Taking the pulse at work: An employment relations scorecard for Australia

Abstract

This study, and the project behind it, is an attempt 100 years on from the Webbs to comprehensively assess the health of the industrial relations/employment relations system by ‘taking the pulse’ of the employment relationship. If, as we argue, the relative health and performance of the employment relationship remains the key dependent variable of the field of employment relations today, there have been remarkably few attempts to audit and measure its critical dimensions. This study, founded on a large representative survey of workers and managers across Australia, the United States, the United Kingdom, and Canada, attempts to do just that, and produces in this article, results of those survey questions for Australia. The article is novel since this kind of employment diagnostic is based on a unique nationally representative survey of employers and employees. The study is also innovative, in that it presents the results of the health of the system in the form of an employment relations scorecard and is the first such attempt to do so in industrial relations.

Keywords

Employee voice employment relations human resource management industrial relations management working conditions

Introduction

The need to understand and resolve the many ‘labour problems’ that emerged across industrialised nations in the late nineteenth/early twentieth century was central to the birth of the field of industrial relations (Ackers and Wilkinson, 2003; Derber, 1967; Feldman, 1928; Kaufman, 1993, 2004a, 2004b). Although the origins of industrial relations/employment relations (IR/ER)¹ analysis and theory date back more than 100 years in Britain to Sydney and Beatrice Webb (1897) and even further back to Adam Smith, a series of related events give us cause to see the period leading up to the 1920s as the starting point for industrial relations as a field of active research and investigation (Commons, 1935; Kaufman, 2010). While academic industrial relations departments were being created in the US, chiefly in Wisconsin in 1920 by the likes of John Commons, (1934) it was Woodrow Wilson’s nine-member Commission on Industrial Relations (CoIR) that first attempted to comprehensively assess the health of the employment relationship. The Commission’s rationale was reproduced in its Final Report which stated:

That the commission shall inquire into the general conditions of labor in the principal industries of the United States, [and] … into existing relations between employers and employees … The commission shall seek to discover the underlying causes of dissatisfaction in the industrial situation and report its conclusions thereon. (CoIR, 1916: 6)

The report was commissioned in 1916 to prevent labour–management conflict and to stem deteriorating industrial relations, which included a violent escalation of the dispute at Colorado Iron and Fuel, dubbed the Ludlow Massacre. The events surrounding this tragedy induced owner/industrialist, and notable welfare capitalist, JD Rockefeller to employ his own IR counsellor, McKenzie King (later Canadian Prime Minister). It was McKenzie King who persuaded Rockefeller to create the first corporate industrial relations department at Standard Oil in New Jersey in 1918. This experiment was a testing ground for the welfare capitalism movement and spawned a near-century long search for company-led systems that could simultaneously improve employee morale and boost workplace performance.

This study, and the project behind it, is an attempt 100 years later to comprehensively assess the health of the IR/ER system by ‘taking the pulse’ of the employment relationship. If, as we argue, the relative health and performance of the employment relationship remains the key dependent variable of the field of ER today, there have been remarkably few attempts to audit and measure its critical dimensions. This study, founded on a large representative survey of workers and managers across Australia, the United States, the United Kingdom and Canada, attempts to do just that, and produces in this article, results of those survey questions for Australia. The article is novel since this kind of employment diagnostic is based on a unique nationally representative survey of employers and employees. The study is also innovative, in that it presents the results of the health of the system in the form of an ER scorecard and is the first such attempt to do so in IR.

The significance and relevance of attempting to gauge the relative heath of ER is indicated in an article in the Australian Financial Review (24 October 2011: 55) that remarks that a ‘limiting factor [in Australian economic performance] is not necessarily unions, but the difficulties experienced by management in structuring their relationship with workforces in a positive and collaborative way.’ Of added worry, since 2000, economic performance as measured by productivity growth has faltered and one suggested explanation is a decline in ‘unmeasured labour quality’ (Dolman, 2009: 257). A number of factors contributing to unmeasured labour quality are included in this project, such as the quality of management, the extent of positive feelings and cooperation between managers and workers, the work ethic and commitment of employees, and the relative flexibility of trade union practices and labour law.

A growing body of empirical research has shown (Black and Lynch, 2001; Gittell et al., 2004) that the effect of poor ER also creates many other problems that handicap business and imposes significant costs on workers and society (e.g., labour turnover, absenteeism, workplace injuries, and industrial disputation). However, getting managers and workers energised and actively partnering together is sufficiently complex and ill-understood that HRM scholars call the process the ‘black box’ of ER research (Becker and Huselid, 2006). By collecting extensive data from workers and managers across all sectors on a wide range employment relationship dimensions, we hope to shed new light on a number of these issues.

Our article is, therefore, important and relevant because the topic: (1) is prominent in current-day business, economic and policy-making discussions in Australia; (2) has, by some accounts, recently worsened in Australia relative to major trading partners and consequently threatens the nation’s continued economic growth and prosperity; (3) still remains a poorly understood ‘black box’ in the research literature; and (4) is timely given that current ER are accompanied by unprecedented change, pressure and stress in the workplace (Findlay and Thompson, 2017).

Literature review

That there has been significant change to Australian ER is both undeniable and well documented. Overall, there has been a shift toward a market-driven model, underpinned by the principles of neo-liberalism (Bray and Underhill, 2009; Isaac and Lansbury, 2005; Wilkinson et al., 2009). This is most visible in once heavily-unionised and protected industries, with extensive collective bargaining coverage, which have declined dramatically in recent decades (Australian Bureau of Statistics (ABS), 2014; Pekarek and Gahan, 2016). In an effort to promote increased competitiveness, regulation through arbitrated awards has been progressively eroded in favour of extensive decentralisation of bargaining (Mitchell et al., 2010). The IR change process and the proliferation of non-standard employment arrangements has produced greater disparity across the labour market, and there remains (or indeed we have seen intensified) evidence of inequality, of which gender inequity has featured prominently in labour market analysis. Although expected to deliver gains to workers as well as firms, evidence suggests that many of the ‘returns’ from market-driven IR reforms have been captured by employers. For employees, particularly outside the neo-liberal heartlands and especially those subject to casualisation, there is growing wage disparity and employment insecurity (Watson, 2016).

While there is an extensive literature examining the ‘structural/institutional’ changes to the employment relationship in Australia, our focus is on the state or health of ‘employment relations’ itself. Our study seeks to examine the views of workers and managers, respectively, about current workplace issues and problems. In doing so, our study is one of a small number of major, nationally representative surveys of Australian ER, and builds on other studies such as those conducted by the Fair Work Commission (FWC), and the Australia at Work (AaW) project. Seeking to fill the data gap on Australian workplace characteristics left after the final AWIRS survey, FWC commissioned its own Australian Workplace Relations Survey (AWRS). The AWRS results produced in the FWC (2015) First Findings Report were derived from data on a wide range of employment characteristics and a survey of the views of employees. In addition, FWC reported, albeit somewhat parenthetically, on management views about issues such as ‘why employers favour certain types of agreements and rosters rather than others’, and ‘whether labour productivity has improved’. Part 6 of the Report contained employee experience indicators, including measures of job satisfaction. Here the report finds, broadly, that managing work and non-work responsibilities is a key driver of employee job satisfaction, especially for females who, consistent with the findings of AaW, report higher overall levels of satisfaction than males. The report also found that quality of work is central to the satisfaction of professional, scientific and technical employees, compared to employees in mining, construction, and rental, hiring and real estate, where explicit rewards figure more prominently.

The AaW project collected data over five years on labour market and employment characteristics and outcomes (for example, wages, work hours, job tenure and union status). A short set of questions probed attitudes about relations between employers and employees. As reported in the project’s Benchmark Report (Van Wanrooy et al., 2007; also see Fact Sheet No. 15 on the AaW website), conflicting trends are evident. Overall, the report notes ‘workers are generally happy with their work and work environment’ (p. 85). Underneath this positive picture, however, lies a significant minority of employees who give negative grades to both objective features of the workplace (for example, work schedules) and subjective features (for example, managers can be trusted). The report also notes that the employees’ rating of their relationship with the company deteriorates with greater perceived job insecurity, a worrisome finding given the new employment relationship trends noted above.

Data and methods

The State of the Workplace Employment Relations Survey (SWERS)

The preceding summary provides context for a discussion of the current state of employer–employee relations (EER) in Australia. In assessing the state of EER, we decided to focus on the workplace level, and developed a custom-designed survey instrument and data set, the SWERS, which collected data in 2016. The SWERS research agenda complements but is different from the AWRS and AaW in several respects. Whereas AaW surveys a broad and diverse range of labour market and workplace topics (for example, in the Benchmark Report see Chapter 1, ‘Awards, Agreements and Contracts at Work’; Ch. 4, ‘Earnings at Work;’ and Ch. 7, ‘Unions at Work’), our research focuses on a specific subset of topics only modestly covered by AaW (for example, portions of Ch. 6, ‘Employees’ Attitudes at Work’). That is, we probe more deeply into the state of EER in Australian workplaces and then go beyond AaW by examining (1) determinants of these relations, (2) problems that contribute to and grow out of these relations, and (3) consequences of these problems and relations for firm performance and worker wellbeing.

Similarly, while FWC examines the incidence of arrangements such as flexible work practices and pay structures, and demonstrates how these vary according to employee demographics such as gender or firmographics such as firm size, it does not determine how these practices reflect the quality of employment relationships. Therefore, left largely unanswered is how managers and employees feel about the degree of workplace flexibility or the current method(s) of payment. Other than by presumption, we cannot ascertain the extent to which low vs. high flexibility reflect employer prerogative or employee preferences, or whether differences in employee reporting of how pay is determined (either negotiated with employer vs. set by an award) reflects choice as to the method of payment, or is simply a function of prevailing practice.

While we seek to probe more deeply into the views of managers and employees about their relationship, another feature of this study that is novel is the attempt to score the responses to give an overall assessment of the state or ‘health’ of the ER system in Australia. There is no accepted method to do so in ER, so we have borrowed from the balanced scorecard (BSC) approach developed by (Kaplan and Norton, 2001, 2007, 2010).² BSC was originally designed to track the performance of a range of measures linked to business strategy by collecting and evaluating financial, customer, and business processes data. Within the BSC framework there was no specific category for employee. Instead they figured principally within the learning and growth perspective – in terms of the strategic skills and knowledge of the workforce to support strategy, and in the cultural shifts required to motivate, empower, and align the workforce behind the strategy (Boxall and Purcell, 2011). Later, Huselid and Becker modified the BSC to produce an HRM scorecard (Becker and Huselid, 2006) but importantly in this schema, the worker perspective is captured by examining worker ‘success’ rather than worker ‘satisfaction’. On this the HR scorecard approach is quite specific: it argues that a strategic choice (or value proposition) should be articulated, such that the workforce can understand and embrace how the employer intends to be successful in its chosen market (Beatty et al., 2003: 109). In other words, the BSC does not specifically suggest that employees are stakeholders in their own right, but only in so far as they can enhance customer satisfaction and financial performance through their ability to support business strategy – not through any moral perspective.

While acknowledging the usefulness of integrating key HR performance drivers into the strategic management framework, Boxall and Purcell (2011) are concerned that the BSC approach does not go far enough in relation to ER. The major concern is that ER is not just about satisfying corporate objectives but also relates to social legitimacy in terms of compliance with labour laws and the provision of policies that build long-run succession and development opportunities for managers and workers. As such, it [traditional scorecard measures] continues to reflect management and organisational interests primarily rather than those of all stakeholders (Marchington et al., 2016). Accordingly, we have developed an ER scorecard that is designed to test both worker and manager interests within the employment relationship.

Data collection

Data generated from this study comes from a broader project funded by the Australian Research Council (ARC), Social Science and Humanities Research Council (SSHRC in Canada) and Innovation Resource Centre for Human Resources in the USA examining ER in Australia, the USA, the UK and Canada including nationally representative surveys of employees and managers.³ In assessing the state of ER, we decided to focus on the workplace level, and developed a custom-designed survey instrument and data set, the SWERS. Unlike the employee voice surveys of Freeman et al. (2007), or the FWC survey, we include both employees (EME) and employers (EMR) in order to get a wider perspective on the state of the ER, as well as broadening the focus to the entirety of the employment relationship rather than simply representation and participation. ‘Employer’ in this case is a manager who, in legal terms, is the employer’s agent.

The employer and employee surveys were given to separate panels of respondents pre-assembled by a professional organisation survey company, Opinion Research Corporation (ORC), who conducted the survey on our behalf. The survey research firm, considered world leader in this technique, has a large panel of employees and managers in Australia drawing from a sample of HR/personnel managers and based on characteristics of the positive responders (industry, firm size, etc.). It then blends in requests to other panel members having the needed characteristics until an (approximately) nationally representative sample is obtained over key observables, which in our case were standard demographics for employees (age, gender, marital status, immigrant status, education) and for employers encompassed industry and workplace size. This is in effect a form of ‘quota sampling’ with online (as opposed to telephone) respondents. Quota sampling (Dodge, 2006: 428) does not have a sampling frame and as such does not generate a response rate since there is no randomly drawn target sample from which a proportion of valid responses can be drawn. Given that quota sampling is designed to ensure a representative sample across various key characteristics, it is not surprising – as seen in Table 1 – that the survey characteristics match very closely the employee population characteristics as given by the ABS figures for 2016.

Table 1.

Comparison of SWERS and Australian Bureau of Statistics (ABS) Means, 2016.

	Percentage (%)
	SWERS Employee sample	ABS Employment data
Age
18–24	11	14
25–34	21	23
35–44	24	22
45–54	25	21
55–64	17	15
65>	3	4
Gender
Male	54	54
Female	46	46
Marital status
Married	47	48
Not married (never married, separated/ divorced, widowed)	53	52
Country of birth
Australia	75	71
Overseas	25	29
Education^a
Advanced degree (Masters, Doctorate)	9	11
Bachelor degree	31	32
Associate degree or less (Certificate, High School, Some High School)	58	57
Tenure
Less than 1 year	12	15
1–2 years	14	19
3–9 years	45	39
10 years or more	29	27
Sector^b
Public	14	16
Private	86	84

Source: Authors calculations from State of the Workplace Employment Relations Survey (SWERS) (2016); Australia Bureau of Statistics (2016).

In SWERS 2% of sample did not list educational level so column does not sum to 100.

Public sector refers to federal, state and local government employees only and not public institutions such as schools or public utility while private is all other categories.

Each selected respondent then answers our survey online with questions that are as identical as possible to those in the employee survey with one distinction: the employee questions are typically framed in terms of the person’s workplace while the manager questions are typically framed in terms of the company/organisation, although in both cases respondents are told if they cannot knowledgeably answer at this level, to drop down to the level where they have sufficient knowledge about employment practices/relations. The company/workplace distinction introduces some non-comparability between report cards but employees cannot be expected to give reliable answers beyond their immediate workplace while upper-level managers and executives should have sufficient knowledge at a higher level (e.g., entire company for a small-medium sized enterprise; a separate business unit such as a store, plant or office; or for a business division in a large, multi-facility company).

The surveys were completed in early 2016 with a sample of roughly 1996 employees and 400 employers. Only workplaces with more than 20 employees were included. This article uses only the Australian data from SWERS, with its scores for 39 ER report card performance indicators. The mean and median values of these indicators are then used to derive a summary numerical score and letter grade indicating the state of the Australian ER on a low-to-high performance scale.

Our aim is to obtain far more specific and detailed information than contained in AWRS and AaW. Thus, rather than ask employees with one broadly framed question to rate EER on a five-point scale, we construct an ER climate index as a weighted average of responses to nine questions probing alternative dimensions of EER. This index ranks the employing companies from low to high. We follow this approach with other questions as well.

One issue in getting employees to do rankings relates to their knowledge of business operations. While employees are unlikely to know hard data on profits, sales, or productivity, our questions probe on qualitative measures of organisational performance such as operational efficiency, competitive position and whether management has optimised performance. Confidence in these subjective-type measures is provided by Forth and McNabb (2008) who find, using WERS data, that they have suitably high correlation with objective data and, indeed, are in some respects preferred because they capture intangible aspects of performance.⁴

Assessing the state of the ER: The ER scorecard approach

There is no universally agreed ER assessment framework so our scorecard attempts to develop one. We note that the first findings report (Fair Work Commission, 2015) provides descriptive data on employee/management practices but not EER as such, or an assessment of overall ER performance. Table 2 therefore presents the first ER scorecard for Australia. The answers to nearly all the questions in this part of the survey were solicited on a 1–7 scale, typically with 1 = lowest/worst and 7 = highest/best (e.g., 1 = lowest quality management, 7 = highest quality). Thus, we converted the 1–7 numeric answers to an F-to-A letter grade, with grades above F also distinguished into minus and plus categories (e.g., C−, C, C+). The conversion scale from numeric to letter grade is made to be as symmetric as possible using the following: 1–2.59 = F; 2.6–3.59 = D, 3.6–4.59 = C, 4.6–5.59 = B, and 5.6–7 = A. In-between letter grades are further defined using the same symmetric approach (e.g., 3.6–3.93 = C−, 3.94–4.26 = C, 4.27–4.59 = C+). Negatively-framed questions were inverted so they follow the positive 1–7 scale.

Our report card purports to measure the state of the ER with a set of 39 diagnostics (33 of which are common across the EME and EMR surveys with six unique to one or the other) covering seven performance areas. The seven performance areas include: (i) Organisational Outcomes; (ii) Employee Outcomes; (iii) Management Inputs; (iv) Employee Inputs; (v) EER Climate; (vi) Employee Relations Management Practices; and (vii) Community Outcomes. These seven areas are common to both the employer and employee surveys thus giving us access to both employer and employee perspectives.

As an example of how the 39 diagnostics intersect with our seven performance areas, we will detail how the last three performance areas are measured beginning with EER climate. Nine of the 39 dimensions are used to capture the EER climate. These include measures for whether: (i) employees are treated fairly and humanely (EME survey only); (ii) management’s receptivity to hearing employees gripes and problems; (iii) relations between managers and employees; (iv) organisation has a family/partnership feeling; (v) organisation has little conflict and infighting; (vi) level of employee morale; (vii) the employees’ connection and interest with what management says and does; (viii) employees collaborate/cooperate with managers (EMR only); and (ix) the organisation is a good place to work for women, minority, and LGBT employees (EME only, average of two separate questions).

Next is Employee Relations Management practices which are captured using seven of the 39 items and include measures of: (i) employee voice; (ii) employee listening and opinion methods; (iii) effective dispute resolution; (iv) effective/extensive internal communication with employees; (v) management style is collaborative/commitment; (vi) management uses positive motivation that emphasises hope of gain over fear of negative consequences; and (vii) there is a perception by employees that HR is a value-added function at the workplace (EMR only).

Finally, Community Outcomes attempt to capture social stakeholder performance on groups such as customers, families, and the local community. Three diagnostic measures are used including: (i) the organisation has happy/satisfied customers; (ii) organisation provides employees with flexible working arrangements; and (iii) organisation ranks high in community involvement and social responsibility.

Report card mean and median values are reported across all 39 individual diagnostic areas given that means and medians can diverge for distributions having considerable skewness. The individual means and medians are then averaged for the seven areas to provide sub-section scores as well as an overall average score (our eighth section). Based on the numerical scores, and using the conversion scale described earlier, corresponding report card letter grades are also listed for each question, section and the overall report.

The overall numerical score and letter grade at the bottom of Table 2 provide our project’s summary diagnostic measure of the health and performance of the Australian employment relationship, as rated respectively by employers and employees. The overall scores and grades are the average of the mean and median responses across all 39 individual components and, thus, in effect assign equal weights to each item.

Table 2.

State of the employment relation report card in Australia, employee and employer surveys.

	Employer (ER) Survey				Employee (EE) Survey
	Average		Median		Average		Median
Section	Score	Grade	Score	Grade	Score	Grade	Score	Grade
I. Companies/Workplaces: Performance outcomes
1. Management has optimized financial performance	5.08	B	5.00	B	–	–	–	–
2. Company/Workplace financial performance	5.28	B	5.00	B	4.91	B	5.00	B
3. Company/Workplace competitive position	4.29	C+	4.00	C	4.62	B−	5.00	B
4. Change in employment	4.62	B	4.00	C	3.92	C	4.00	C
5. Company/Workplace operational efficiency	5.08	B	5.00	B	4.17	C	4.00	C
6. Employer/Employee value proposition	5.33	B+	5.00	B	4.66	B−	5.00	B
Section I subtotal	4.95	B	4.67	B−	4.46	C+	4.59	C+
II. Employees: Satisfaction/Performance outcomes
1. Job/Employment satisfaction	–	–	–	–	4.61	C+	5.00	B
2. Great place to work	5.39	B+	6.00	A	4.91	B−	5.00	B
3. Pay/Benefits	4.85	B	5.00	B	4.32	C+	4.50	C+
4. Employment security	5.08	B	5.00	B	4.76	B−	5.00	B
5. Friendly/Sociable workplace environment	5.22	B+	5.00	B	4.92	B−	5.00	B
6. Employee advancement opportunities	4.88	B	5.00	B	4.03	C	4.00	C
Section II subtotal	5.08	B	5.20	B+	4.59	C+	4.75	B−
III. Internal management capabilities/performance
1. Quality of management team	5.19	B+	5.00	B	4.35	C+	4.00	C
2. Quality of people management	5.09	B	5.00	B	4.14	C	4.00	C
3. Confidence/Trust in management	5.16	B	5.00	B	4.10	C	4.00	C
4. Effectively deals with underperformer/problem employees	4.48	C+	5.00	B	3.52	D+	4.00	C
Section III subtotal	4.98	B	5.00	B	4.03	C	4.00	C
IV. Internal workforce capabilities/performance
1. Quality of workforce	5.40	B+	5.00	B	4.85	B−	5.00	B
2. Engagement	5.22	B+	5.00	B	4.67	B−	5.00	B
3. Provided with good job resources	5.26	B+	5.00	B	4.59	C+	5.00	B
4. Low employee turnover	3.82	C−	4.00	C	4.77	B−	5.00	B
Section IV subtotal	4.92	B	4.75	B−	4.72	B−	5.00	B
V. Employer–employee relations and climate
1. Employees treated fairly/humanely	–	–	–	–	4.57	C+	5.00	B
2. Mgrs’ interest in hearing EEs’ gripes/problems	5.14	B	5.00	B	–	–	–	–
3. Relations between management and employees	5.28	B+	5.00	B	4.36	C+	4.00	C
4. Family/partnership feeling	4.98	B	5.00	B	4.14	C	4.00	C
5. Little conflict/infighting	3.97	C	4.00	C	4.14	C	4.00	C
6. Morale	5.21	B+	5.00	B	4.25	C	4.00	C
7. Employees’ connection/interest with what mgt. says/does	3.57	D+	3.00	D	3.68	C−	4.00	C
8. Employees collaborate/cooperate with managers	5.27	B+	5.00	B	–	–	–	–
9. Good workplace for women/minority/LGBT employees	–	–	–	–	5.00	B	5.00	B
Section V subtotal	4.77	B−	4.57	C+	4.31	C+	4.29	C+
VI. Employee relations practices
1. Employee voice/involvement in way work is done	5.15	B	5.00	B	4.35	C+	4.00	C
2. Employee listening and opinion methods	4.70	B−	5.00	B	3.83	C−	4.00	C
3. Effective dispute resolution	5.13	B	5.00	B	4.34	C+	4.00	C
4. Effective/Extensive internal communication	5.02	B	5.00	B	4.15	C	4.00	C
5. Collaborative/Commitment management style	5.03	B	5.00	B	3.81	C−	4.00	C
6. Positive forms of employee motivation	4.99	B	5.00	B	4.12	C	4.00	C
7. Value-added from HR function at organisation	5.48	B	6.00	A
Section VI subtotal	5.07	B	5.14	B	4.10	C	4.00	C
VII. Community’s Performance Outcome
1. Customer Satisfaction	5.20	B+	5.00	B	4.77	B−	5.00	B
2. Corporate Social Responsibility	5.08	B	5.00	B	4.55	C+	4.00	C
3. Flexible Work Arrangements	5.06	B	5.00	B	4.38	C+	4.00	C
Section VII subtotal	5.11	B	5.00	B	4.57	C+	4.33	C+
VIII. Overall	4.97	B	5.00	B	4.40	C+	4.41	C+

Source: Authors calculations from State of the Workplace Employment Relations Survey (SWERS) (2016).

Before we go into the key findings, it is worth noting that for all individual scores and the overall score, we focus on mean results and only discuss selected cases where median scores diverge.

Results and findings

What is evident from Table 2 is that employers and employees have rather different perspectives on the state of ER. Of the 33 individual indicators spanning both EMR and EME columns, employer and employee mean letter grades are not the same in ANY cases. For 28 indicators, the mean score given by employers is at least one plus/minus grade higher than given by the employees, and in only four cases is the employers’ mean score lower (company competitive position, employee turnover, conflict and infighting, and employee connection with what management says/does). So, employers have a more positive assessment of the state of the ER than do employees. The largest gap is for Section VI. Item 5 ‘collaborative/commitment management style’ where employer rating is B and employee rating is C−. Looking over the grades in the entire table, 31 of the 35 employer grades are in the B or A range while for employees only 11 grades are in the B or A range and disturbingly almost all (10) are a B minus.

The overall score for the state of ER from employers is 4.97, which translates into a report card grade of B. The score from employees is 4.40, which comes in as a C plus. When we look at the overall scores measured by the median instead of mean response (the median numerical score for each question cumulated and then averaged), neither of the grades change.

If one compares the EE and EMR scores in Table 2, two features of the table stand out, respectively, the higher mean scores given by employer respondents and, second, the larger dispersion in EE scores which range from 3.08 to 5.62 compared to EMR scores of 3.98 to 6.06. Using the employer data, the overall mean grade is B and only employer respondents in the top (90th) percentile have scores in the A range. The overall mean for employees is C+ with a more pronounced low end where scores in the bottom 30th percentile are in the D range.

These features are clearly evident when we plot the percentile distributions of overall scores for employers and employees. In Figure 1 (available in the Online Appendix), by comparing the bar graphs for employers and employees, we see the positive employer skew in scores relative to the employee responses. It is not until the 40th percentile of the employee scores, for example, that average employee scores surpass those of the employer’s bottom 10th percentile score. This is graphic evidence that employers and employees do not see the Australian employment relationship in the same way.

One note of caution in these preliminary findings is that our employer responses, because they are solicited by senior management, may be based on their assessment of the ‘core’ workforce for whom better workplace relations and practices are more common. That is, it may be that by construction the survey has found a divergence between employees and their employers due to the fact that key employee groups are treated better than others, so one would expect higher manager ratings (consistent with the results presented here). In practice, however, the wording of each sectional survey descriptor actually varied between asking senior managers/owners for assessments of the organisation as a whole versus focusing on the largest most representative employee group.⁵

So although the source of the employer–employee divergence in scores is definitely a legitimate issue, the fact that there were actually a greater number of scorecard questions that asked the senior manager to report on the company as a whole (Sections I, III, V and VII) versus modulating answers for the largest group of applicable employees (Sections II, IV and VI) lends some credence to the notion that there are ‘real’ perceptual differences between employers and employees as to the state of health of the Australian workplace. The possibility that these differences are reflective of real differences in treatment amongst employees within an organisation is equally not negated and is perhaps one reason why we see such noticeable differences in employers and employees.

ER scorecard differences by individual respondent characteristics (gender, education, occupation, and union status)

An important question is whether evaluations of the state of ER differ according to some key categories. Table 6 reports (available in the Online Appendix) results on socio-demographic differences by gender, education, occupation and union status.

For each of the 39 measures, a multiple-regression estimation was conducted with the item score as the dependent variable and a binary 0/1 variable as the key independent variable with 0 = respondents in the base group and 1 = respondents in the comparison group.⁶ For gender, base = female and comparison = male; for education, base = high school degree or less and comparison = above high school; for occupation, base = blue collar/service and comparison = white collar and for union status base = not covered by collective bargaining contract and comparison = covered by collective bargaining contract.⁷

Table 6 vertically displays the 39 report card questions with three different entries for each cell: not applicable (NA because the question is not asked in the survey), a dash (–) indicating no statistical difference in means across base/comparison groups (at 95% confidence level), and a number expressed at two decimal points which represents the difference in mean scores between comparison and base groups when the difference is statistically different at the 5% level. Thus, for example, the male–female entry for Section I, Question 5 under employee survey column means the average score given to this question is −0.23 lower when the respondent is male relative to female. In other words, female employees are significantly more likely to respond positively than men when asked about the operational efficiency of their workplace or company.

More generally, amongst the four respondent attribute columns, the number of statistically significant differences in mean scores on a high-low scale starts with occupation at top (32 out of 35 questions), union status second (18 out of 35), gender (11 out of 35), and education lowest (4 out of 35). The differences between each comparison grouping (e.g., male vs. female; HS or above vs. Below HS; White Collar vs. Blue Collar; and Union vs. Non-Union) are generally what would be expected.

Of the 32 numerical scores in the occupation column, only one is negative meaning that most white-collar workers view their workplaces and mangers more favourably than do blue-collar/service workers. Thus, white-collar respondents report, on average, higher job/work satisfaction, job security, trust in management, engagement and ER climate than blue collar counterparts. Similarly, union workers report higher pay and benefits and job security, as one might expect, but lower quality of management, confidence in management, fair treatment and ER climate scores than non-union respondents. In the employee and employer survey, men give more positive scores for most questions (where statistically different) but in the employer survey (column 1 of Table 7), gender makes an impact on fewer measures.

And while we feel it is appropriate not to focus on differences that are statistically insignificant, Table 6 results implicitly suggest that any statistically significant differences are meaningful from a practical perspective. For example, is a −0.18 coefficient (found between male vs. female for great place to work in Column 2 of Table 6) a meaningful difference in practice? Is it large enough for managers to be concerned with? We think, as a guide, a 0.18 difference is meaningful as it represents a 3% change in score relative to the 7-point scale responses and we think this meets a minimum standard of importance. For example the 0.45 difference between white-collar and blue-collar workers is reflective of a 7% difference with respect to the 7-point scale. These examples are perhaps at the low end of the importance spectrum since we also see differences (when looking at firm characteristics in Table 7) greater than 1, which are reflective of a full point (or 15% change) in the 7-point scale used.

ER scorecard differences by firm characteristics (firm size, industry, type and presence of worker representation)

Perhaps an even more compelling question is whether evaluations of the state of ER differs according to some key workplace characteristics. This was, after all, the finding of Bryson and Freeman (2013) when they estimated perceived workplace problems on the part of employees. In their study, workplace characteristics were the more important predictor of whether a worker thought their employer was ‘good’ or ‘bad’. We report scorecard results by firm size, industry, firm type and presence of collective representation for employers and employees, respectively.

Table 7 reports (available in the Online Appendix), on each of the 39 questions asked of employers, where we again (as in Table 6) conducted a statistical test to determine if the mean score reported by respondents differs according to each of the four firm/workplace attributes. For each question, a multiple-regression was estimated with the individual’s score as the dependent variable and a binary 0/1 variable as the independent variable with 0 = respondents in the base group and 1 = respondents in the comparison group. For firm size, base = small (fewer than 500 employees) and comparison = large (500 or more employees); for industry, base = all non-manufacturing industries and comparison = manufacturing; for type of firm, base = all not for profit enterprises and comparison = private for profit.

In terms of worker representation, several measures across the EMR and EME surveys are available. In the employer survey, because we asked managers whether their workplace was unionised, we have as base = no union present and comparison = union present. For employees, we have a question capturing the presence of a non-union employee consultative committee, so our base is = no non-union consultative committee and comparison = a non-union consultative committee is present. It should be noted that the question concerning the presence of internal consultation was asked of employers and employees and was not mutually exclusive of the presence of a union. In other words some unionised firms have joint consultative committees that function independently of the collective bargaining process and this question was designed to pick that up. This facet of the survey is exploited more fully in our regression results (see Tables 4, 5, 10 and 11).⁸

What did we find for employers?

In Table 7, we see that firm size is generally not a significant differentiator, but where it is significant, the coefficients are negative meaning that managers in large employers are less positive about the items on the scorecard than respondents in smaller enterprises. For example, the employee connection with what management says/does (item 7 under Section V) is highly negative (−0.45), suggesting that larger employers are having a harder time convincing employees that they are on the same page.

Unlike firm size, a column with a lot of significant results was firm-type (Column 3). Here we divided firms into private, for-profit firms and the rest, which included not-for-profit, government agencies and broader public sector employers like hospitals, schools and so on. Here the clear result is that employers in the private, for-profit sector are much more positive about the state of ER than employers elsewhere. Apart from worrying more about the company’s competitive position (−0.28), experiencing higher turnover and employee retention concerns (−0.50 and −0.58 respectively), and again sensing that there is a lack of connection/interest with what management says does (−0.43), all other responses tilted positive.

In the last column of Table 7 the results were mixed. Union presence was associated with the strongest negative relationship between management getting employees to connect with what management says and does (−1.04). It was also the most highly related to conflict (−1.02), and unionised employers were much more concerned about their competitive position (−0.20). On the other hand, employers in unionised firms were reporting higher positive employment growth (0.70), had more positive employee advancement opportunities (0.43) and felt (perhaps because of internal grievance procedures) that they had a better handle on dealing with problem employees.

What did we find for employees?

In Table 8 (available in the Online Appendix), we find that employees, not surprisingly, see the world in a much different way than their employers. For example, taking firm size (our first column) we find many more significant differences between large (500+ employees) and smaller (<500 employees) firms. This column is almost exclusively negative, suggesting that for most items, employees in large firms perceive management and company outcomes more negatively than similar employees in smaller firms. The one area where large firms ‘outperform’ smaller firms is in the last item on pay, larger firms pay more than smaller firms. Of note, as seen in Table 7, this is one area of congruence between management and employees.

As was the case for employers, firm type for employees was the most meaningful category, only this time the findings were completely the opposite of what senior management perceived. Recall that employer respondents in for-profit companies had a much rosier picture of the health of their workplaces than those in the non-profit and government sector. Not so for employees, in that respondents working in the private for-profit sector scored significantly lower than employee respondents in the non-profit and government sectors on almost all items, which is why the column entries are practically all negative.

Finally in the last column of Table 8, a really striking result – employees that had a non-union consultative committee were more positive on almost every aspect of the ER scorecard. The presence of a consultative employee system was by far the strongest (positive) correlate of scorecard success and merits further investigation. Indeed, we have a particularly interesting opportunity to look within the union and non-union sectors in order to move beyond simple union/non-union comparisons. Increasingly, researchers (see Black and Lynch, 2001; Gittell et al., 2004) have recognised that variations within each sector matter in this sense that it is the quality of union and non-union relationships that might be more important than the average effect of union presence per se.

Multivariate estimates of the seven ER scorecard outcomes

Since the differences noted in Tables 6 through 8 are based on separate regression equations for each individual survey item, we do not display the independent relationships between size of firm, profit/non-profit and any other relevant variables with our outcome measures. Thus, we have obscured the net effects of different variables once others are controlled for. This is now done in Tables 3 and 9. In both tables we create an average (unweighted) score based on all the items listed under each of the seven ER scorecard sections. So for example, under Section I Table 6 of the scorecard (e.g., Company/workplace Performance Outcomes) we have six items, each based on a question asked of employees and all measured on a 7-point scale. We now take those six items and estimate the average (unweighted) sectional score and call this ‘Company Performance’. This is now our single dependent measure of Company/workplace Performance, which is regressed against a series of independent variables that include the categories already employed in Tables 6 through 8 plus additional variables of interest. We do this for all seven scorecard sections.

In Table 3, using employee data only, we see seven columns, each representing a regression conducted for one of the seven summary scorecard measures. Most independent variables were explained in Tables 6 through 8 and are self-explanatory but at least two require some elaboration: Private sector actually refers to a binary variable capturing all non-federal, state and local government employees; and Major disruption refers to whether the company/workplace experienced a major disruptive change over the last several years, such as a restructuring, downsizing, merger-acquisition, or bankruptcy.

Table 3.

Estimates of ER scorecard outcomes by key individual and workplace characteristics, employees only.

	Company performance	Employee satisfaction	Management performance	Workforce performance	Employment climate	Employment practices	Community performance
[Mean Score = 1 to 7]	[4.46]	[4.59]	[4.03]	[4.72]	[4.31]	[4.10]	[4.57]
	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Male	−0.0774	−0.0378	0.0246	−0.0656	0.0575	0.199***	−0.0931
	(0.0483)	(0.0712)	(0.0522)	(0.0654)	(0.0672)	(0.0751)	(0.0711)
≤High school	0.00602	−0.148*	−0.0937	−0.0152	−0.126	−0.182**	−0.0492
	(0.0567)	(0.0855)	(0.0609)	(0.0783)	(0.0801)	(0.0882)	(0.0827)
White Collar	−0.0212	0.0583	0.0205	0.129*	0.129*	0.205***	0.0759
	(0.0502)	(0.0749)	(0.0539)	(0.0688)	(0.0708)	(0.0790)	(0.0757)
Age ≤45	0.0500	−0.192***	0.0274	0.0174	−0.00606	−0.352***	−0.0957
	(0.0501)	(0.0734)	(0.0545)	(0.0668)	(0.0713)	(0.0787)	(0.0740)
Married	0.0353	0.150**	0.0445	0.122*	0.126*	0.139*	0.145**
	(0.0470)	(0.0692)	(0.0522)	(0.0633)	(0.0656)	(0.0733)	(0.0697)
Tenure ≤10 yrs	−0.107**	−0.0771	−0.164***	0.00353	−0.152*	−0.170**	−0.0671
	(0.0542)	(0.0785)	(0.0591)	(0.0718)	(0.0783)	(0.0822)	(0.0770)
Private sector	0.0506**	0.0481	0.0232	0.0330	0.0312	−0.00235	0.121***
	(0.0256)	(0.0396)	(0.0291)	(0.0364)	(0.0353)	(0.0408)	(0.0381)
For profit	0.341***	−0.0483	−0.0269	−0.0592	−0.0232	0.0458	0.00714
	(0.0525)	(0.0757)	(0.0556)	(0.0699)	(0.0714)	(0.0818)	(0.0755)
Large organisation	−0.0784*	−0.0769	−0.116**	−0.0875	−0.175***	−0.126*	−0.0931
	(0.0472)	(0.0691)	(0.0509)	(0.0629)	(0.0661)	(0.0725)	(0.0700)
Major disruption	−0.422***	−0.322***	−0.191***	−0.204***	−0.489***	−0.341***	−0.128*
	(0.0497)	(0.0707)	(0.0528)	(0.0629)	(0.0680)	(0.0738)	(0.0697)
Unionised	−0.0653	−0.0527	−0.120**	0.0404	−0.252***	−0.0242	−0.0906
	(0.0502)	(0.0727)	(0.0552)	(0.0651)	(0.0709)	(0.0779)	(0.0723)
Committee	0.443***	0.874***	0.606***	0.794***	0.820***	1.132***	0.816***
	(0.0481)	(0.0695)	(0.0512)	(0.0625)	(0.0662)	(0.0723)	(0.0699)
Constant	4.013***	4.446***	4.271***	4.297***	4.313***	4.092***	3.960***
	(0.120)	(0.182)	(0.131)	(0.163)	(0.164)	(0.185)	(0.173)
Observations	1,225	1,225	1,225	1,225	1,225	1,225	1,225
R-squared	0.155	0.146	0.124	0.140	0.166	0.213	0.131

Robust standard errors in parentheses; Significance levels denoted by *p < 0.1; **p < 0.05; ***p < 0.01.

Notes: All dependent variables are (unweighted) average scale indices of the section categories found in Table 2. Individual employee responses under each sectional measure were averaged to create a single 7-point score for each respondent. Independent variables are all dichotomous (or dummy) measures taking on the value ‘1’ if the category in question is present for a respondent and ‘0’ otherwise. The excluded reference categories are female; ≤ High School; Blue Collar; ≤ Age 45; Non-immigrant; Unmarried; ≤ Tenure 10 years; Public Sector (Government employees only); Not-for profit; Small organisation (<500 employees); No major disruption at company in recent past; Non-manufacturing; No-union coverage/membership; No consultative committee. For space constraints Manufacturing and Immigrant status coefficients not reported but available upon request.

Table 4.

Estimates of ER Scorecard Outcomes by Presence of Union and Non-Union Consultative Committees, Employees only.

	Company performance	Employee satisfaction	Management performance	Workforce performance	Employment climate	Employment practices	Community performance
[Mean Score = 1 to 7]	[4.46]	[4.59]	[4.03]	[4.72]	[4.31]	[4.10]	[4.57]
	(1)	(2)	(3)	(4)	(5)	(6)	(7)
[Non-Union × No committee]
Non-Union × Committee	0.408***	0.831***	0.569***	0.783***	0.759***	1.061***	0.763***
	(0.061)	(0.0893)	(0.0648)	(0.0825)	(0.0858)	(0.0908)	(0.0902)
Union × No committee	−0.111	−0.110	−0.169**	0.0254	−0.331***	−0.117	−0.160
	(0.077)	(0.109)	(0.0839)	(0.0942)	(0.102)	(0.113)	(0.107)
Union × Committee	0.383***	0.828***	0.492***	0.836***	0.577***	1.118***	0.733***
	(0.061)	(0.0919)	(0.0668)	(0.0845)	(0.0874)	(0.101)	(0.0926)
Controls included^a	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Constant	4.028***	4.465***	4.287***	4.302***	4.340***	4.123***	3.984***
	(0.121)	(0.184)	(0.132)	(0.166)	(0.166)	(0.185)	(0.175)
Observations	1,225	1,225	1,225	1,225	1,225	1,225	1,225
R-squared	0.155	0.146	0.124	0.140	0.167	0.214	0.132

Robust standard errors in parentheses; Significance levels denoted by *p < 0.1; **p < 0.05; ***p < 0.01.

Controls are identical to those used in Table 9.

Table 5.

Estimates of company performance and employee satisfaction by presence of union and high-performance work systems, employees only.

	Company performance	Company performance	Employee satisfaction	Employee satisfaction
[Mean Score = 1 to 7]	[4.46]	[4.46]	[4.59]	[4.59]
	(1)	(2)	(3)	(4)
[Non-Union × Low performance HRM]
Union × Low performance HRM	−0.225***	−0.157***	−0.0521	−0.0757
	(0.0587)	(0.0586)	(0.0691)	(0.0705)
Non-Union × High performance HRM	0.568***	0.542***	1.577***	1.563***
	(0.0446)	(0.0454)	(0.0530)	(0.0540)
Union × High performance HRM	0.442***	0.507***	1.624***	1.567***
	(0.0474)	(0.0499)	(0.0580)	(0.0605)
Controls included^a	No	Yes	No	Yes
Constant	4.143***	3.881***	3.877***	3.796***
	(0.0318)	(0.111)	(0.0415)	(0.130)
Observations	2,002	1,903	2,002	1,903
R-squared	0.132	0.173	0.422	0.435

Robust standard errors in parentheses; Significance levels denoted by *p < 0.1; **p < 0.05; ***p < 0.01.

Controls are identical to those used in Table 9 but also include full measures of industry and occupation.

What do we find in our regression estimates of summary employee scorecard outcomes?

The major news is that two workplace/company level variables – Major disruption and Consultative Committee – have a significant and meaningful association in every regression. The presence of a consultative committee is uniformly positive with respect to all seven outcomes. There is a positive association with company economic performance (column 1) through to community performance (column 7). The strongest association is linked to progressive employment practices (column 6) with a coefficient 1.132 (or 30% higher relative to the mean score of 4.10). This means that employees with consultative committees scored 1.132 higher on employment practices than observably similar employees without them. Equally strong were the coefficients associated with ER climate (column 5) and employee satisfaction (column 2), each with score coefficients of 0.820 and 0.874, respectively, corresponding to a 20% higher score for employees with consultative committees versus those without. Interestingly there is even a noticeable association between consultative committees and company performance (column 1), with a coefficient of 0.443 corresponding to an 11% higher performance rating for employees with committees present versus those without.

The reverse association across all seven scorecard outcomes was found with employees experiencing a major disruption (e.g., restructuring, downsizing, merger-acquisition, or bankruptcy) at their company/workplace. All associations were negative and strongly so, with company performance (column 1) being −0.422 (or 11% lower relative to the mean company performance score); employee satisfaction (column 2) being −0.322 (or 8% lower); ER climate (column 5) being −0.489 (or 11% lower); and assessments of progressive employment practices (column 6) being −0.341 (or 8% lower) for employee respondents facing a major disruption.

Apart from those two variables all other variables were less uniformly associated (positively or negatively) with the seven outcome measures. The next most significant predictor was working for a large organisation (employing 500 or more employees), which had four out of seven significant negative relationships – it seems large organisations (after controlling for observable characteristics of workers and workplaces) generate a negative association with employment scorecard outcomes especially ER climate (column 5).

Being a union member was associated negatively with most scorecard outcomes, though only two outcomes were significant (i.e., assessment of management performance (column 3) and ER climate (column 5)). Tenure (i.e., more than 10 years with the same company) and age (i.e., greater than 45 years of age) were negatively associated with scorecard outcomes, but only significantly so in the case of tenure in terms of assessments of company performance (column 1), management performance (column 3), ER climate (column 5) and progressive employment practices (column 6). For older workers, employee satisfaction (column 2) and employment practices (column 6) were negatively associated with age. More experienced workers perhaps have greater expectations of what an employer can do and/or simply have a more jaundiced view of their employers than younger workers.

Lastly, for-profit companies were (perhaps unsurprisingly) associated positively with company performance (column 1) but almost nothing else. Private sector firms were significantly associated with community performance (column 7) but only marginally so (a coefficient of 0.121 or 3% more than a public sector organisation).

In Table 9 (available in the Online Appendix), using employer data only, we see the same seven columns, each representing a regression conducted for one of the seven summary scorecard measures. Most independent variables were explained in Tables 6 through 8 and are self-explanatory, but at least one new variable requires some elaboration: Top management refers to only the highest level and CEO/Owner designations in our sample of senior managers. If a manager responded to the employer questions and self-identified as middle-to-lower level management, we coded them as 0 for the purposes of this analysis. This was done to see if there was any variation within the management respondents based on level in the organisation.

What significant variables of interest emerge in regression estimates of our employer scorecard outcomes?

It turns out where management sits in the organisational hierarchy seems to have a large and positive correlation with respect to the way it perceives the workplace and its outcomes. Uniformly, top management is associated positively with all seven ER scorecard outcomes relative to lower level management. The coefficients range from 0.286 for company performance (column 1) to 0.458 for employment practices (column 6).

The second largest predictor of ER outcomes was private sector with five significantly positive coefficients. Next was being a for profit employer, which was positive and significantly associated with company performance (column 1), employee satisfaction (column 2) and management performance (column 3). Large organisations, which had a modestly negative association amongst employee responses on a number of outcomes, was only significantly negatively associated with one outcome, ER climate (column 5).

Finally, and perhaps most interestingly, union presence was not seen by employers as a significantly negative predictor of scorecard outcomes. In fact, three important outcomes (company performance, management performance, and employment practices) were all positively and significantly associated with union presence. This is in keeping with some well-known findings from the ER literature showing that managers familiar with unions tend to have a less negative view of unions than managers with no contact or experience with unions (Campolieti et al., 2013). Perhaps, as noted by previous IR scholarship (Verma, 2005), unionised employers’ more positive impressions of unions are rooted in real evaluations of how a company performs with a union in place.

Examining the ‘quality’ of union and non-union employment relationships

In Table 8 it was clear that the presence of some employee representation was by far the strongest (positive) correlate of scorecard success. It would appear, based on these findings that we have a particularly interesting opportunity to look within the union and non-union sectors a bit more closely in order to expand upon the simple union/no union comparison. Increasingly, researchers (see Black and Lynch, 2001; Gittell et al., 2004) have recognised that variations within each sector matter, in the sense that it is the quality of union and non-union relationships that might be more important than the average effects of each category.

We approach this question in two ways. First we make use of the consultative committee question present in both EME and EMR surveys to identify presence of a non-union voice channel at work. For employees, we know whether they are covered by a union collective agreement. We do not know if the company is entirely unionised but we infer from the employee union membership/coverage question that there is some form of bargaining present at that workplace/company. So we can therefore code four mutually exclusive categories of employee voice for each respondent: (i) non-union × no-consultative committee; (ii) non-union × consultative committee; (iii) union × no consultative committee; and (iv) union × consultative committee. We can think of this as one categorical variable that scales essentially from little or no formal voice (e.g., no union × no committee) to most formal voice (union × committee). In-between are the intermediate categories of moderate formal voice provided by just having the committee or the union present.

For the employer respondents a similar categorical variable is constructed using the same non-union consultative committee question present in the EMR survey. With regards to union presence in the employer survey, we have knowledge of whether a union is present or not. We ask employers to estimate whether none or some positive number of the company is unionised. For those responding ‘0’ we code these as non-union employers and 1 otherwise.

We then regress this new four category workplace voice measure (with no union × no consultative committee as the excluded reference or ‘base’ category) against the seven aggregated ER outcome measures used in regressions found in Tables 3 and 9 and with the same controls.

Our expectation would be that moving from no voice (No union × No consultative committee) to most voice (Union × Consultative committee) would engender rising positive coefficient estimates. The question of quality however is likely answered in our in-between categories of either having a union or consultative committees but not both. We can also evaluate whether presence of a consultative committee acts as a ‘moderator’ of union presence or vice versa.

In Table 4 we report the employee findings.

It is important to compare these results with those found for unionisation and consultative committees found at the bottom of Table 3, as those initial estimates represent the ‘average’ effect of having each of these institutions present at the workplace, without taking account of the interactions between the two. Take columns (3) and (5), which estimate employee evaluations of managerial performance and ER climate respectively. In both cases these outcomes were significantly and negatively related to unionisation. These were in fact the only two outcomes that were significantly related to union presence. In Table 4 the presence of a union (alone) is again significantly and negatively associated with managerial performance and ER climate. However after controlling for the simultaneous presence of union and a non-union committee, the co-efficient turns even more strongly negative (−0.120 in Table 3 versus −0.169 in Table 4 for management performance and −0.252 in Table 3 versus −0.331 in Table 4).

The reason for these negative results is that the ‘average union’ effect being picked up in Table 3 was actually masking two effects: the highly negative relationship between being unionised with no formal consultative system, and the highly positive effect of being unionised in a workplace that has a consultative committee system. This is picked up in the last row of Table 4 which shows that across the seven ER categories, the effect of having a union and a consultative committee is strongly positive and significant. This is all being measured against the base category of no formal voice (i.e., Non-union × No committee). Interestingly non-union firms with a committee seem to have equivalent coefficients to those employees responding from workplaces with both voice systems, suggesting that the presence of consultative committees has a strong and positive moderation effect on union presence.

One can interpret this in several ways but it seems plausible that we are picking up ‘deep’ organisational features in the workplace-voice variables captured here. In other words, firms offering up voice on their own volition (Non-union × Committee) and in combination (Union × Committee) are likely a ‘good high-performing employer’ even in the absence of these two intuitions, whereas a unionised workplace that offers up no consultative system is perhaps emblematic of the ‘bad performing employer’ that is only being held in-check by union presence. Bryson and Freeman (2013) found that the accumulation of workplace problems was the strongest predictor of desired unionisation amongst workers. This would explain why employee assessments of management and climate are most negative in the union only category as compared to the no-voice reference category (No Union × No Committee).

In Table 10 (available in the Online Appendix) we report the employer findings.

In the equivalent employer Table 9 findings, where we only measured the average effect of union presence with respect to the seven ER scorecard outcomes, union presence was associated positively with all seven ER scorecard outcomes. Here the positive relationship between unionisation and ER outcomes continues, but significance is lost in all but one outcome estimate (company performance). However, by examining the complete set of union and consultative committee categorisations – with No union × No committee set as the reference category – we see that the ‘quality’ of union–management relations seems to definitely matter. Workplaces with unions and consultative committees are associated significantly and positively across all seven outcome measures. The positive association is also present in workplaces without unions but which have consultee committees. This is an indication that union presence is moderated strongly by the presence of consultative committees, which themselves may be proxies for the ‘health and quality’ of the union-management relationship at the firm.

Table 10 results indicate that Australian workplaces lacking in any formal voice score the poorest on key elements of ER scorecard health. Since employers in the survey are reporting ‘factually’ to the presence (or lack therefor) of these institutions at the workplace and then separately in the survey on their perception of the ER health and performance of the organisation, it seems plausible to conclude that this lack of employee voice is ‘causally’ linked with negative ER outcomes, rather than just an association being made in the minds of our employer respondents.

High-Road versus Low Road HRM

We also mentioned there were two ways to look at the ‘quality of employment relations’ in union and non-union workplaces. The second approach is to look at the category VI scorecard measures (i.e., Employee Relations Practices) and create something akin to a high-performance (i.e., transformed/high-road employment relationship) measure. We do this by taking the seven practices listed under category VI and estimating the median score for this category. Respondents with scores above the median are placed into a high-performance category and those below into a low performance/low road work system category. We then create a four way categorical variable that is combined with union and non-union status. Our excluded reference category is Non-union × Low performance HRM.

In Table 5 we show the employee estimates for two outcomes; one of direct interest to employers (company performance) and one for employees (employee satisfaction). Here again we find that it matters whether a unionised employee is responding from a workplace that has employment practices that are considered ‘high road’ versus an employment situation that is stuck in the ‘low road’ approach. A unionised employee in a low performance workplace is significantly associated with negative company performance (relative to the respondent in a non-union × low performance system). However, being unionised is strongly and positively associated with company performance and employee satisfaction when combined with high-performance employment systems. There is no statistical difference between union and non-union high-performance workplaces, suggesting that unions are not inhibitors of high performance effects but neither are they capable, on their own, of reversing the ill associations of a ‘low road’ employer with respect to key ER outcomes.

In Table 11 (available in the Online Appendix) we show the employer respondent estimates for the same outcomes above. The difference is that we create our union variable from the employer response to the presence of unionised employees at the workplace/company; i.e., those employers responding with 0 union members were treated as non-union and the non-zero estimates assigned a value of 1 and treated as having union presence at the workplace. We find that the interaction of unions and high-performance employment systems is strongly and significantly associated with company performance and employee satisfaction. It should be noted that these associations are some of the largest in the analyses undertaken. So for example, the Union × High Performance HRM coefficient for employee satisfaction (with all controls included) is 1.415 or 30% greater (relative to the mean satisfaction score) than the excluded reference category of no union presence and low road employment practices.

The state of Australian ER: Healthy or needing a boost?

First, it should be no surprise to students of industrial relations that different parties assess the overall state of ER differently. Unlike unitarist approaches (i.e., managerial analyses) to the study of work, a long line of critical and pluralistic ER scholarship has acknowledged that the employment relationship engenders differing interests and can sometimes produce conflict. It is also probably not a surprise to find that on nearly all measures, senior managers (employers) gave higher scores on ER than employees. It is true that managers would tend to believe they are good leaders and communicate well (illusory superiority principle), however, the rating is not just about themselves but about relations with employees where one might expect to see a less rosy picture. One perspective might put greater trust in managerial views given their broader perspective, but another might say that ER is only part of their overall set of responsibilities, and a second order one at that, whereas employees’ lived experience provides more accurate ratings. Any explanation about the underlying cause(s) of differences between manager and worker ratings is necessarily speculative. However, one possibility is that lower employee ratings reflect changes to the nature of work and its regulation discussed earlier. The interaction of neo-liberalism and changes in the institutional ER regulatory environment, which are supportive of employer interests, have been felt broadly in Australia and have been shown to diminish employee power and create wage and working-condition disadvantages (Watson, 2016). Casualisation has become an undesirable, yet entrenched feature of employment (Watson, 2013).

A striking feature of our scorecard is what the grades are. We found scores from the D to A range across individual indicators; however, the mean and median scores/grades for the seven main ER functional dimensions were consistently in the B range for employers and C range for employees. It is also worth noting that the areas where employees give the lowest grades relate to employee connection and interest in what management says (C−), management listening/opinion methods (C−), and collaborative/commitment management style (C−). Meanwhile on the employer side, lowest grades were given to employee connection with what management says/does (D+), little conflict or infighting (C) and low employee turnover (C−). These grades are cause for concern, for they surely point to a strong level of disconnect in certain areas of the employment relationship. If, as Boxall (2014) argues, building alignment with workers is the key strategic imperative to gaining competitive advantage, then these results are troubling from a performance-management standpoint.

In our scorecard we also provided a bottom line assessment. The employers rating of the health and performance of the ER is, on average, a B (mean/median), while employee’s rate it as a C+ (mean/median). If both EE and EMR bottom line grades were combined and equally weighted, we would be left with a B- result. This is a somewhat uninspiring scorecard for Australian workplaces, which we are told need to move up the value chain given the proximity to regional markets with low labour costs. If ‘high performance work systems’ are here in Australia, they are clearly not a significant enough part of the market in terms of boosting these overall numbers. That said, in our regression estimates, which provide a picture of what is significantly associated with our ER scorecard measures, we do find considerable variation within union and non-union sectors. Specifically, we find that high performance work systems (where they exist) are associated – even in the presence of a union – with higher employee satisfaction and company performance scores. This is in keeping with what some in the literature have stressed is the key takeaway from looking inside the ‘black box’; namely that variations within multiple sectors are likely to be more important than the average effects within a single group.

Conclusion

This article reports the findings from nationally representative surveys of Australian employers and employees. The results offer mixed evidence and implications for the state of the employment relationship in Australian workplaces. Some companies and workplaces rate relatively high and others relatively low on ER health and performance, while the bulk cluster in a middle category rated at a B level by employers and C level by employees. By traditional grading standards, a C is average and a B is modestly above average, which by the very meaning of ‘average,’ is where we might well expect organisations to cluster. On the other hand, grades of C and B are not encouraging because of the large gap between the reality of just ‘average’ health and performance and the potential of high health and performance in the A range. In a world of global competition, average or modestly above-average performance in EER may signal danger ahead for all stakeholders. Our article also provided some detailed evidence showing that the presence of formal voice at the workplace matters and that the quality of that workplace voice – be it union or non-union – plays by far the most significant and important role in determining employee and employer perceptions of the state of ER health and self-reported performance. We hope that our scorecard measures and our findings regarding voice will be seen as having wide applicability and a robust measure of ER to be tested in other contexts.

Supplemental Material

Supplemental material for Taking the pulse at work: An employment relations scorecard for Australia

Supplemental material for Taking the pulse at work: An employment relations scorecard for Australia by Adrian Wilkinson, Michael Barry, Rafael Gomez and Bruce E Kaufman in Journal of Industrial Relations

Footnotes

Acknowledgement

We would like to thank Guenther Lomas for his excellent RA support.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: the ARC (DP140100194) SSHRC (435-2015-0801), and from the Innovation Resource Center for Human Resources (IRC4HR).

Notes

Biographical notes

Adrian Wilkinson is Professor and Director of the Centre for Work, Organisation and Wellbeing at Griffith University, Australia. Prior to his 2006 appointment, Adrian worked at Loughborough University in the UK where he was Professor of Human Resource Management from 1998, and Director of Research for the Business School.

Michael Barry is Professor of Employment Relations in the Griffith Business School. Michael was Foundation Head of the Department of Employment Relations and Human Resources at Griffith, from 2010 to 2016.

Rafael Gomez is Professor of Employment Relations at the University of Toronto and since 2015 has been the Director of the Centre for Industrial Relations and Human Resources.

Bruce E Kaufman is Professor at Georgia State University and Principal Research Fellow, Griffith Business School. Kaufman is co-editor of the annual research volume Advances in Industrial and Labor Relations. In 2009 he received the ‘Excellence in Teaching’ award from the Andrew Young School; he also received the ‘Alumni Distinguished Professor’ award – the highest honor given each year to a GSU faculty member.

References

Ackers

Wilkinson

(2003) Understanding Work and Employment: Industrial Relations in Transition, Oxford: Oxford University Press.

Australian Bureau of Statistics (ABS) (2014) Employee, Benefits and Trade Union Membership, Australia, August 2013. Canberra: Australian Bureau of Statistics. Australian Financial Review, 22 October, 55.

Australia Bureau of Statistics (2016) 6333.0-Characteristics of Employment, Data Cube 6. Available at: http://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/6333.0August%202016?OpenDocument (accessed 20 October 2017).

Beatty

Huselid

Schneier

(2003) New HR metrics: Scoring on the business scorecard. Organizational Dynamics 32(2): 107–121.

Becker

Huselid

(2006) Strategic human resources management: Where do we go from here? Journal of Management 32(6): 898–925.

Black

Lynch

(2001) How to compete: The impact of workplace practices and information technology on productivity. Review of Economics and Statistics 83(3): 434–445.

Boxall

(2014) The future of employment relations from the perspective of human resource management. Journal of Industrial Relations 56(4): 578–593.

Boxall

Purcell

(2011) Strategy and Human Resource Management, New York, NY: Palgrave Macmillan.

Bray

Underhill

(2009) Industry differences in the neoliberal transformation of Australian industrial relations. Industrial Relations Journal 40: 372–392.

10.

Bryson

Freeman

(2013) Employee perceptions of working conditions and the desire for worker representation in Britain and the US. Journal of Labor Research 34(1): 1–29.

11.

Campolieti

Gomez

Gunderson

(2013) Managerial hostility and attitudes towards unions: A Canada–US comparison. Journal of Labor Research 34(1): 99–119.

12.

Commission on Industrial Relations (CoIR) (1916) Final Report and Testimony. Vol. 1. Washington: GPO.

13.

Commons J (1934) Myself, Madison: University of Wisconsin Press.

14.

Commons J (1935) Communism and collective democracy. American Economic Review 25(2): 212–223.

15.

Derber

(1967) Research on Labor Problems in the United States, New York, NY: Random House.

16.

Dodge

(2006) The Oxford Dictionary of Statistical Terms, 6th edn. Oxford: Oxford University Press.

17.

Dolman

(2009) What happened to Australia’s productivity surge? The Australian Economic Review 42(3): 243–263.

18.

Fair Work Commission (2015) Annual Report 2014–15, Melbourne: Commonwealth of Australia.

19.

Feldman

(1928) Survey of Research in the Field of Industrial Relations, New York, NY: Social Science Research Council.

20.

Findlay

Thompson

(2017) Contemporary work: Its meanings and demands. Journal of Industrial Relations 59(2): 122–138.

21.

Forth

McNabb

(2008) Workplace performance: A comparison of subjective and objective measures in the 2004 Workplace Employment Relations Survey. Industrial Relations Journal 39: 104–123.

22.

Freeman

Boxall

Haynes

(2007) What Workers Say: Employee Voice in the Anglo-American Workplace, Ithaca, CA: Cornell University Press.

23.

Gittell

VonNordenflycht

Kochan

(2004) Mutual gains or zero sum? Labor relations and firm performance in the airline industry. Industrial Labor Relations Review 57(2): 163–180.

24.

Isaac

Lansbury

(2005) Labour Market Deregulation: Rewriting the Rules, Sydney: Federation Press.

25.

Kaplan

Norton

(2001) The Strategy-focused Organisation: How Balanced Scorecard Companies Thrive in the New Business Environment, Boston, MA: Harvard Business School Press.

26.

Kaplan R and Norton D (2007) Using the balanced scorecard as a strategic management system. Harvard Business Review July/August: 150–161.

27.

Kaplan

Norton

(2010) Kaplan and Norton on Strategic Management, Harvard: Harvard Business Publishing.

28.

Kaufman

(1993) The Origins and Evolution of the Field of Industrial Relations in the United States, Ithaca, NY: ILR Press.

29.

Kaufman

(2004a) The Global Evolution of Industrial Relations, Geneva: ILO.

30.

Kaufman

(2004b) Theoretical Perspectives on Work and the Employment Relationship, Urbana-Champaign, IL: LERA.

31.

Kaufman

(2010) The theoretical foundation of industrial relations and its implications for labor economics and human resource management. Industrial and Labor Relations Review 64(1): 74–108.

32.

Marchington

Wilkinson

Donnely

et al. (2016) Human Resource Management at Work, 6th edn. London: Chartered Institute of Personnel and Development.

33.

Mitchell

Gahan

Stewart

et al. (2010) The evolution of labour law in Australia: Measuring the change. Australian Journal of Labour Law 23: 61–93.

34.

Pekarek

Gahan

(2016) Unions and collective bargaining in Australia in 2015. Journal of Industrial Relations 58(3): 356–371.

35.

State of Workplace Employment Relations Survey (SWERS), (2016). Available from authors upon request.

36.

Van Wanrooy

Oxenbridge

Buchanan

et al. (2007) Australia at work: The benchmark report, Sydney: Workplace Research Centre.

37.

Verma

(2005) What do unions do to the workplace? Union effects on management and HRM policies. Journal of Labor Research 26(3): 415–449.

38.

Watson

(2013) Bridges or traps? Casualisation and labour market transitions in Australia. Journal of Industrial Relations 55(1): 6–37.

39.

Watson

(2016) Wage inequality and neo-liberalism: The Australian experience. Journal of Industrial Relations 58(1): 131–149.

40.

Webb

(1897) Industrial Democracy, London: Longman, Greens.n.

41.

Wilkinson

Bailey

Mourell

(2009) Australian industrial relations in transition. Industrial Relations Journal 40(5): 358–371.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.13 MB