected Countries ^1/

(In percent)

	1980			1990			1997
	Total	Male	Female	Total	Male	Female	Total	Male	Female
Australia	69.2	86.2	51.9	73.0	84.4	61.5	72.5	81.9	63.0
Canada	71.8	85.9	57.8	76.8	84.9	68.6	74.9	81.8	68.0
New Zealand	66.1	85.8	46.1	73.0	83.0	63.2	75.5	84.1	67.1
United States	72.5	85.8	59.9	76.5	85.6	67.8	77.4	84.2	70.7
United Kingdom	73.1	89.2	57.0	77.8	88.3	67.2	76.2	84.4	68.0

Sources: ILO (1999).

^1/For the 15–64 age group.

View Table

Table III.7.

New Zealand: Labor Force Participation in Selected Countries ^1/

(In percent)

	1980			1990			1997
	Total	Male	Female	Total	Male	Female	Total	Male	Female
Australia	69.2	86.2	51.9	73.0	84.4	61.5	72.5	81.9	63.0
Canada	71.8	85.9	57.8	76.8	84.9	68.6	74.9	81.8	68.0
New Zealand	66.1	85.8	46.1	73.0	83.0	63.2	75.5	84.1	67.1
United States	72.5	85.8	59.9	76.5	85.6	67.8	77.4	84.2	70.7
United Kingdom	73.1	89.2	57.0	77.8	88.3	67.2	76.2	84.4	68.0

Sources: ILO (1999).

^1/For the 15–64 age group.

81. The rate of unemployment, which now stands at 6.1 percent, fell from a peak of 11 percent early 1992 to 5.7 percent in the second half of 1995 before rising to 7.6 percent in the recession of 1998. As with other macroeconomic variables, one needs to separate the effects of other influences (in this case the relatively robust expansion for most of the 1990s) to isolate the effect of labor market reform. Econometric work by Maloney (1994), based on industry data in the two years following the implementation of the ECA, shows that while there is no direct link between the ECA and employment growth, there is an indirect link in that lower union density is associated with increased employment. He estimated that 22 percent of the employment growth between the second quarter of 1991 and the third quarter of 1993 was indirectly a result of the ECA.

82. Overall, there is no consensus view on the outcome of labor market reforms in New Zealand. Indeed, given the depth of the reform and the wide—range of other reforms implemented, the answer to why New Zealand has not closed the gap in terms of output and productivity growth with the comparator group consisting of the United Kingdom (which did) undertake a significant labor market reform) and the US and Australia (which did not) remains a puzzle (Gregory, 1999). There are several proffered (non-mutually exclusive) explanations:

• Labor market outcomes were dwarfed by other outcomes. This line of thinking would argue that given the high income level of New Zealand and the existence of well-developed factor markets, the impact of the type of reforms undertaken in the labor market would be relatively small, and that GDP growth is more sensitive to more basic variables such as technological change and/or savings. Adherents of this view would note that the US did not undertake any meaningful labor market reform and that Australia did not undertaken any fundamental labor market reform, but that both countries outperformed New Zealand in the 1990s.

• The reforms need more time to take effect. While the reforms in the labor market were undertaken rapidly, there is hysterisis in labor market behavior, even though bargaining structures seem to have changed rather quickly. Relatedly, in the period following the implementation of the ECA, monetary policy credibility was paramount, meaning that the relatively weak macroeconomic performance in New Zealand was the cost of ridding the system of inflation.

• Social benefit systems continue to act as a deterrent. This hypothesis would acknowledge the disincentive problems in the labor market (and the associated macroeconomic costs) of high effective marginal tax rates of low income families (resulting from generous public benefits) and “passive” labor market measures.²⁷ A recent OECD study (Martin, 1998) shows that, despite narrowing the gap, New Zealand continues to have the highest summary measure of entitlement benefits among the English-speaking group. Also, the same study shows that the share of active labor market policy expenditure to total labor market policy expenditure has fallen by 20 percentage points over 1985–96 to close to the OECD average. Perhaps a more general reading of this view would be that while the product and factor markets underwent fundamental reform in the 1980s and early 1990s, the social sectors did not. By extension, only when the social sectors are reformed will the optimal “interface” with the labor market be achieved.

• The ECA did not achieve its aims, or was somehow flawed. This view states that since the relative economic performance of New Zealand—in particular labor productivity—has not improved, the ECA did not deliver on its promise of a more efficient labor market. Alternatively, some observers are implicitly arguing that the objective function used to formulate the ECA was flawed in that it put too much weight on efficiency (i.e., purely contractual) concerns.

While it is beyond the scope of this paper to provide a resolution as to which (combination) of these hypotheses is most accurate, the discussion above does suggest that the first three have merit. However, such a resolution may ultimately be impossible owing to the lack of a counterfactual.

83. While it is difficult to pinpoint the effects of the ECA on industrial relations and, more broadly, macroeconomic outcomes—although these would appear to be positive on balance—the ERA has been framed in terms of a broader social objective function. That is, the government has signaled a shift in emphasis from purely efficiency terms in evaluating performance to include considerations of equity and fairness. Although the presence of differing objective functions would make a straightforward comparison of the two Acts impossible, the next section tries to look at how the central elements of the ERA could effect the efficiency gains achieved under the ECA.

D. The Employment Relations Act

84. Following the election of November 1999, the coalition Labour-Alliance government came to power (replacing the National Party, which had been in power for nine years), and moved quickly on its campaign promise to replace the ECA with more union-friendly legislation. In March 2000, the Employment Relations Bill was introduced in to Parliament with the intent to “provide for balance in the conduct of employment relationships [by] aiming to improve mutual trust and confidence between employers and employees” (New Zealand Labour Party, 2000)

85. There was considerable turmoil surrounding the Employment Relations Bill, with the opposition charging that: (i) the Bill was a backward-looking departure from the status quo; (ii) the Government had not made the case that the ECA was defective in achieving good labor market outcomes; and (iii) only one measurable outcome (productivity, see below) was provided for the Bill, leaving the public with no way to measure the legislation’s success or failure (NZ National Party, 2000). In response to these concerns and those of business community, some changes were made—the definition of employee was clarified to exclude independent contractors and others from the coverage of the ERA, as were the conditions of union access to the workplace and the application of “good faith” bargaining—but the thrust of the legislation remained unchanged. The ERB was passed into law in August 2000.

86. The remainder of this section will be devoted to identifying areas where the ERA could be assessed regarding its impact on industrial relations and macroeconomic outcomes, acknowledging that many of these will not be observed for some time. To begin, we note that-the main features of the ERA are as follows:

• Acknowledging the inherent inequality of bargaining power in employment relationships, the Act specifies that such relationships must be built on “good faith.”

• All collective bargaining must be undertaken through unions (which will be a signatory to any resulting employment contact) and only union members may be covered by a collective bargaining agreement.

• Collective bargaining is promoted by the Act through the requirements to bargain in good faith over a collective agreement, and the additional guidance over the process by which orderly collectively bargaining should be conducted. The Act also provides that industrial action can be taken in pursuit of single or multi-party collective agreements, and provides for mediation services which will be able to be utilized by the parties to collective bargaining (see below).

• Individual choice is protected—union membership is voluntary and “freedom of association” is maintained.

• New institutions—a Mediation Service (staffed by the Department of Labor) and the Employment Relations authority (an independent statutory body)—will be set up to promote informal and low-level resolution of problems.

• The legislation establishes an entitlement for those covered by a collective agreement to take leave for training in employment relations, at the employer’s expense. The union allocates the use of this right to eligible employees who must be union members covered by the relevant collective agreement.

87. The effects of most of these items on labor market performance are not directly measurable although inferences as to their effects will be possible over time. In fact, the only macroeconomic claim made by proponents of the ERA is that productivity will be increased owing to improved mutual trust and confidence in the workplace. However, as discussed earlier, particularly in the case of labor productivity, the effect of labor market reforms can go either way.²⁸

88. It is worth noting that even after the passage of the ERA, employment arrangements in New Zealand will remain closer to those in North America than to those in many European countries. In fact, one could argue that at least in some respects the labor market in New Zealand will remain more “liberal” than those in the US and Canada.²⁹ First, although there was much debate over the “good faith” bargaining requirement in the ERA, such a requirement has always existed hi North America. Second, while the ERA effectively specifies that only union members can be covered by collective agreements, this is also permitted in North America. Finally, the ERA (like the EC A) lacks statutory job protections such as a minimum notice period or severance pay, which have a significant body of case law in North American jurisdictions.

89. Another indirect piece of evidence suggesting that the ERA may be less of a throwback to the pre-ECA days than suggested by some commentators, is the electoral reforms that went into effect in New Zealand in 1993. Following what were perceived as sub-optimal, violent swings in policies under successive governments occupying differing poles of the political spectrum, the electoral system was changed from one of pure representation to a “mixed” system based on the German model comprising elements of both pure and proportional representation. The ensuing period has been characterized for the most part by a string of coalition governments, which has ostensibly lessened the amplitude of policy changes. In addition, some observers note that the political system in New Zealand since 1993 is likely to prevent large swings in policies, and to the extent that legislation must reflect the views of the coalition, it is likely to stay short of a radical change.

90. That said, the ERA does appear to have the potential to impart some rigidities—as well as uncertainties—into the New Zealand labor market. Below are the main areas to watch as the effects of the ERA unfold, divided into those that are likely to be readily observable in the short-term and those that may only be realized after a long time. It is worth emphasizing that the effects of uncertainty regarding both the changes in substantive legislation and in the institutions enforcing these changes, underlay many of these considerations.

E. Conclusion

91. The months leading up to the passage of the Employment Relations Act have been a contentious period in New Zealand characterized, inter alia, by acrimonious political exchanges regarding the effects of the legislation and a sharp drop in business and investor confidence. At issue is the extent to which the new legislation represents a necessary correction in policies that will have modest, positive results or whether it represents a reversal in labor market policies that would lead New Zealand back to the pre-ECA era of relatively high labor unrest and relatively inflexible labor markets.

92. In the end, this is an empirical question, although coming to a definitive answer is fraught with difficulties. First, the objective function underlying the new labor market policy has not been clearly spelled out. While the Employment Contracts Act was designed solely to improve labor market efficiency, the Employment Relations Act includes equity concerns, which are arguably more difficult to measure. (Moreover, the relative weights between efficiency and equity have not been spelled out.) Second, although many of the data needed to come to a judgement on labor market outcomes are available, it will be a challenge—as it has been in assessing the impact of the ECA—to separate the effects of labor market reforms from the cyclical and structural trends in the economy. Third, some of the information needed to assess the ERA is either unobservable (like the effects of the build up of case law) or will only be available with very long lags (e.g., NAIRU). Overall, barring a return to 198Os-style labor market outcomes, it seems realistic to say that a consensus on the effects of the ERA on New Zealand’s labor market and macroeconomic performance will not be reached in the near future.