P a g e | 1

THE INFLUENCE OF SOCIAL CAPITAL THROUGH NETWORK TIES ON HUMAN RESOURCE SELECTION POLICIES AND DECISION MAKING

Sunia Vosikata

A thesis in fulfillment of the requirements for the degree of Doctor of Philosophy

School of Business University of New South Wales UNSW Canberra March 2014 P a g e | 2

THE UNIVERSITY OF NEW SOUTH WALES

Thesis/Dissertation Sheet Surname or Family name: Vosikata

First name: Sunia Other name/s:

Abbreviation for degree as given in the University calendar: PhD Faculty: UNSW School: Business Canberra

Title: The Influence of Social Capital though Network Ties on Human Resource Selection Policies and Decision Making in the Fiji Public Service”

Abstract

The impact of social capital through network ties on human resources (HR) selection decisions and policies has not been adequately explained in much of the debate over the operation of state institutions and the public service. Social capital issues in Fiji have gradually shaped HR selection decisions within the Fijian Public Service (FPS) ever since

Fiji became a colonised territory in 1874. The influence continued to the pre and post 1970 independence era and materialised its ugly head to aggressively influence social and political stability from 1987. Within the transitional era, human resource management (HRM) practices, such as HR selection decisions, have been influenced by social networks and dominant social groups. This statement of this thesis study was declared as follows: “Social capital through network ties influence HR selection policies and decisions within the Fiji Public Service”. Specifically, distinctive pockets of instrumental and affective ties have influenced HR selection decisions since networkings informally infiltrate and corrupt the merits of contemporary HR selection framework, tarnishing the image of the FPS. However, the influences of instrumental ties are more obvious than affective ties.

Instrumental based societies and networkings are governed through obligations that are mutual and contractual, whilst affective ties oriented ones are governed by established norms of social obligations through status and seniority. The study has also revealed that key heterogeneity variables and demographics of the FPS correlate with a sustained level of social capital and networking which has been influencing FPS employees’ perception of justice on the meritocracy of HR selection decisions. Of significance here is ethnicity. FPS employees have also highlighted their perception of injustices on the application of strategic remuneration policies and the ways in which HR processes and polices are generally administered. The influence of the current social, political and economic environment may be crucial here. This would also place FPS managers in in the position of not making unitary decisions about HR selection. In general, HR selection for entry positions of the FPS do not generally correlate with the required HR selection meritocracy levels, because the application of HR selection policies hasn’t attained a desired industry-driven level, in view of national disruptions as stated. All in all, social capital through network ties has indeed influenced HR selection policies and decisions within the FPS.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

…………………………………………………………… ……………………………………..……………… ……….……………………...…….… Signature Witness Date The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

P a g e | 3

ABSTRACT

The impact of social capital through network ties on human resources (HR) selection decisions and policies has not been adequately explained in much of the debate over the operation of state institutions and the public service. Social capital issues in Fiji have gradually shaped HR selection decisions within the Fijian Public Service (FPS) ever since Fiji became a colonised territory in 1874.

The influence continued to the pre and post 1970 independence era and materialised to aggressively influence social and political stability from 1987. Within the transitional era, human resource management (HRM) practices, such as HR selection decisions, have been influenced by social networks and dominant social groups. This statement of this thesis study was declared as follows:

“Social capital through network ties influence HR selection policies and decisions within the Fiji

Public Service”. Specifically, distinctive pockets of instrumental and affective ties have influenced

HR selection decisions since networkings informally infiltrate and corrupt the merits of contemporary

HR selection framework, tarnishing the image of the FPS. However, the influences of instrumental ties are more obvious than affective ties. Instrumental based societies and networkings are governed through obligations that are mutual and contractual, whilst affective ties oriented societies are governed by established norms of social obligations through status and seniority. The study has also revealed that key heterogeneity variables and demographics of the FPS correlate with a sustained level of social capital and networking which has been influencing FPS employees’ perception of justice on the meritocracy of HR selection decisions. Of significance here is ethnicity. The influence of the current social, political and economic environment may be crucial here. This would also place

FPS managers in in the position of not making unitary decisions about HR selection. In general, HR selection for entry positions of the FPS do not generally correlate with the required HR selection meritocracy levels, because the application of HR selection policies has not attained a desired industry-driven level, in view of national disruptions as stated. All in all, social capital through network ties has indeed influenced HR selection policies and decisions within the FPS.

P a g e | 4

TABLE OF CONTENTS

Chap 1 Introduction Page 1.1 Background to the research 16 1.2 Aims and objectives 18 1.3 Research Problem 21 1.4 Justification for the research 24 1.5 Methodology 26 1.6 Outline of the thesis 27 1.7 Definitions 30 1.8 Delimitations of scope of the study 34 1.9 Conclusion 34

Chap 2 Literature Review 36 2.1 Introduction 38 2.2 Human Resource Management (HRM) and 45 the Fiji Public Service 2.3 Human Resource selection and the Fiji Public 52 Service 2.4 The strategic approach of human Resources 56 management and the Fiji Public Service 2.5 Social Capital and the Fiji Public Service 62 2.6 Networking Ties 68 2.7 Justice and Employee Discretion in the Fiji 71 Public Service 2.8 Conclusion 77

Chap 3 Methods 80 3.1 Introduction 82 3.2 Research questions, units of analysis and data 83 3.3 Response rate 85 3.4 The embedded case study 87 3.4.1 Case study organization sample 90 3.5 Data collection methods 91 3.6 Design and administration of data collection 92 methods 3.7 Qualitative data analysis 97 3.7.1 Qualitative data analysis methods 97 3.7.1.1 Pattern Matching 97 3.7.1.2 Explanation Building 98 3.7.1.3 Time Series Analysis 99 3.8 Quantitative data analysis 99 3.8.1 Survey data analysis 99 3.8.2 Method 1: Descriptive Analysis 101 3.8.3 Method 2: Factor Analysis 101 3.8.4 Rules of Thumb for retaining factors 106 3.8.4.1 Rule of Thumb 1 – Kaiser Guttman Criterion 106 3.8.4.2 Rule of Thumb 2 – Gorsuch’s subjective 107 Scree Test 3.8.5 Rotation 108 3.8.5.1 Retaining factor loading greater than an 110 arbitrary threshold 3.9 Ethical considerations 110 3.10 Conclusion 111

P a g e | 5

Chap 4 The Historical, Political and Social context 112 of Human Resources Selection Decisions in the Fiji Public Service 4.1 Introduction 114 4.2 Historical background of labour and 115 networking in Fiji 4.3 The Role of Chiefs in the Fiji Public Service 116 4.4 Indian Immigrants and Indenture 118 4.4.1 End of Indenture 120 4.5 Fiji Independence 1970 122 4.6 The political context of employment 124 4.7 Political context and HRM Challenges in the 128 Fiji Public Service 4.8 The Legislative Charter of Fiji and its 129 autonomy 4.9 The Role of the Fiji Military Forces in Fiji’s 131 political Context 4.10 Demographic Structure of Fiji – 2011(base 135 year) 4.11 Social Networking in Fiji 136 4.11.1 Network ties in Indian Social Structure 138 4.11.2 Network ties in Fijian Social Structure 140 4.12 Demographic Structure of Fiji Public Service 142 - 2011 4.13 Conclusion 144

Chap 5 The strategic approaches of human 145 resource selection within the Fiji Public Service 5.1 Introduction 147 5.2 Response rate 149 5.3 Archival data and publicly available 150 documents 5.4 Human Resources Management in the Fiji 151 Public Service 5.4.1 Mimicking of western HRM practices as a 155 challenge facing the Fiji Public Service 5.4.2 Management of the new Republic of Fiji 156 5.4.3 Cultural constraints 158 5.4.4 Pressure for public service reform from 162 developing economies and donor agencies 5.4.5 Changing the mindset of Fiji public servants 164 and effective management of the Fiji Public Service 5.4.5.1 Fijian perceptions on social capital 166 5.4.5.2 Indian perceptions on social capital 171 5.5 Human resources selection and the Fiji Public 175 Services 5.6 The strategic approach of Human Resources 184 Management and the Fiji Public Service 5.6.1 Choices about particular HR policies that best 185 achieve organisational fit 5.6.2 Building theoretical frameworks to guide 189 policy formulation for the Fiji Public Service 5.6.3 Clarity in HR policies impact HR selection 190 decisions for the Fiji Public Service 5.7 Conclusions 194

P a g e | 6

Chap 6 Characteristics of Instrumental and 196 Affective ties that build and maintain social capital influence on human resource selection decision within the Fiji Public Service 6.1 Introduction 198 6.2 Social capital network approach and Human 200 Resource Selection in the Fiji Public Service 6.3 Archival data and publicly available 200 documents 6.4 Social Capital and the Fiji Public Service 202 6.4.1 Representations in the distinctive social 202 systems of the dominant social groups 6.4.2 Representations and influence in government 205 and cabinet 6.5 Networking Ties 211 6.5.1 Instrumental ties and Human Resource 211 Selection decisions in the Fiji Public Service 6.5.2 School and religious influences in the Fiji 212 Public Service 6.5.3 Political influences 227 6.6 Affective ties and Human Resource Selection 232 decisions in the Fiji Public Service 6.7 Demographics of the Fiji Public Service and 239 Social Capital 6.7.1 Representations of dominant ethnic groups 239 within the Fiji Public Service 6.7.2 Age distribution analysis within the Fiji 241 Public Service 6.8 Conclusion 245

Chap 7 Fiji Public Service Employee’s Perception 246 of Justice in Human Resources Selection Decisions 7.1 Introduction 248 7.2 Fiji Government Ministries under survey 251 7.3 Response Rate 251 7.4 Analytical method 252 7.4.1 Data screening 253 7.4.2 KMO and Bartlett’s test 255 7.5 Justice and employee discretions in the Fiji 257 Public Service 7.5.1 Descriptive Analysis – Fairness of current 259 HR selection policies in your organization 7.5.2 Factor Analysis – justice and fairness of HR 266 selection policies 7.6 Conclusion 272

Chap 8 Discussions and Policy Implications 273 8.1 Introduction 275 8.2 Sub- research question one: What are the 279 HRM challenges facing the Fiji Public Service and what are the implications on HR selection decisions? 8.3 Sub - research question two: What are the 284 critical HR selection strategy challenges facing the Fiji Public Service in view of HR selection decisions? 8.4 Sub- research question three: Do the HR 287 selection policies of the Fiji Public Service P a g e | 7

contribute to the concepts of fit and flexibility to develop a highly qualified HR pool for the Fiji Public Service? 8.5 Sub research question four (a): Could the key 289 relations of the dominant social groups in Fiji that build and maintain social capital influence HR selection decisions within the Fiji Public Service ? 8.6 Sub research question four (b): What are the 291 key characteristics of instrumental and affective ties within the Fiji Public Service that have the potential to influence HR selection decisions? 8.7 Sub research question four (c): Do the 294 demographics of employees in the Fiji Public Service ministries examined correlate with a level of sustained social capital and social networking? 8.8 Sub research question five (a): Do employees 295 perceive justice and/or injustice with HR selection decisions of the Fiji Public Service?

8.9 Conclusion 297

Chap 9 Conclusion 301 9.1 Introduction 303 9.2 Conclusion about research questions 304 9.3 Conclusion about research problem 306 9.4 Implications for theory 307 9.5 Implications for policy and practice 308 9.6 Limitations 314 9.7 Implications for further researcher 315 9.8 Conclusion 316

P a g e | 8

LIST OF TABLES

Table Chapter Page Table Chapter Table

2.1 2 70 7.1 7 252 3.1 3 83 7.2 7 254 3.2 3 86 7.3 7 256 3.3 3 104 7.4 7 259 4.1 4 136 7.5 7 262 5.1 5 149 7.6 7 263 5.2 5 151 7.7 7 267 5.3 5 161 7.8 7 268 5.4 5 167 7.9 7 270 5.5 5 187 7.10 7 271 5.6 5 190 8.1 8 283 5.7 5 193 6.1 6 201 6.2 6 203 6.3 6 203 6.4 6 204 6.5 6 206 6.6 6 207 6.7 6 207 6.8 6 208 6.9 6 210 6.10 6 215 6.11 6 215 6.12 6 217 6.13 6 218 6.14 6 219 6.15 6 219 6.16 6 219 6.17 6 220 6.18 6 224 6.19 6 229 6.20 6 233 6.21 6 235 6.22 6 237 6.23 6 239 6.24 6 241 6.25 6 242 6.26 6 244

P a g e | 9

LIST OF FIQURES

Figures Chapter Page Figures Chapter Table

2.1 2 79 7.1 7 265 3.1 3 90 7.2 7 269 3.2 3 106 8.1 8 277 3.3 3 108 5.1 4 153 5.2 5 155 5.3 5 165 5.4 5 169 5.5 5 174 5.6 5 178 5.7 5 185 5.8 5 188 6.1 6 216 6.2 6 217 6.3 6 218 6.4 6 222 6.5 6 227 6.6 6 231

P a g e | 10

ABBREVIATIONS

Abbreviations Definitions Abbreviations Definitions

ACR Annual confidential report QVS Queen Victoria School ACS Adi Cakobau School RBV Resource based view ADC Aide de Camp RFMF Republic of Fiji Military Force AFWPC Australian Fair Work Pay RKS Ratu Kadavulevu School Commission SCI Social Capital Identity CA Competitive Advantage SDA Seventh Day Adventist CEO Chief Executive Officer SDL Soqosoqo Duavata ni Lewenivanua CSR Colonial Sugar Refining SFCRDSED Strategic Framework for Change and Roadmap for Democracy and Socio- Economic Development CTD Centre for training and development SHRM Strategic Human Resources Management EDP Employment department pass SPI South Pacific Island EEO Equal employment opportunities UK United Kingdom EFA Exploratory factor analysis US United States EIM Eritrean Institute of Management USP University of the South Pacific EU European Union FLP Fiji Labour Party FNPF Fiji National Provident Fund FPS Fiji Public Service FPSC Fiji Public Service Commission FPSIM Fiji Public Service Institute of Management FWA Fair Work Australia GCC Great Council of Chiefs GTC Government Training Center GWE Government wage earners HMOCS Her Majesty Overseas Cadet Service HR Human Resource HRMIS Human Resource Management Information System ICR Internal consistency reliability IHRM International human resources management IMF International Monetary Fund KSA Knowledge, Skills and Abilities LMS Lellean Memorial School MDG Millennium development goals MPSCR Maltese Public Service Reform Commission Report MQR Minimum qualification requirement MV Measured variables NEP New Economic Policy NFP National Federation Party NLFC Native Lands & Fisheries Commission PACO Performance Assessment for Contract Officers PCA Principle component analysis PRC Personnel replacement chart

P a g e | 11

APPENDIX

Appendix Page Title

9.1 319 Daily Mean Wages of Wage Earners by Industry Group in Fijian Dollars 9.2 320 Key Designations held and influenced by the Fiji Military Forces, 2011 9.3 322 Fiji Public Service Commission – Annual Confidential Report 9.4 324 Fiji Public Service Commission – Annual Performance Assessment 9.5 333 HR Selection: Appointment and Promotion - The Fiji Public Service - Process & Merit Criteria 9.6 343 Pilot Study – Fiji Public Service Survey 9.7 354 The First Public Service in Fiji – Civil List 1 - 1874

P a g e | 12

STATEMENT OF ORIGINAL AUTHORSHIP

I hereby declare that this submission is my own work and to the best of my knowledge it contains no material previously published or written by another person, or substantial proportions of material which have been accepted for the award for any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in this thesis. Any contribution made to research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the projects’ design and conception or in style, presentation and linguistic expression is acknowledged.

Signed ……………………………………………......

Date ……………………………………………......

P a g e | 13

ACKNOWLEDGEMENTS

Above all, I am deeply grateful to my supervisor, Dr Denise Faifua, for her advice, encouragement, and comradeship over the last three (4) years. I am also thankful to Professor

Satish Chand who gave very generously of his time, advice and overwhelming support during the scholarship tenure of the thesis.

Many people have contributed to the development of my ideas and to the smooth production of this thesis and of significance are Professor Michael Hess and Professor

Michael O’Donnell. I am grateful for their interest, and for the time they have spent reading and commenting on aspects of my work.

My family (Ilisapeci and Joe) in Australia and Fiji (Jiutasa) have endured a great deal during the growth of this thesis, and their support and love have been unwavering. Words can never convey my immense gratitude, and the great debt I owe them. However, during the tenure of thesis, the arrival of my grandchildren Vetaia, Mairini and Maraia have been uplifting.

To my mentor, guardian, father, uncle and advocate – Jiutasa Rasaca Vosikata whose sweat, toil and hardship has finally paid dividends. If only you were here (RIP).

Lastly, I would like to acknowledge my maker without whom I would not have completed the thesis

Philippians 4:13 I can do all this through him who gives me strength.

P a g e | 14

CHAPTER 1

INTRODUCTION

P a g e | 15

Table of Contents

1.1 Background to the research

1.2 Aims and objectives

1.3 Research problem

1.4 Justification for the research

1.5 Methodology

1.6 Outline of the thesis

1.7 Definitions

1.8 Delimitations of scope of the study

1.9 Conclusion

P a g e | 16

1.1 Background to the research

The purpose of this study is to analyse the influence of social capital through network ties on the human resource management (HRM) selection system in the Fiji Public Service (FPS). As the researcher of this thesis, I have worked as a civil servant in Fiji and come across situations where the human resources (HR) selection decision appeared to be influenced by the societal and political environment in Fiji. This subjective point of view was the motivation for undertaking an objective study of the influence of social capital and social networks on HR selection systems. This is particularly relevant in a country such as Fiji, with a history of political instability, and where since 2006 a non-democratically elected government has been in power. As a researcher, a citizen of Fiji and a previous public servant of the FPS, there were assumptions about unstable HR selection decisions that I wished to test. My personal observations were that social capital and social networks influenced the selection system, with different groups achieving different outcomes depending on ethnicity, gender, provincialism, settlements, politics, school, family and friends. Bourdieu confirmed that “social capital has two components: it is first, a resource that is connected with group membership and social networks. The volume of social capital possessed by a given agent depends on the size of the network of connections that he can effectively mobilize” (1986:

249). He went on further to state that “it is the quality produced by the totality of the relationships between actors, rather than merely a common quality of the group. This finding is also significant in justifying the correlations between the variables (instrumental ties, affective ties, heterogeneity variables, HR selection and decisions), under examination in this thesis.

P a g e | 17

This experience drove the motivation of this thesis and study and also formulates the assumptions for devising a scientific research using grounded theories to ascertain that unpredictable HR selection decisions may either corrupt the merit, quality and virtue of a public service institution or add value, hence influencing the quality of public service rendered, and tarnishing the image of the FPS. Further, the researcher also noticed the aspirations of different groups and institutions in Fiji’s social, economic and political spheres as they aspire to gain prominence in securing jobs and status within the FPS. In the process, networks are either directly or indirectly formed as per the following social structure streams: ethnicity, gender, provincialism, settlements, politics, school, family and friends.

Concomitantly, interpersonal influence and interpersonal selection are vital elements of the networking theory and are considered crucial in this thesis because important shared pasts in social networks can result in newly selected employees bringing established network ties to the organisation. Once established, these ties exert themselves and, similar to cell populations, go through a particular type of exponential growth, and snowball over time.

With this background and motivation, the writer began to search for relevant literature using credible and reputable sources to substantiate the research questions for the thesis, which took into account the research gaps. These were then conceptualised, and subsequently used to build a robust and rigorous theoretical framework that acted as a guide to the research.

The parent discipline is HR and Strategic Human Resource Management (SHRM), whist the core discipline of the study is HR selection. Throughout the thesis even though HR selection has been discussed as a standalone HR function with internal and external links,

Social Capital was prominently highlighted in view of the contextual basis it provides for the

Fiji Public Services. The study analysed the impact of social capital through network streams generated by Fiji’s social structure on the FPS HR selection system, and their impact on HR P a g e | 18

selection decisions. Specifically, it critically examined the hiring, selection and HR selection decisions of HRM in the FPS using a mixed method approach; a structured survey, followed by an analytical narrative where archival data were meticulously gathered and examined. As

Page and Meyer (2000) explained, the term ‘archival’ refers to any historical records and documentation held by companies, libraries, specialist institutions and so on, that may or may not be within the public domain, hence the use of this term in this thesis resembles this definition.

This study is important because it examines a state institution’s HR selection system in a country in the Pacific region with a non-democratically elected government and a history of ethnically fractured and political instability. Further, the findings in this thesis are crucial for developing nations of the world, because it examines the impact of social capital on a government’s HR selection system, particularly on HR selection decisions.

1.2 Aims and objectives

The major aim of this study is to analyse the impact of social capital in the FPS HR selection system and their influence on HR selection decisions.

In trying to set specific targets that align with the measurement framework and the key research question, the prime objective of this study was to conceptualise the theoretical gaps within the current literature into key substantive outcomes which have been empirically tested using the theoretical framework of the study. Specifically, the substantive outcomes eventuated after the confirmed gaps of the current literature were substantiated. In this regard, the study used literature on HRM, strategic human resources management (SHRM), social capital and justice perceptions and their examination of public service HR selection systems P a g e | 19

and decisions. The shortcomings were then critically capitalised and empirically examined to portray the context and fibre of this thesis.

The first key objective was to conceptualise the theoretical gap, and present the outcome based on substantive contributions from a general overview of the historical, political and social context of HR selection decisions in Fiji. In the process, it has provided literature, exemplified in Chapter 4 of the thesis, as evidence to help the reader understand underlying issues behind the main research question of the FPS study.

An additional objective was to conceptualise the theoretical gap and emphasise substantive contributions from the HRM, HR selection and social capital context of HR selection decisions in Fiji. The strategic fit of HR selection policies was also assessed. For this objective, the relevant literature and its shortcomings in view of the main research question of the FPS study and evidence are presented Chapter 2 to help the reader understand their association to the central research question.

Another crucial objective of the study was to conceptualise the theoretical gap and emphasise substantive contributions to meticulously examine the impact of social capital through network ties on HRM policy and practice in the FPS. Specifically, the theoretical framework and analysis set the parameters to critically analyse the HR selection processes in the FPS, examining pieces of legislation that have constitutional significance for the Fiji government and in particular the FPS with respect to HRM and HR selection decisions. This substantive outcome was deduced through an analysis based on a survey and the collection of case documents and archival data. It brought to light the association between social networks or relational ties, justice perceptions and worker performance within selected FPS organisations. They were selected because they have similar organisational structures and their operations depict outcomes that serve members of the public in Fiji in terms of the essentiality of services rendered. P a g e | 20

A key substantive outcome for the study was to highlight that current employees’ justice perception of HR selection decisions within the FPS has never been critically examined before. The literature review conducted and presented in Chapter 2 of the thesis revealed that perceptions of justice of HR selection decisions and procedures by existing employees has not received much attention, since much of the focus has been directed to applicant reactions (Gilliland, 1993, Gilliland & Hale, 2005). This gap in the literature justifies this key substantive outcome of the research which focused on perceptions of current employees in public service institutions.

Another objective of the study was to conceptualise the theoretical gap and emphasise substantive contributions through an analysis of the association between social networks or relational ties, justice perceptions and HR selection decisions across three (3) Fiji

Government ministries using the same survey instruments adopted for organisations in the embedded-depth case study. The survey of the three (3) Fiji Government ministries justifies this key substantive outcome of the research, which will focus on the justice perceptions of current employees on HR selection within the FPS. Further, this substantive outcome is also justified because these selected institutions provide vital and essential services to the Fiji public through Works, Transport and Public Utilities, Lands and Mineral Resources and

Agricultural functions.

Lastly, a key objective of the study was to conceptualise the theoretical gap and emphasise substantive contributions through an analysis of the numerous pitfalls and measurement defects that need to be considered in building an effective HR selection measurement system for the FPS. Pfeffer (1997) raised an important issue that potentially dwarfs the other concerns, and this is the politics of the measurement process and its likely effect on the HR function. The FPS currently uses the Performance Assessment for Contract

Officers (PACO) for workers who are either on short-term or long-term contracts and the P a g e | 21

Annual Confidential Report (ACR) performance measurement system for workers on full- time and part-time appointments. Hence, the study assessed the work performance and appraisal system given the magnitude of social capital impact on the FPS, and this has been presented in Chapter 4 of the thesis. Access to exit interviews and management records was obtained for this substantive outcome, with prior approval authorised by the Fiji Government.

1.3 Research problem

The central research question to address the research problem of the study is:

How does social capital through network ties influence human resource management

selection decisions and policies in the Fiji Public Service?

The answers to the above question can provide a theoretical confirmation of the influence of social capital on the FPS HR selection systems, particularly on HR selection decisions. At the outset, if the findings suggest that the HR selection system has a significant level of social capital and social networks influence, then the FPS needs to accord priority and urgently reform HR selection policies. On the other hand, it may also need to ascertain the benefits of social capital in public service operations, where the formal participation of members of the public is an important ingredient. However, if the findings of this survey suggest that the level of social capital and social networks influence on HR selection decisions is insignificant, then it may be ascertained that the FPS’s HRM focus is strategically aligned with the organisations’ business environment. With this rationale as a guide, a pilot study was conducted on a Fiji Government department (Department of

Immigration) to study characteristics of instrumental and affective ties and the reliability and validity of survey instruments (see Appendix E).

To conclude, firstly the historical, political and social context in Fiji have gradually shaped HR selection in the FPS, which is evident in Chapter 4. Further, the key strategic P a g e | 22

challenges of HRM and HR selection have emanated from social capital and social network influence through the key relationships of dominant social groups that build and maintain social capital, and this is evident in Chapter 5. In addition, the demographics of the FPS in view of the influence of social capital have provided the impetus to influence the justification of HR selection policy application, and this is shown in Chapter 6. Further, evidence from

FPS employee perceptions on HR selection have confirmed that social capital and social networks do influence HRM selection systems and selection decisions in the FPS through instrumental and affective ties. All in all, the net effect of instrumental and affective ties is shown to negatively influence the selection systems in the FPS, resulting in continued challenges to HR selection decisions in an environment of continued political instability. In this regard, I propose a new agenda for future research once Fiji’s political context has stabilised.

Four key areas of literature are examined is this thesis. First, the literature on HRM is examined with a view to theoretically examine the sub-research question: What are the HRM challenges facing the FPS and what are the implications for HR selection decisions?

Secondly, the literature on HR selection is critically examined to address the sub-research question: What are the critical challenges of HR selection strategies facing the FPS in view of

HR selection decisions? Thirdly, the literature on the strategic aspects of HRM is examined to theoretically address the sub-research question: Do the HR selection policies of the FPS contribute to the concepts of fit and flexibility to develop a cadre of highly qualified HR for the FPS? The essence of these sub-research questions is to critically examine whether the current FPS HRM policy framework is achieving the desired organisational fit between stated and achieved goals. Fourthly, the literature on social capital is examined to theoretically address the sub-research question: Could the key relations of the dominant social groups in

Fiji which build and maintain social capital influence HR selection decisions within the FPS? P a g e | 23

The literature is from the parent disciplines that inform this research on selection systems and

HR selection decisions in the FPS. The research problem’s immediate focus is selection systems and HR selection decisions. This third sub-area of literature also focuses on attribution and justice theory on HR selection, which is also a major focus in the literature review.

The following are the sub-research questions developed from the literature review to drive the data collection:

Research sub-question one:

What are the HRM challenges facing the FPS and what are the implications for HR selection decisions?

Research sub-question two:

What are the critical challenges of HR selection strategies facing the FPS in view of HR selection decisions?

Research sub-question three:

Do the HR selection policies of the FPS contribute to the concepts of fit and flexibility to ddevelop a cadre of highly qualified HR for the FPS?

Research sub-question four (a):

Could the key relations of the dominant social groups in Fiji which build and maintain social capital influence HR selection decisions within the FPS?

Research sub-question four (b):

What are the key characteristics of instrumental and affective ties within the FPS that have the potential to influence HR selection decisions?

P a g e | 24

Research sub-question four (c):

Do the demographics of employees in the FPS ministries examined correlate with a level of sustained social capital and social networking?

Research sub-question five (a):

Do employees perceive justice and/or injustice in the HR selection decisions of the FPS?

1.4 Justification for the research

This study is important because it examines a state institution’s HRM selection system in a developing country in the Pacific region with a non-democratically elected government. In selecting personnel to fill job vacancies, civil service managers in the FPS are continually challenged to adopt SHRM practices at the strategic and operational level but often encounter constraints in application. These civil service managers are challenged by complex historical, political and demographic forces which hinder the rational decision-making prescribed by

HRM principles. These complex forces are defined as the selection context, and provide a large part of the explanation for influencing HR selection decisions.

This research contributes a social capital perspective to the HRM discipline. From a social capital perspective, it offers an explanation of how instrumental and affective ties of ethnic groups in developing countries can negatively influence HRM selection systems. In essence, social structure generates social capital, and through social capital actors can gain direct access to economic resources (subsidised loans, investment tips, protected markets); they can increase their cultural capital through contacts with experts or individuals with institutional authority and power who embrace and worship cultural capital; Alternatively, they can affiliate with institutions that confer valued credentials (i.e. institutionalised cultural capital) (Portes, 1998). Coleman (1988) asserted Ben-Porath (1980) finding that in P a g e | 25

economics, the exchange systems are centered around the “F – connection”, where families, friends and firms constitute social organisations that affect economic exchange. He also confirmed that in the new institutional economics, there is a failure to recognise the importance of concrete personal relations and networks of relations in what he calls

“embeddedness” in generating trusts, in establishing expectations, and in creating and enforcing norms.

In adopting a social capital perspective to HR selection systems, it is possible to examine how employees use their social capital to gain direct access to economic resources, for instance by being appointed to a public service institution designation or being promoted to a higher position within the organisation. In that regards, the distortions of HR selection decisions could be evident on HR selection outcomes. Further, the actors can increase their social capital through contacts with experts or individuals within or external to FPS institutions and form strategic alliances. There are certain third-party institutions that Fiji public servants perceive as having the potential to impact the rationality of operations within the FPS, thereby making that workplace safe, comfortable and desirable for them. In using their social capital, public servants in Fiji may also affiliate with institutions that confer valued credentials in manufacturing social capital which might promote social mobility beyond economic means. In other words, these public service institutions provide other social benefits apart from the formalities of existing intrinsic and extrinsic values. Bourdieu (1980) produced the first systematic contemporary analysis of social capital and defined the concept as the aggregate of the actual or potential resources which are linked to possession of a durable network of more or less institutionalised relationships of mutual acquaintance or recognition. Coleman also affirmed that “the concept of social capital is a tool that aids the importation of the economists’ principle of rationale action for use in the analysis of social systems proper without discarding social organisation in the process” (1988:97). The early P a g e | 26

Indian settlers in Fiji for that matter utilised their social capital in the application of rationality towards business ethics which enhanced their economic success whilst maintaining their social organisations.

By examining existing employee perceptions, this research also contributes to existing research on the perceptions of justice and injustice of selection decisions, i.e. the attribution theory of employee perspectives on selection. It is different to past studies in this area where the focus has been on applicant reactions (see Gilliland, 1993, Gilliland & Hale, 2005). It draws on psychological perspectives of group biases, misconceptions and tacit understandings; i.e. the attributions or causal ascriptions people make of events and behaviours as the basis of judgements, emotional reactions and behaviour (Ployhart & Ryan,

1997, Harold & Ployhart, 2004). Attribution theory assumes any number of causes may be attributed to the perceptions of justice and injustice in selection decisions. Based on Weiner’s

1986) work it is possible to view attributions from the point of view of internalised traits or externalised situations. That is, whether a cause is perceived to be due to factors internal to a person; the extent to which the cause is seen as stable or unstable in the future; and the amount of control the individual or the person making the selection decision has over the outcome. Attribution theory is used to guide the study of existing employee perceptions of justice and injustice in HRM selection systems and HR selection decisions.

1.5 Methodology

The methodology adopted in this thesis is positivism, and is based on research questions that are tested both qualitatively and quantitatively. An embedded case study approach is adopted given the study is based on the FPS and three ministries within it. The ministries are the

Ministry of Agriculture, the Ministry of Works, and the Ministry of Lands and Mineral P a g e | 27

Resources. The case study approach enables research findings to be presented at the level of the organisation’s strategic overview of HRM, the level of operational HRM and the level of the employee. In other words, throughout the presentation of case findings, there are three units of analysis, the strategic, the operational and the existing employee. The case study approach evidence is a mix of qualitative data and quantitative data; i.e. qualitative data collection and analysis, e.g. historical documents, organisation documents and policy documents are collected and analysed to address the research questions, and the quantitative data collection and analysis is based on the survey responses of existing employees.

1.6 Outline of the thesis

Again, the problem addressed in this research is:

How does social capital through network ties influence human resource management

selection decisions and policies in the Fiji Public Service?

Chapter 2 presents the literature outlining the key theoretical bases including HRM and HR selection literature, highlighting its strategic development and applicational aspects in the FPS. Of note, this chapter demonstrates that research on SHRM within the FPS is important to identify and align policies between the internal and external environment, in view of unforseen environmental changes. Consequently, FPS managers need to observe the conditions of growth and competitive advantage (CA) at a comparative HRM level. Further,

Chapter 2 also emphasises the need for an alternative approach, a social capital and social network approach, that acknowledges the growing complexity of inter- and intra- organisational social networks and then goes on to introduce social network theory as a tool for rethinking how managers might adequately deal with complex social contexts that detract, mitigate or challenge the successful implementation of the principles and practice of HRM. It P a g e | 28

is well recognised that there are challenges in implementing Western ideas or systems, such as HRM, in countries where communalism and collectivist ideologies have a strong foothold.

This kind of ideological foothold is evident in the Fiji context and is a strategic challenge to the implementation of HRM. Further, Chapter 2 examines employee perceptions of justice and injustice of HRM systems and selection decisions, and also employee reactions to perceptions of injustice and justice in the HRM system and selection decisions. It also includes literature on employee reactions to perceptions of justice and injustice and reviews the likely responses of employees to perceived distractions in HR selection decisions.

Chapter 3 describes the research method used for this thesis. It is divided into three

(3) key parts. The first part presents the data collection method, which includes gathering historical, organisational and policy documents. The part also outlines a description of the development of the research instruments for the survey and the research framework for the embedded case study analysis, which includes the selection of and justification for the country of survey and specific public service institutions; the administration of the questionnaire; the response rate; and the demographic statistics of the respondents which detailed the sample and population in the research. The second part discusses the instruments available for the measurement of variables of interest in this study and the justification of the selection of the particular instruments used. The third part of the chapter explains the descriptive statistics, the correlation matrix of the variables and the case study analogy analysis.

Chapter 4 briefly describes the historical, political and social context surrounding HR selection decisions within the FPS. Specifically, the historical background of labour, the role of the chief system in the FPS, the political context of employment and the demographic structure of Fiji in have been examined. The purpose of this chapter is to describe the HR selection context. P a g e | 29

Chapter 5 examines HR selection approaches within the FPS, considering the concepts of fit and flexibility to develop a highly qualified HR pool for the FPS. In this chapter, key evidence is presented to help the reader understand the association between the theories and the central research question of the thesis. The chapter has been labelled Case Study One and presents the answers to the sub-research questions one, two and three, as related to key theoretical constructs of the survey.

Chapter 6 is the embedded case study, and is labelled Case Study Two. It presents instrumental and affective ties characteristics for FPS employees which build and maintain social capital influence on HR selection decisions within the FPS. The case study presents the answers to the sub-research questions four (a), (b) and (c).

Chapter 7 presents the general distributional properties of the data in the survey on existing employee perceptions of justice of the HR selection decisions within the FPS. The chapter is labelled Case Study Three and presents answers to sub-research questions five (a).

In Chapter 8, the findings from the three case studies are amalgamated and discussed in view of the central research question of the thesis. Specifically, the answers to the sub- research questions used the different units and levels of analysis in the embedded case study and the survey findings on employee perceptions on HR selection decisions.

Finally, Chapter 9 presents conclusions and major implications and ties up the thesis statement, the initial declaration of the major findings and the thesis proper. It also touches on future implications for HR selection decisions in the FPS, and major theoretical and practical potential.

P a g e | 30

1.7 Definitions

Definitions adopted by researchers are not uniform, so in this thesis, key and controversial terms are defined to establish positions taken so as to reinforce uniformity and consistency. In this regard, it is crucial that key definitions are presented (Perry, 2012).

Human resource management (HRM)

In his study, Stone established that “HRM involves the productive use of people in achieving the organisation’s strategic objectives and the satisfaction of individual employee needs” (2011:4).

Social capital:

Coleman confirmed that “social capital consists of the relations between persons of the sort that exist in the community and particularly to closure of these relations” (1988: 8).

Bourdieu established that “social capital is the aggregate of the actual or potential resources which are liked to possession of a durable network of more or less institutionalised relationships of mutual acquaintance and recognition or in other words, to membership in a group which provides each of its members with the backing of the collectivity owned capital, a ‘credential’ which entitles them to credit, in the various sense of the word” (1986:10).

According to Putman (1993b), social capital refers to “features of social organization, such as networks, norms and trust that facilitate action and cooperation for mutual benefit

Adler and Kwon (1999) have defined ‘social capital’ as a resource which accrues to actors through network ties. Further, Portes argued that:

it is probable that social structure generates social capital, and through social capital,

actors can gain direct access to economic resources (subsidized loans, investment tips,

protected markets); they can increase their cultural capital through contacts with experts

or individuals of refinement (i.e. embodied cultural capital); or, alternatively, they can P a g e | 31

affiliate with institutions that confer valued credentials (i.e. institutionalized cultural

capital). (1998: p4).

In addition, Harrington (2001) developed a model of the causal impact of social capital on organisational performance, with particular attention to specifying the contingencies that transform some network ties into social capital or social liability.

Social network:

In their research, Acquaah & Eshun found that “only managerial social networking relationships with top managers of other firms, government bureaucratic officials and community leaders enhance organisational performance” (2010, p668). They further argued that a social system that is highly collectivistic, and embedded in cultures and traditions that thrive on communal bonds, interpersonal relationships and strong allegiance to community leadership, provides the sources of networking that may influence and distort HR selection processes. Further, Argyle et al. (1981) found that ‘social rules’ are defined as shared beliefs about the behaviours that should or should not be performed in particular situations.

Selection context:

Mellahi & Frynas (2003) confirmed an inherent difficulty in transferring HRM practices between different cultures and the inability of Algerian managers to understand the potential problems that might be created by applying Western HRM practices in Algeria.

Bauer et al. (2001) found that studies on the relationship between the perceived fairness of selection systems and applicant reactions have increased over the years. The inherent difficulty in transferring HRM practices has provided a key HR selection context for this thesis.

P a g e | 32

Selection systems:

De Cieri and colleagues (2011) defined ‘selection’ as the process by which companies decide who will or will not be allowed into their organisations. De Ceiri et al. (2007) found that evaluation of HR functions, such as selection, are increasing the importance of establishing the link to the organisation’s bottom line. Hence, this definition is crucial for this thesis.

Selection decisions:

Amin (2004) argued that government policies can affect labour market outcomes for different ethnic groups. He asserted that, in 1971, the Malaysian Government implemented the New

Economic Policy (NEP), which gave differential economic benefits to Malays compared to non-Malays. Compared to other ethnic groups, the Malays are, on average, less educated, have lower incomes and have a higher fertility rate. Hence, this context has provided the texture for selection decisions in this thesis. Further, Davidson & Friedman (1998) stated that demographic characteristics, such as ethnicity, age and competency levels, may be likely to influence an individual’s expectations of justice because they capture some issues of a group’s variability in experience or beliefs systems of past fairness. In this regard, it has also provided an HR selection decision context for this thesis.

Instrumental ties:

Harrington found that “instrumental based societies and networking exemplifies rationalising decision making so that a particular decision may produce the greatest good for the greatest number of individuals” (2001:p 92). She further stated that instrumental ties have low intrinsic relationship value outside task setting.

Affective ties:

In her study, Harrington (2008) also confirmed that an affective based society and networking are controlled and managed by social norms and their respective obligations, including the P a g e | 33

traditional duty to reciprocate in cash or kind. She also found that affective ties have high intrinsic relationship value.

Justice:

The literature review revealed that existing employees’ justice perceptions of HR selection decisions and procedures has not received much attention, since much of the focus has been directed to applicant reactions (Gilliland, 1993, Gilliland & Hale, 2005). Greenberg (1990) defined ‘organisational justice’ as the study of fairness within organisational settings and derived this from work in social psychology aimed at understanding fairness issues in social interactions. Ployhart & Ryan (1997) stated that, organizational justice refers to how people perceive and react to fairness in work related contexts. Recent research suggests that justice perceptions are most aptly conceptualised along four dimensions: distributive, procedural, interpersonal and informational

Employee attributions:

Ployhart & Ryan (1997) argued that attribution theory makes clear predictions about how an individual will respond to a selection decision, and also provides information about why the reaction occurs (i.e., nature of causal ascriptions).

Fiji Public Service

The FPS is the umbrella institution representing all government departments and ministries in

Fiji.

Fiji Public Service Commission

The Fiji Public Service Commission (FPSC) under the States Services Decree No. 6 of 2009 is the Central Personnel Authority for the FPS and is responsible for the effective management and development of the HR, organisational and systems management and for ensuring the continuous improvement of the entire service. (www.psc.gov.fj) P a g e | 34

1.8 Delimitations of scope

Three key FPS institutions were selected for the case study because they are crucial in public service delivery in essential areas, e.g., lands and survey, works, transport, energy, rural electrification, meteorology, agriculture and the primary industries sector. As exemplified, the three government ministries were chosen because of the extensive nature of services rendered through their respective ministries. The stipulated ministries include: Ministry of

Lands and Mineral Resources and the Ministry of Works, Transport and Public Utilities and the Ministry of Agriculture and Primary Industries. However, there are certain limitations, as these ministries are guided by government regulations.

The timeframe for my study is four years, beginning. April 2010, which signifies the implicit boundaries for the tenure of the scholarship. The scholarship provider is the Fiji

Government under the Itaukei Scholarship Board. Even though the scholarship provider has been operating through an undemocratic system of government, this awareness has been perceived and controlled by the student during the scholarship tenure. In other words, the scholarship tenure and its contractual obligations have always been perceived as unstable by the recipient.

1.9 Conclusion

This chapter outlines the structure of the thesis, acknowledging Perry’s (1998) field approach to presenting a thesis, which is a structure common to thesis presentation in the areas of HRM and management. It introduced the parent disciplines or literature of HRM, HR selection, social capital and the justice and attribution theory of selection perceptions as a sub-area or literature of HRM. The sub-research questions and the research hypotheses were presented to portray the significance of those disciplines and highlight the substantive contributions of the P a g e | 35

thesis. All in all, the research was justified; the definitions presented; the methodology outlined; the chapters outlined; and an explanation of the limitations of scope was given.

Chapter 2 will now outline the literature review of the thesis, highlighting the critical research gaps to justify the substantive contribution of the thesis.

P a g e | 36

CHAPTER 2

LITERATURE REVIEW

P a g e | 37

Table of Contents

2.1 Introduction

2.2 Human resource management and the Fiji Public Service

2.3 Human resource selection and the Fiji Public Service

2.4 The strategic approach of human resources management and the Fiji Public Service

2.5 Social capital and the Fiji Public Service

2.6 Networking ties

2.7 Justice and employee discretion in the Fiji Public Service

2.8 Conclusion

P a g e | 38

2.1 Introduction

Central Research Question: How does social capital through network ties influence

human resource management selection decisions and policies in the Fiji Public

Service?

The above research question was formulated in view of the need to assess those social capital and social network variables through network ties which have been influencing HR selection decisions in the FPS. In this regard, the literature review for this thesis explores network ties through social networks and social rules theory. It also examines FPS employees’ justice perceptions and applicant reactions, since the literature review has revealed that justice perceptions of HR selection decisions by existing employees has not received adequate attention because much of the focus has been directed to intending employees’ reactions

(Gilliland, 1993, Gilliland & Hale, 2005). Key variables have emerged from social network theory; i.e. instrumental ties and affective ties (Harrington, 2001) are assumed to have affected the perceptions of justice in the hiring processes in the state institutions under study.

Existing employees’ differing perceptions of justice on HR selection decisions may have a positive or negative influence on the utilisation of group heterogeneity variables including age, ethnicity and competency (KSA).

For this thesis, cited works have been incorporated in the relevant discussions and arguments through an analysis of the theoretical framework, followed by the analytical or empirical framework. It will then identify the limitations of the findings and highlight the relevance of the article to the study of the hiring processes in the FPS.

This chapter assimilates critical components of the relevant literature, taking into consideration the boundaries and constraints governing the research problem, so as to develop a scientific basis for a theoretical framework to carefully guide the application of the P a g e | 39

research instruments. A theoretical framework is a conceptual model of how one theorises or makes logical sense of the relationships among the several factors that have been identified as important to the problem (Sekaran, 2003).

To date, the research has achieved a number of substantive outcomes. These included substantive contributions from the social capital context of HR selection decisions within the

FPS. The critical outcome from this part of the study involves a detailed analysis of the demographic characteristics of the employees in the FPS on the basis of ethnicity or race, political affiliation, religious affiliation, education background, and community and provincial affiliations. The demographic characteristics present the social capital context within the FPS which may influence HR selection decisions.

The second substantive outcome is an analysis of the effectiveness of HR selection decisions in the FPS; i.e. the validity and reliability of the selection tools that are part of HR selection decisions, which are used as predictors for candidate work performance. These analyses will examine the relationship between the HR selection decisions and success on the job. This fits with the argument by De Ceiri et al. (2007) that evaluation of HR functions, such as selection, are increasingly important in establishing the link to the organisation’s bottom line. This also fits with the argument that there are potential biases impacting HRM systems. In his research, Pfeffer (1997) discussed the numerous pitfalls and measurement biases that need to be considered in building effective measurement systems in HRM. He raised an important issue that potentially dwarfs the other concerns: that the politics surrounding measurement processes is likely to have a biased effect on HR functions. The politics may have emerged from social capital influence through instrumental and network ties. Pfeffer (1997) elaborated that departmental leaders have well-developed skills at massaging numbers to get them to support their units. This is an area of crucial concern for the FPS where the researcher has analysed HR selection decisions of the institutions under P a g e | 40

study, particularly how managers have been effectively utilising HR selection decision policies.

The third substantive outcome of the study concerns the reactions of FPS employees to current HR selection decisions. As exemplified earlier, they may either perceive managerial HR selection decisions positively or negatively. The focus on HR selection decisions is justified given the provision of essential services by the FPS. The three FPS organisations were selected because they have similar organisational structures and they are concerned with essential services. These organizations are the Fiji Government Ministries:

The Ministry of Lands and Mineral Resources and the Ministry of Works, Transport and

Public Utilities and the Ministry of Agriculture. Absenteeism and employee exit and turnover data were used to show the effectiveness of HR selection decisions and policies within the

FPS.

A fourth possible substantive outcome will be the perception of justice with regard to selection and promotion. Questions about selection and promotion were included in the survey.

The chapter has been deliberately designed to emphasise the substantive contributions from the influence of social capital in Fiji through instrumental and network ties on HR selection decisions within the FPS. In the process, relevant literature on the underlying theories and evidence are presented to help the reader understand their association in answering the main research question of the FPS study. In his study Bourdieu (1986) found that these relationships and networking may also be “socially instituted and guaranteed by the application of a common name (the name of a family, a class, or a tribe or of a school, a party etc) and by a whole set of instituting acts designed simultaneously to form and inform those who undergo them; in this case, they are more or less really enacted and so maintained and in exchanges. In the Fijian social systems, these relationships and networking have been socially P a g e | 41

instituted and traditionally entrenched within the social structures to depict the collective obligation of the tribe in terms of their traditional duties to the patriarch.

Further, HRM and SHRM are examined in regard to the developmental status of their application within the FPS, ascertaining whether the FPS has fully adopted SHRM, or is still engrossed with HRM. Furthermore, Fiji’s HRM system may be classified as undeveloped because it depicts a simple and dynamic environment comprised predominantly of simple structures. In his study, Begin (2006) asserted that most developing countries would fall in this category due to their degree of industrialisation and the unpredictable operation of their economic and political systems. He also found that in developing countries, “both the societal and organisational dimensions are limited in scope producing a poor standard of living in view of the substandard economic performance of the nation” (2006:pp4 - 8).

Acknowledging the HRM literature as the basis of the study, managers in the FPS need to understand the challenges faced in managing a culturally diverse cohort of employees, given the influence of the social context in which FPS organisations operate.

Further, FPS managers need to gain an awareness of the growing importance of social justice and the contributions SHRM can make to sustainable economic initiatives in the developing nation of Fiji. In this regard, considering the influence of social capital, SHRM is important for the continuing implementation and success of HRM within the FPS, linking key strategies from within to business outcomes.

HRM in developing countries like Fiji is often developed over time as a result of the interaction between contemporary management ideologies as against a labour force that closely aligns itself with societal and cultural norms. Specifically, the encounter triggers a chain reaction between the clash of what are perceived as a well-defined and homogeneous

HR selection policy framework versus a labour force infused with a collectivist culture, tradition and networking. This chapter argues that the mindset of FPS workers needs to be P a g e | 42

continually aligned with a conducive HR selection policy framework that controls the influence of social capital.

The chapter also emphasises the need for an alternative approach, one which acknowledges the growing complexity of inter- and intra-organisational social networks and then goes on to introduce social capital theory as a tool for rethinking how managers might adequately deal with complex social contexts that detract, mitigate or challenge the successful implementation of the principles and practice of SHRM (Jackson et al., 1989).

In this chapter, Harrington’s (2001) model of social capital has been adapted and modified as the theoretical base for the FPS study, acknowledging the social selection context of SHRM in developing countries. She examined the mechanisms by which certain kinds of social relationships at the individual level facilitate or impede the attainment of objectives at the firm level. She found that recruitment and selection based on weak instrumental ties increase an organisations’ compositional diversity and commitment to task performance.

Harrington also confirmed that groups based on affective ties can become “over embedded”, limiting the pool of information available for decision-making, and inhibiting debate in favour of social cohesion (2001). These findings highlight Harrington’s contingency thesis of network ties and its relevancy to the study on the FPS. Further, Coleman (1988) compounded this finding when he confirmed that social capital comprises certain attributes of the social structure. His findings confirmed that a given merchant community is ordinarily very close, both in the frequency of interaction and in ethnic and family ties. In that regards, he exemplified the wholesale diamond market in New York City which is Jewish, with a high degree of intermarriage, living in the same community in Brooklyn, and going to the same synagogues. In essence, it is a closed community. In the Fiji context, the merchant communities also exemplifies the same social structures, where the Muslim communities go to their respective mosques in different towns in Fiji; those merchants with Hinduism P a g e | 43

background go their temples, whilst those with Chinese background converge on those attributes of their social structure that specifically depict their social capital. As similar to the

Jewish community in Brooklyn, these Fiji merchant communities are also very close, both in the frequency of interaction and in ethnic and family ties. Coleman confirmed that these

“close ties; through family, community and religious affiliation provide the insurance that is necessary to facilitate the transactions in the market” (1988:99). On the outset, the normative structure within a Fijian village ensures that unattended children will be “looked after” by adults within the vicinity, while no such normative structures exists in most urban or urban metropolitan areas. In that regards, we can deduce that families in a Fijian village have available social capital that does not exist in the urban areas in Fiji.

At the level of the employees, the FPS study’s theoretical framework highlights how the social context can influence HR selection decisions which lead to employee perceptions of justice and injustice in HR selection policy and practices. Thus, the model highlights how social capital is a determining factor in FPS managers’ HR selection decisions. At the organisation level, the model highlights the tensions created by FPS employees’ perceptions of justice towards the application of HR selection policies. At the national level, the model exemplifies the influence of social capital in policy formulation and application. In addition, a critical theme of the literature is that social rules emerge from social values built on instrumental ties and affective ties. On this basis, social rules, social values, instrumental and affective ties are presumed to affect employee perceptions of organisational justice, in particular and for the interests of this study, perceptions of the hiring process in the FPS.

The social context of the FPS is considered a key challenge to the successful implementation of SHRM. Stakeholders of the FPS need to understand the reality of the paradigm shift from the traditional colonialist system of governance to the implementation of the Western rationale of SHRM. In implementing SHRM, developing countries need to P a g e | 44

acknowledge and embrace globalisation, where ownership of commodities and management ideologies are being extending across the horizon.

One of the key challenges for public service organisations in developing countries is the application of HRM systems suggested by the developed nations. These developed nations are part of a tripartite group that are globally perceived to staff the donating agencies and institutions that have been tasked to assist the developing nations. However noble and laudable their reform objective is, the process is open to serious critical analysis from stakeholders, including those in the political arena and in academia. This is because of the inherent difficulty in transferring HRM practices between cultures and the inability of managers to understand the potential problems that might be created when applying Western

HRM practices, given the robustness of their existing internal policy and structural framework (Kim & Hong, 2005). In this context, it is understandable that public service policy and structural frameworks for developing countries are fragile and may not contain the advanced mechanisms of Western HRM practices. The argument that developing countries are vulnerable has also been advanced by Kim and Hong (2006), whose work has provided policy initiatives and advice for many developing countries where there is a need to initially establish a practicable and effective HRM framework. This would act as an infrastructural guide to achieve sustainable high performance in public service institutions. Further, an HRM system for developing nations should then be tailored and built on the infrastructure, based on a hands-on approach around impartiality, professionalism and responsiveness to environmental changes.

The study of perceptions of justice is important for this thesis (see Figure 2.1 at the end of the chapter) because FPS employee perceptions of HR selection decisions moderate the relationships between social networking and employee reactions to their perceptions. The outcome of such a correlation deduces either positive or negative aspects of network ties. P a g e | 45

Perceptions of justice have a strong contingent effect on the relationships between social network ties, social capital, HR practices and performance. Hence, the theoretical proposition deduces that the positive or negative impact of instrumental ties on HR selection decisions will be greater in organisations with perceptions of justice than in organisations with perceptions of injustice. The thesis specifically examined FPS employee perceptions of justice in HR selection decisions whereby employees’ discretion were induced after workers had assessed how just or unjust these decisions were. In the process, they individually assess how their outcomes have been aligned in proportion to their inputs, triggering an equity theorem that affects work behaviour. Specifically, employees normally seek to perceive equity between the inputs that they bring into a job versus the outcomes that they receive from it against the perceived input and outcome of others (Adams, 1963).

The literature gathered to provide the theoretical underpinning for the study has also highlighted significant gaps, which provide the substantive contribution of the chapter to theory and practice. In the process, sub-research questions have been designed to substantiate the contributions and propose a riposte to the central research question of the thesis.

2.2 Human resources management and the Fiji Public Service

Against a social capital backdrop that provides the contextual basis, the investigation of the parent discipline of HR and Strategic Management was eminent to pave the way for the critical examination of the core discipline of the study which is HR selection. In their study,

Ferris, Hall, Todd and Martocchio confirmed that “the field of human resources management

(HRM) has emerged from the conceptual, empirical, and practical intersection of several disciplines including psychology, sociology, economics, and management/organization sciences” (2004:p231). In that regards, they concurred that “theoretical development in the field of HRM has been more piecemeal, and has focused on more middle-range P a g e | 46

conceptualizations, driven by theory in these core disciplines” (2004:p231). Further, Boxall and Purcell (2011) coined a profound definition where they ascertain that HRM are those activities associated with the management of work and people in organizations. In a later study, Stone confirmed that HRM “focuses on managing people within the employer– employee relationship, and involves the productive use of people in achieving the: organisation’s strategic business objectives, and satisfaction of individual employee needs”

(2013:p34). However, this thesis has also considered Boudreau & Ramstad (2007)’s findings which observed that HRM are shifting their focus from providing transactional services to providing top management with decisions that “inform and support”. This critical observation poses a challenge for contemporary managers, particularly in developing countries like Fiji in managing those sub-functions of HRM that deals with transactional practices like HR selection, recruitment, HR planning and so forth.

The impact of globalisation influenced HRM in the FPS, particularly by modifying the methodology of operations and service delivery. In his study, Holden (2002) stated that this issue, together with the serious economic downturn of the 1990s particularly amongst developing economies, caused many to abandon regulatory state measures for more liberal economic policies. This issue will, of course, have considerable implications for HRM, in that relevant work practices, performance management policies and strategic HRM will complement the liberal ideologies. In the present day, HRM in the FPS can be classified as a

‘cross-cultural HRM’, where a multicultural concept has been spinning diverse networking intricacies so as to ascertain the social, legal, political, religious and economic grip for dominant the social grouping, reinforcing their stronghold and status.

Further, it is important to conceptualise the ways in which HRM ideologies confront the diversity in Fiji, particularly how they are embraced and adapted. In developing his paper,

Poole (1982) established the effective harnessing of HR as the discipline base for his theory, P a g e | 47

which provides the theoretical framework for his study. He linked the theoretical constructs to the research design by highlighting key HR variables, grouping them and operationalising them to form two (2) key research questions. Firstly, what are some of the general forces that affect the personnel function in organisations and secondly, what are some ways in which

HRM may be different in ‘emergent’ countries from other developed nations? A substantial contribution of this thesis has been presenting HR selection in a Pacific Island country which has a unique social context.

The key HRM challenges facing the FPS today vary in accordance to its respective doctrines for survival and good governance, taking into account its political and economic environment. In acknowledging the ‘known’ HRM challenges in the FPS, this paper will rather focus on ‘unknowns’ and emerging challenges. In her study of the FPS, Vallance highlighted several ‘known’ challenges and these included: brain drain; poor education quality within the HR pool; and lack of manpower planning (1996:p93). While these ‘known’ challenges have intensified through the post-independence and post republic era in Fiji, this thesis will highlight and present other ‘unknown’ challenges to the FPS. These include: constraints in the application of readymade managerial concepts because of lack of resources; difficulty of transferring and utilisation of Western HRM practices; pressures on the public service from donor agencies and their peers; the need to adhere to regional change initiatives; and the need to change the mindset of public servants (Hewitt et al., 2002). These

‘unknowns’ have provided substantive contributions to cover gaps in the literature studied.

This chapter will now introduce significant findings and challenges for some developing countries which have the same governance framework as most South Pacific

Island (SPI) countries including Fiji. Firstly, in their research on the Algerian civil service,

Mellahi & Frynas found that under the immense pressure for survival, “managers in the civil service have desperately searched for new ready-to-implement management concepts as their P a g e | 48

previous practices have become obsolete and inadequate in the new business environment”

(2003: p 75). They found that these managers hoped that such adoption would enhance their effectiveness, and give them a fighting chance in the fiercely competitive global markets.

Further, Mellahi & Frynas found that, “mimicking western HRM practices in an alien social context” are often ineffective and the objective for such integration into public services are often vigorously scrutinised by stakeholders (2003: p77). This is because of the inherent difficulty in transferring HRM practices between different cultures and the inability of

Algerian managers to understand the potential problems that might be created by applying western HRM practices in Algeria (Mellahi & Frynas (2003).

For Fiji, pressures for reform usually emanate from stakeholder claims that the public enterprises were operating at unacceptably low levels of efficiency and effectiveness. These stakeholders include internal private sector institutions and external financiers, including certain leading developed economies and their financiers. Some of the reasons identified for this included a lack of accountability, over-staffing, corruption, cronyism, etc. (Department of

Public Enterprises, 1998). Also, other external agents including donor agencies, particularly the International Monetary Fund (IMF), World Bank and the Asian Development Bank, have been exerting pressure for reforms and have played a key role in public service reforms in

Fiji. This pressure-cooker situation capitulate Fiji civil service managers to implement reforms dictated by ‘change agents’, and their reform specifics included the application of

Western HRM practices which is not conducive to embrace the merits of public service reform. They had no other choice. The SPI countries’ public service reform patterns all have the same unique framework, whereby the donor agencies dictate and indirectly control the reform processes (Vallance, 1996). In Fiji, for example, ever since the inception of the FPSC,

Western democracies including Australia and New Zealand were leading the ‘charge’ for efficiency, expediency and managerial autonomy through decentralisation of the public P a g e | 49

services. However, Fiji and other SPI countries neither have the resources nor the flexibility to transform their public services quickly or to keep pace with global trends in public management. The outcome has made SPI countries over reliant on the developed economies within Oceania, hence reinforcing their status as micro states.

FPS institutions need to effectively manage their HR if they are to get the maximum contribution from their employees. In their study on the Eritrea civil service, Tessema and

Soeters found that, just like most developing countries, “the Eritrea Government introduced civil service reforms from 1995 to 1997, which led to: (i) streamlining of about 34 per cent of the civil service, (ii) the establishment of the Eritrean Institute of Management (EIM) in

1995, (iii) the introduction of a new salary scale in 1997, and (iv) the launching of the

Eritrean human resources development (HRD) project” (2006: p 87). However, in spite of the implementation of civil service reforms, the critical challenge that Eritrea faced was the use of HR to enable civil service organisations to operate effectively (Tessema & Soeters, 2006).

Generally speaking, they identified that the environment in which Eritrean civil servants are employed does not seem to attract, motivate and retain competent civil servants, hence the availability and use of capable civil servants is of utmost importance. To this effect, enhancing government performance and national development requires moulding a technically conducive environment which would build a public service that is impartial, professional and responsive.

Public service reforms were often initiated through a tripartite arrangement where political mileage was often prioritised against the rationality of incorporating desired HRM practices to improve performance. In their study on the Maltese public service, Cassar and

Bezzina highlighted that “the structural and operational changes in the Maltese public service were spearheaded through the European Union (EU), due to Malta’s accession to the EU”

(2005: p213). Further, they also stated that it is important to note that under the supervision of P a g e | 50

the EU, the Maltese Public Service Reform Commission Report (MPSCR) mandate was primarily to execute structural and policy changes. The attempt to change old mindsets

(Hewitt et al., 2002) was relatively ignored, except for retraining recommendations. This is

“considering a work force environment which is characterised by: relatively small-scale market conditions, community oriented business giving rise to a clan mode of organising work” as found by Ouchi (1980), and “dominant cultural norm of consensus and tradition ingrained in traditional values and beliefs” (Baldacchino, 1999). Further, Cassar & Bezzina found that Malta although in a transitional state, “still operates an institutionalised employment relationship based on collective agreements in which many conditions of employment are regulated by powerful societal systems and preset by national legislations and networking” (2005:p 207). The critical lesson for public service organisations in developing countries from the Maltese civil service study is that ‘people must change before institutions can’. This is again a crucial finding for the Fiji public servants, where the civil servant mindset has to be changed to fully embrace newer ideologies and good governance.

The influence of networking ties on the FPS HR selection decisions poses a challenge for HRM in the FPS and recent political events has made the FPS study interesting. Since gaining independence from Britain in 1970, Fiji has experienced a period of military coups d'état that have radically changed the government and the bureaucracy. The ramification of this event on HR selection decisions within the FPS has been severe and has influenced the

HR competency databank of the FPS. Vallance found that through emigration and brain drain, “chronic shortages of skilled staff characterise the public service sectors of many

Pacific Island countries including Western Samoa, Vanuatu, Solomon Islands, Tonga and

Kiribati” (1996:p93 ). In Fiji, however, politics have compounded the crisis. Compared to other Pacific Island countries, the population of Fiji has recorded a much more diverse ethnicity. However, this research has found that the coup d'état era alone cannot be blamed P a g e | 51

for the HRM constraints and associated challenges, as the root cause of instability lies deeply embedded within the societal environment of Fiji. Due to the micro-state structure and fragility of many Pacific Island countries, remuneration has always been a major disincentive to those employed by their respective public services. The vulnerability of many Pacific

Island countries, including Fiji, has also affected their ability to operate their bureaucracies, compensate public servants and retain the required competency level for each public service function. This has posed as a critical challenge for HRM in the FPS.

The national culture of a nation often provides the national temperament which often influences HR selection policies. In some very collectivist and family-oriented societies, selecting someone you know is often a norm and considered ethical. In their study, of the

Iranian public and private sectors Cassar and Bezzina confirmed that “a distinctive challenge facing the Iranian public sector selection programme is nepotism or hiring someone who is known to someone else in the organization” (2007:p165). They found that this is due to the

Iranian national culture of being family oriented and distrusting of outsiders. Hence, a known person will be more committed and loyal to the organisation than a complete stranger. This finding has also been taken into account in the formulation of the theoretical framework for the Fiji study, and expanded to probe the influence of networking relationships (Harrington,

2001).

In regard to this part of the chapter, the following sub-research question was formulated to substantiate the contribution of the specific literature on HRM and the FPS:

What are the HRM challenges facing the FPS and what are the implications on HR

selection decisions?

P a g e | 52

2.3 Human resources selection and the Fiji Public Service

The study of the HR selection function has always been examined under the collective discipline of HRM and assessed as a critical component of the HRM model. Dessler defined

HRM as, “The process of acquiring, training, appraising, and compensating employees, and of attending to their labour relations, health and safety, and fairness concerns” (2012:p4 ).

However, the Personnel Psychology discipline was also highly involved because, as an organisation tries to select the right employee, a key strategy of selection involves effective screening of employees so as “to help reduce the dysfunctional behaviours” (Dessler, 2012: p193). In this regard, personality measures are critical components in personnel selection.

The inception of the HR selection function as a stand-alone theory in HRM began through industrial and organisational psychology studies. Freyd (1923) established the classical view of personnel selection research by proposing a 10-step model for selecting employees in industries. Guion expanded the prognosis through a classic overview of the field by examining “the scientific study of the relationship between man and the world of work” (1965: p817). In his study, McEvoy furthered the investigation and found “that of all the aspects of personnel management, selection is the one area that could be most easily improved to incur the greatest benefit” (1984:p2). He stated that time spent on the development of a personnel selection program could be a very good investment for the owner/manager of a small business.

The early studies on HR selection emphasised the importance of formalising certain

HRM functions including HR selection. In expanding the study on HR selection, Little

(1986) confirmed that as small firms grow from employing just a few people, there is an increasing need for more formalised HRM practices. She was specifically concerned with the areas of: compensation, particularly health insurance and recruiting a qualified work force. In a later study, Guion (1987) found that it is extremely important to formulate the key P a g e | 53

instruments in HR selection. Hence, he found that changes in views of personnel selection predictors, and these include choice of predictors, choice of criteria, methods of data collection, and evaluation of predictors, one of the critical components “is the distinction between the job relevance of a predictor and the valid measurement of a trait” Guion (1987: p201).

HR selection function was then guided as a stand-alone theory in HRM and Personnel

Psychology through the works of Gatewood and Field (1987). They then guided and described the development of a multi-step selection strategy that is useful for small businesses. This strategy is based on the principle of incorporating ‘behavioral consistency’ into selection devices used to make decisions among job applicants, a principle which has been useful in developing various types of selection programmes. They found that the essence of effective selection lies in making predictions about the future job performance of applicants for a given position. But such predictions are difficult, partly because only a limited amount of information can be gathered in the selection process, and because important characteristics of job applicants are hard to measure accurately. In this regard, they confirmed that the quality of selection decisions depends on the accuracy and completeness of the information gathered from applicants. They also confirmed that since the amount of information collected is often severely limited by cost and logistical constraints, an appropriate solution is therefore to develop selection instruments that collect only the most directly relevant information. This is an important lesson for FPS managers.

The studies to assert the critical acknowledgement of HR selection as a stand-alone discipline of HRM is critical for this study. Rynes (1991) complemented Gatewood and

Field’s (1987) finding and initiated the examination of HR recruitment and selection as a discipline. He found that Guion (1976) offered two substantive conclusions on recruitment: that little recruitment research existed before1976, and where it did, it was not characterised P a g e | 54

by a ‘search for understanding’. Since that time, the empirical literature on recruitment has expanded considerably and researchers have begun to speculate as to how recruitment might influence applicant and employee behaviours (Rynes, 1991).

HR selection is more closely related to recruitment than it is to the other HRM programmes, because recruitment and selection are concerned with placing individuals into jobs (Gatewood, Barrick and Field, (2011)). Studies by Rynes (1991), Barber (1998) and

Taylor and Bergmann (1987) contributed to a conclusive definition which confirmed that recruitment as the process of attracting a group of individuals to an organisation as the organisation strives to attract applicants who will fulfil its needs. Further, recruitment has three (3) major purposes and these are: to increase the pool of job applicants with reasonable costs, to meet the organisation’s legal and social obligations regarding the demographic composition of its workforce and to help increase the success rate of the selection process by reducing the percentage of applicants who are either poorly qualified or have the wrong skills

(Rynes, 1991).

The value of recruitment to the organisation increases as the pool of qualified applicants grows, which essentially means that a larger pool results in more qualified applicants for the same number of positions (Rynes, 1991). However, if the recruitment programme produces a small number of applicants relative to the number of available positions, the situation may be approaching what is termed ‘hiring’ rather than ‘selection’.

This element is crucial in the hiring processes in the FPS, because if turnover is high and well-qualified personnel are moving to ‘greener’ pastures, either internal or external, then a critical research question that needs to be examined is whether the FPS has been engaging a recruitment process without objective selection decision processes. Hence, if the recruitment process overrides selection, then this may affect HR selection decisions in a drastic manner. P a g e | 55

HR selection is the process of collecting and evaluating information about an individual in order to extend an offer of employment, and such employment could be either a first position for a new employee or a different position for a current employee (Rynes,

1991). Hence, the selection process is performed under legal and environmental constraints and addresses the future interests of the organisation and of the individual. Barber et al.

(1994) also stated that selection is the process by which companies decide who will or will not be allowed into their organisations. In selecting personnel to fill job vacancies, managers are usually faced with important questions that must be addressed and these include: what tasks are new employees required to perform? what knowledge, skills and abilities (KSA), i.e. employee or job specifications, must new employees possess to perform these tasks effectively? If certain specifications are used, is it possible to find or develop selection instruments, such as tests or employment interviews, which could be used in making selection decisions, and for any job under study, what factors or measures represent job success? (Rynes, 1991).

HR selection function as a stand-alone theory in HRM exemplifies the strategic importance of HR selection within HRM. Complementing Gatewood et al., (2011), this finding provides the basis for presenting HR selection as a distinctive function of HRM in this thesis. In this regard, HR selection is critically linked to HRM when HR selection decisions are aligned to the business outcomes of organisations. In other words, employees must efficiently perform their individual work tasks necessary for the firms to reach their goals, and align with the organisation’s overall directions. The above finding also provides justifications to highlight the linkages between HR selection and other HRM sub-functions as the organisation aspires to meet its stated goals.

In view of this part of the chapter, the following sub-research question was formulated to substantiate the contribution of the specific literature on HR selection and the FPS: P a g e | 56

What are the critical HR selection strategies challenges facing the Fiji Public Service in view of HR selection decisions?

2.4 The strategic approach of human resources management and the Fiji Public

Service

For the FPS, ministerial and departmental HR strategy linkages with the overall strategy of the FPS through the FPSC are critical to enhance the credibility of services rendered. These services include HR selection, and the credibility of how stakeholders perceive its functioning. The key lesson highlighted in this study is the need to frame and formalise a new perspective of HR strategically for the FPS. With this in mind, a chronology for SHRM needs to be presented so that the reader can understand the current HRM development phase of the

FPS and the merits of why it needs to be expanded strategically.

The commentators on SHRM during the last 30 years began with the work of Miles and Snow when they proposed “a strategy typology (defenders, prospectors, analysers and reactors) to highlight specific elements of HR practices to be matched to each strategy”

(1984: p2). This contingency approach meant that the implementation of a particular set of

HR practices would depend on an organisation’s strategy. This new perspective was furthered by Schuler and Jackson when they confirmed that “different strategy types (cost reduction, quality improvement, and innovation) require different types of employee role behaviour which must be aligned with the appropriate HR practices “to ensure that those behaviours take place” (1987: p209).

In trying to align the linkages between HR strategy and the organisation’s strategy,

Lengnick -Hall (1988) proposed that for human resources to affect strategy there needs to be a fit between an organisations business strategy and its HR strategy. Her idea was that the P a g e | 57

firm needs to be flexibile on the application of the strategy which needs to be conducive with the business cycle of the organisation. This is a critical lesson for the FPS, since strategies needs to be flexible with the business cycle of the FPS. In a later study, Baird and

Meshoulam (1988) expanded this notion of fit to incorporate the organisation’s external and internal environments. They proposed that an external fit requires aligning a firm’s HR practice with its strategy, while an internal fit encompasses aligning a firm’s HR practice with each other, mutually reinforcing each other. They went on to complement the work of

Hall and Lengnick (1988), proposing that the “fit between HR practices and strategy will change as an organisation advances through its life cycle stages” Baird and Meshoulam

(1988:p117).

It is important for public service managers to continually assess, modify and reinforce the link between the internal organisational characteristics and relevant HR function. Jackson,

Schuler, and Rivero “demonstrated an empirical relationship between organisational context characteristics (industry sector, innovation strategy, manufacturing technology, organisational structure, organisational size and unionisation) and HR practices” (1989:p727 ). A critical finding from their work was that “strategy is related to HR practices, and that multiple HR systems exist within a single organisation” (Jackson et al., 1989: pp783). This should be an important lesson for FPS managers so to ensure that HR practices and strategies do not conflict or overlap. Jackson et al. found that “personnel practices vary as a function of organisational characteristics” (1989:p781). Likewise, when HR selection practices are strategically aligned to the changes in external and internal environmental factors, the outcome of HR practices may not show significant deviations in ascertaining that reformed

HR practices have contributed positively to productivity. In view of assessing the environmental condition, HR selection practitioners of the FPS may vary strategies along a number of dimensions, therefore considering a number of strategy typologies. The rationale P a g e | 58

from this ideology is to implement HR reform policies relative to the specific economic and political phase of development. This is a key lesson for the FPS.

A major school of thought which has been developing since the 1990s is based on the resource based view (RBV) of the firm. This school argued that organisations need to obtain and use their resources, including their people, so they become the best organisation in their sector (Barney, 1991). Barney argued that “firms possess resources, a subset of which enables them to achieve competitive advantage, and a subset of those that lead to superior long-term performance” (1991: p 100). Resources that are valuable and rare can lead to the creation of competitive advantage. Hence, the success of an organisation therefore depends on building resources such as people, culture and knowledge that are difficult to copy and obtain.

According to this view, “financial, material and technological resources can be copied, but it is the culture, relationships and people that are unable to be copied” (Barney, 1991: p116).

Hence, organisations are therefore able to maximise their competitive advantage by building and developing internal capabilities. A key lesson from Barney’s (1991) work for the FPS is that when making choices regarding HR practices, it is important to acknowledge the influence of formal institutional requirements such as legislation, professional association requirements and trade union influences. A key limitation of Barney’s work in view of the

RBV of the firm is that when organisations build their resources base, which includes people and culture, they are actually building network ties. This key finding has been supported by

Bourdieu (1986) who found that the resource base through social capital depends on the size of the network of connections. In this regard, the FPS study is highly significant because it acknowledges the specific network ties that are indirectly fused into the organisation and impact HR selection policies and decisions. Hence, the limitation of Barney’s work in view of the RBV of the firm for the FPS is that building network ties may obstruct the attainment of CA. P a g e | 59

In their study, Milliman, Von Glinow and Nathan (1991) expanded the HR practice

“fit concept” towards international human resource management (IHRM), while Mesch and

Perry (1995) assessed and analysed the differences between bureaucratic HRM and SHRM on government systems. Even though Mesch and Perry’s (1995) findings did not provide evidence for any significant differences, the study made substantive advances since it was one of the first SHRM studies to be conducted in a public sector organisation. In a later study,

Jackson and Schuler (1995) made a significant contribution to the SHRM literature by identifying the important components of organisational environments. They ascertained that the organisational environments were divided into internal and external elements. The findings by Mesch and Perry (1995) have been expanded in this study towards the FPS.

Adopting a strategic approach to HRM improves organisational performance (Martell and Carroll, 1995). This is an important observation that is reticent to the FPS, where improving organizational performance had been a cumbersome task for managers and other senior executives. In their study, Martell and Carroll (1995) formulated a theoretical framework with measures focussing on the issue of the degree of integration between HRM and strategic processes. A key finding with significance for the FPS survey highlighted that senior HRM executives must possess the skills and knowledge to perform strategic partner roles. . In that regards, they need to be highly competent to fully understand the process of integration. Martell and Carroll further stated that, “in order for HRM executives to transform themselves into key strategic partners, they will need to re-evaluate their priorities instigating an organisational reform and restructuring that must be supported by senior managers and board members” (1995:p255). Hence, the importance of adapting a strategic approach to HR selection for the FPS require that HRM executives are highly competent with the processes of integration and public sector reforms which must be supported by the Fiji Public Service

Commission.. P a g e | 60

In view of political instability in Fiji, the current developmental stage of the FPS could not really adopt a strategic approach to HRM; hence a fit approach may not work at this point in time. In his study, Delery “critiqued the literature on fit in SHRM for being focussed entirely on horizontal fit” (1998:p291), and this advantage for research was used later by

Kepes and Delery when they proposed four types of “internal fit based upon a multi-level conceptualisation of HR systems” (2007:p387).

The SHRM literature advanced as scholars either built on early work by SHRM researchers and complemented it with contemporary findings or critiqued the established

SHRM theories and justified their positions through substantive outcomes. Boxall (1998) complemented Barney and Wright (1998) when he explained how the RBV of the firm could be applied to organisations. Specifically, firms should strive to attain human capital advantage, where the quality of their HR should be better than their competitors’. This element is also applied to organisational processes, and the organisational urge to attain the best organisational processes within the industry.

In their study, Wright and Snell expanded on the earlier work regarding the concepts of fit and flexibility, when they asserted that “strategy should fit with three general conceptual values: HRM practices, employee skills, and employee behaviours” (1998:p757).

In a new dimension, Cabrera and Bonache “focused on the need to align an organisation’s culture with its strategy” (1999: P52). They found that a culture could be an asset for an organisation if it encourages behaviours that support its intended strategy. This finding is crucial for the FPS study, since network ties thorough social capital is a critical element of an organisational culture.

The study of ‘fit and organisational flexibility’ continued to evolve, as researchers tried to discover contemporary methodologies that would propel and maintain organisational performance. Richard, Murthi and Ismail (2007) presented a new paradigm in SHRM when P a g e | 61

they examined the impact of diverse HR strategies in manufacturing versus service-oriented industries. They found a “U shaped relationship” between racial diversity and productivity which was stronger in service-oriented industries. The inverted U-shaped relationship (∩) is derived from the integration of two predictions, that is, the positive and negative competing predictions which describe the linear relationships between gender diversity and performance

(Richard et al., 2007). The integration of these theories means that different ranges of gender diversity (e.g., low to moderate levels of gender diversity) are associated with different worker performance dynamics (e.g., Richard et al., 2002; Richard et al., 2007). This study is also important for HR selection within the FPS study in view of the diversity of its workforce and worker performance potential. Understanding well-grounded theoretical frameworks of

SHRM in the operating mechanisms of the FPS may be handy in managing the inflexibility that often arises. The work of Wright and McMahan (1992) ascertained the need to construct an SHRM theoretical framework specifically for the FPS.

In its endeavour to improve public service delivery, the current regime in Fiji formulated a strategic framework for change, using key milestones to direct and drive the public service machinery. One of the key milestones is enhancing public sector efficiency, performance effectiveness and service delivery, and a significant target output is to reduce the size and cost of government (Fiji Government: The State of The Nation and The Economy

Report, 2008). However, FPS managers should always ensure that the motive for downsizing should be related to shifts in the business environment of the public service, where sound data analysis depicting accurate performance levels of the economic growth phases must dictate and defend the rationale for downsizing. It is important for the FPS to adopt a strategic approach so as to assert the relevance of SHRM and improve organisational performance. As organisations adopt a strategic approach, HRM issues should be explicitly discussed in their strategic plans. FPS managers should be actively involved in the planning process as P a g e | 62

‘strategic partners’ at the corporate level of decision-making with FPS customers, integrating

FPS business strategies with the strategic directions of the Fiji Government.

In view of this part of the chapter, the following sub-research question was formulated to substantiate the contribution of the specific literature for SHRM and the FPS:

Do the HR selection policies of the FPS contribute to the concepts of fit and flexibility to develop a highly qualified HR pool for the FPS?

2.5 Social capital and the Fiji Public Service

The origins of social capital in Fiji provided the basis for the inception of subsystems that formed networking within the major ethnic groupings. The key objective was the attainment of security, prominence and survival. Bourdieu (1980) produced the first systematic contemporary analysis of social capital and defined the concept as the aggregate of the actual or potential resources which are linked to possession of a durable network of more or less institutionalised relationships of mutual acquaintance or recognition. Within the networks, social rules are strictly obeyed and this is prevalent in the Fijian and Indian network systems in Fiji. Social rules are defined as shared beliefs about the behaviours that should or should not be performed in particular situations (Argyle, Furnham & Graham, 1981).

Complementing the work of Bourdieu (1980), Portes argued that:

“it is probable that social structure generates social capital, and through social capital,

actors can gain direct access to economic resources (subsidized loans, investment tips,

protected markets); they can increase their cultural capital through contacts with experts

or individuals of refinement (i.e. embodied cultural capital); or, alternatively, they can

affiliate with institutions that confer valued credentials (i.e. institutionalized cultural

capital) (1998:p5)”. P a g e | 63

Organisational researchers have begun to attribute more importance to social capital theory in an effort to understand the contingencies that govern the relationship between social context, social structure and goal attainment (Leenders & Gabbay, 1999). It is crucial to specifically identify social capital, particularly when it is deemed as an informal factor of production for organisations in both the public and private sectors. With increasing clarity about its definition, Adler and Kwon defined social capital as a resource which accrues to actors through network ties (1999:p5). With the definition universally confirmed, attention has now shifted to identifying the preconditions under which social capital can be attributed with either assets or liabilities. When discussing social capital, the aligning references are

“those stocks of social trust, norms, and networks that people draw upon to solve common problems” (Szreter, 2000:p58). He then defined social capital as the social relationships between people that enable productive outcomes (2000:p57). Developing his ideas Bourdieu confirmed that “depending on the field in which it functions, and at the cost of the more or less expensive transformations which are the preconditions of its efficacy in the field in question, capital can present itself in three fundamental guises: as economic capital, which is immediately and directly convertible into money and maybe institutionalised in the forms of property rights; as cultural capital, which is convertible, on certain conditions, into economic capital and may be institutionalised in the forms of educational qualifications; and as social capital, made up of social obligations (‘connections’), which is convertible, in certain conditions, into economic capital and may be institutionalised in the forms of a title of nobility” (1986:3). In the Fijian context, a person of noble birth automatically inherits credit slips acquired through the conversion of social obligations of the tribe and the intricate network ties entwined within its social framework.

P a g e | 64

Social relationships exist at different levels within an organisation, and the relationships between these social contexts define the social capital of the organisation.

Harrington examined the mechanisms by which certain kinds of social relationships at the individual level facilitate or impede the attainment of objectives at the firm level. She found

“that recruitment and selection based on weak instrumental ties increase organisations’ compositional diversity and commitment to task performance” (2001: p84). Further, she also confirmed that groups based on affective ties can become “over embedded”, limiting the pool of information available for decision-making, and inhibiting debate in favour of social cohesion. These findings highlight Harrington’s contingency thesis of network ties. By testing theories of social capital about group composition – drawn from other areas of organisational research – this study seeks to shed light into the “black box” linking social networks, social structure and employee performance (2001: p 83). This finding is crucial for this thesis because social capital may be informally established as a key determinant in the production of goods and services. Hence, this may go against the formality of public service rendering within the FPS. Harrington argued that social capital effects are contingent on the content of network ties, that is, on the nature of relationships among organisation members, whether primarily instrumental or affective. She classified network ties into two (2) ideal types: instrumental bonds, which are the network ties among co-workers, and affective bonds, which are the network ties based on similarity/attraction processes, such as those between family and friends.

In her research on social capital, Harrington (2001) found that investment clubs were suitable subjects since most have been founded recently and the founding members are still active participants. In contrast, the FPS was founded in 1874, and all of the founding members are deceased. Nonetheless, the current network ties within the FPS began in the founding year. Hence, one can deduce that it was possible to reliably measure the content of P a g e | 65

network ties among group members, and since those ties are still active, it would be intriguing to study how they have shaped employee performance over time. The Fijian traditional institutional arrangement was part of the first government of Fiji in 1874. The measures are still evident today in key institutional arrangements like the traditional land owning units, which had been traditionally aligned to the basic social structure for clans for each village.

In Ghana, religious leaders, whether traditional or modern (e.g., Christian, Islamic, etc.), have also become very influential in decision-making among individuals and business leaders (Salm & Falola, 2002). Therefore, community leaders in Ghana, such as the local chiefs, kings and religious leaders are the guardians of societal norms, shared understandings and expectations, which define socially acceptable practices and behaviour in a community’s business environment (Salm & Falola, 2002). All the above issues have been highlighted because of their relevance to the current study of three FPS institutions in Fiji. Importantly, the issues include similarities in the social structures, traditional and cultural set-up, religious commitments etc. Further, it has been ascertained that within the social systems, for example the community, subsystems emerge which include community relations, family relations and friendship and these have the potential to influence HR selection decisions. In view of the above, Adam and Roncevic (2003) expanded the definition of social capital conceptually and asserted that it is linked to concepts such as civil society and social connectedness.

As evident in Fiji, Acquaah and Eshun’s (2010) study highlighted that the social system of Ghana is highly collectivist and embedded in cultures and traditions that thrive on communal bonds, interpersonal relationships and strong allegiance to community leadership.

Hence, the extended family and broader community (especially kings of ethnic groups, chiefs of towns and cities and village heads, who are seen as the cultural standard-bearers) play a significant role in the lives and activities of individuals and organisations. P a g e | 66

It is important for this thesis to highlight that governments in most developed and developing countries are seeking to build social capital which encourages participation from customers in public organisations based on the belief that it will lead to public service improvement (OECD, 2001). An example of this was evident in Fiji where the concept of community policing was introduced by the Fiji Police Force in the last 20 years. The social capital produced is lower crime rates and improved quality of life, hence triggering public confidence in law and order and public service improvement. The Fiji Police Force project on social capital generation was then relayed to public service improvement when the Fiji Police worked with communities in supporting income-generating projects so that people could improve their living standards and welfare. The key lesson here is that crime rates can be reduced when people are able to subsist sustainably. In the process, once they are financially stable, there is no need for the commission of crimes (www.fijitimes.com.fj). Further, this theory of community public service orientation and networking may also be aligned to other key Fiji government ministries and departments as future policy initiatives.

The relationship between social capital, management capacity and the operations of public services was critically examined in Andrew and Brewer’s (2013) study. They found that performance of public service institutions may be positively influenced by high levels of social capital, and this effect may be strengthened where organisations have considerable management capacity. Hence, their findings suggested that public organisations should undertake processes of community building to help create social resilience in the long run.

Further, aligning citizen participation activities with managerial routines and systems can also enable governments to better realise the benefits of social capital for public service improvement. Explicitly, public servants are often strongly involved in building social capital more generally through a variety of extra-role behaviours. P a g e | 67

Bourdieu also found that “the volume of the social capital possessed by a given agent thus depends on the size of the network of connections he can effectively mobilize and on the volume of the capital (economic, cultural or symbolic) possessed in his own right by each of those to whom he is connected” (1986:10). In Fiji, this finding is highly evident in the social structures of the two main ethnic grouping (Fijians and Indians) where the size of the network of connections have been actively mobilized by actors to infiltrate the private and public sectors.

Finally, critics have questioned the existence (and direction) of a relationship between social capital and government performance (e.g. Jackman and Miller, 1998). In this regard, there are contrasting effects of social capital on organisations. Initially, the blueprint introduced by Harrington (2001) suggests that groups based on instrumental and affective ties will impact the HR selection processes, triggering different outcomes. The outcomes may be a decline or an increase in employee performance. The Fiji Police Force example highlighted that the organisation itself formally approached the network groups within the societal context, and in the process generated social capital which impacted on the core services of the public service institutions. The United States (US) Government has also implemented a policy similar to the community policing adopted by the Fiji Police Force and was more advanced in their approach by enacting legislation to formally guide the key government institution responsible. At the same time, although social capital is often thought to be conducive to better public services, it can sometimes have a ‘dark side’ which harms the efforts of public organisations in improving service delivery. The dark side emulates from a passion driven by social obligations (‘connections’) which has the tendency to disrupt the meritocracy of established organizational processes. Hence, the study sought to ascertain the positive and negative impacts of social capital on HR selection policies, and in particular its influence on HR selection decisions through the main research questions of the thesis. P a g e | 68

In view of this part of the chapter, the following sub-research question was formulated to substantiate the contribution of the specific literature on social capital and the FPS:

Could the key relations of the dominant social groups in Fiji that build and maintain social capital influence HR selection decisions within the FPS?

2.6 Networking ties

In public service institutions, like the FPS, group processes emerge from patterns established by the social context and relevant network ties which initially drew members into the group with its social structure and norms. The benefits and profits which accrue from membership in a group are the basis of the solidarity which makes them possible, and this highly significant in a social grouping that fosters rotating credit association like the Indian merchants in Fiji. However, this may not be relevant in a Fijian social structure which basis itself in a non-rotating credit association model which is very much obligation driven.

Nonetheless, the patriarch is responsible for maintaining security and sustaining subsistence for the village and tribe.

The HR selection processes not only create the resource pool available to public service institutions by regulating who gets in, but also indirectly set up the network commitments for the many decisions that networking members must make together under conditions of uncertainty and anonymity. The essence is that if a civil servant is selected and appointed through one of the streams of networking ties, the network ideology may be adding stress in terms of how the worker conducts himself within the organisation. The appropriate methodologies for HR selection employed by public service institutions for new and current employees of the organisation are among the first decisions that define a work group or team, and such early decisions snowball over time. While this is not an irrevocable process, the P a g e | 69

decisions made in the foundational period of a group’s existence as ascertained by

Harrington, acquire a momentum such that “decisions that have been taken in the past increasingly amount to an imperative for future policy implications” (2001: p86)

Examining the alignment of FPS HR selection policies with Fiji’s social system is crucial in this thesis for a couple of reasons. Firstly, it is argued that networking ties which distort the HR selection processes are generated through social capital emanating from the current social structure framework. Secondly, it is important to ascertain the origins of networking ties in view of the assumption that FPS employees may use their social capital to: gain direct access to economic resources, by being appointed to a public service institution designation, being promoted to a higher position within the organisation, or by increasing their social capital through contacts with experts or individuals with power and authority within the hierarchy, or they may also affiliate with institutions that confer valued credentials in manufacturing social capital which might promote social mobility beyond economic means. In essence, these public service institutions may indirectly provide other social benefits apart from intrinsic and extrinsic values rendered from within, and these include job security, comforts and safety of belonging to superior religious, school, work, community, extended family and friendship clusters.

To this extent, Table 2.1 below summarises the key definitions for instrumental and affective ties. A key definition stated that instrumental ties have low intrinsic value of relationship outside task setting, while affective ties have a high intrinsic value of relationship

Harrington (2008). In an instrumental-based society, members would perceive that any relationships outside the work context have low intrinsic value and would be considered low in significance. On the other hand, an affective-based society would perceive that any relationships outside the work context have high intrinsic value (Harrington, 2008). This finding is highly prevalent in family-oriented social systems where the collectivistic P a g e | 70

ideologies mandate members to oblige the collective decision of the network. The instrumental-based society and networking are governed by mutual utility and contractual obligations, while affective ties are governed by norms of social obligations characterised by status and seniority. Harrington found that, “instrumental based societies and networking exemplifies rationalising decision making so that a particular decision may produce the greatest good for the greatest number of individuals”

(2008:P175). However, an affective based society and networking is controlled and managed by social norms and their respective obligations including the traditional duty to reciprocate in cash or kind (Harrington, 2008). Lastly, Harrington (2008) found that an instrumental-based society would provide utilitarian support, including economic and information resources, while an affective-based society and networking system would provide socio-emotional support. As a result, an instrumental based society and networking would provide incentive for economic gains, whilst an affective ideology may support the contrary (Harrington, 2008).

Table 2.1 Characteristics of instrumental and affective ties

Instrumental Affective

Low intrinsic value of relationship outside High intrinsic value of relationship; would task setting persist outside utilitarian context

Low psychological attachment among actors High psychological attachment among actors

Provides utilitarian support, including Provides socio-emotional support, potentially economic and information resources. economic or utility gains

Governed by mutual utility, contractual Governed by norms of social obligation, obligations including reciprocity and ascribed characteristics such as status or seniority

Source: Harrington (2008)

P a g e | 71

In view of this part of the chapter, the following sub-research question was formulated to substantiate the contribution of the specific literature on networking ties:

What are the key characteristics of instrumental and affective ties within the FPS that have the potential to influence HR selection decisions?

Do the demographics of employees in the FPS ministries examined correlate with a level of sustained social capital and social networking?

2.7 Justice and employee discretions in the Fiji Public Service

Fiji public servants’ perceptions of justice have been selected because employees’ discretions are induced after workers have assessed how just or unjust their work environment is in relation to their work status. Employee discretions are an employee’s ability to decide how much effort to contribute to work, i.e. the labour effort effect. By convention, equity theory indicates that outcomes should be distributed in proportion to inputs (Adams, 1963). To that effect, it also explains why giving one person a promotion or pay rise can have a demotivating effect on others and their peers. In the process, when people feel fairly or advantageously treated, they are more likely to be motivated. In contrast, when they discern that they have been unfairly treated, they may be demotivated and display disaffection behaviour. On this note, employees seek to perceive equity between the inputs that they bring into a job versus the outcomes that they receive from it against the perceived input and outcome of others. Equity theory complements justice theories in this thesis because Fiji public servants value fair treatment, which causes them to be motivated to keep the fairness maintained within the relationships of their co-workers and the organisation. In his study,

Goodman (1980) presented a theoretical framework for determining how people evaluate their pay, and ascertained that employees use multiple referents. In this regard, he found that P a g e | 72

people identify three (3) classes of referent (other, system and self) which are significantly associated with their feelings about pay satisfaction. Hence, when Fiji public servants discern that HRM work policies, including HR selection, rewards and compensation, performance, training and development, organisation of work etc., are fair and just, they will then produce and present the desirable employee work performance expected of them which are exhibited through the various desired work attitudes and behaviours.

In this study (refer to Figure 2.1 at the end of the chapter), employee perceptions of justice have a strong contingent effect on the relationship between the independent variables (network ties) and the dependent variables (employee reactions to perception of justices).

In his study, Gilliland found “a positive relation between selection procedural justice and the likelihood of applicants recommending the organisation to others as a place of work, or endorsing the organisation’s products and services” (1993: p 695). Further, the perception of selection fairness has been shown to be related to job acceptance (Singer, 1993), work attitudes and work performance (Gilliland, 1994), and to perceptions of organisational attractiveness (Bauer, Maertz, Dolen & Campion, 1998). It is obvious that employees of most organisations perceive procedures to assess their value against the group, organisation or authority using the procedure. Fair procedures lead to positive feelings about one because they signify respect by the group or authority that enacts the procedure. In contrast, unfair procedures will lead to negative feelings about one-self for they indicate low regard by the group or authority figure.

In administering HR selection strategies, FPS managers should take note that each stage may present different types of perceptions and reactions and the relationships between process fairness and behavioural intentions at both the pre-application and post-offer stages of the selection process. In their study, Ployhart & Ryan (1997) used the organisational P a g e | 73

justice and attribution theory frameworks as their discipline base to understand the processes by which applicants perceive and react to selection procedure and decisions. They found that behavioural intention variables were chosen because they have been studied in past research and have practical significance for organisations. In the study on the FPS, the focus is on current employees’ perspectives of past and current selection processes. This highlights a key substantive finding for the thesis.

Schroth and Shah (2000) highlighted that even though their study examined the effects of procedural justice on self-esteem, future research should examine the reverse, i.e. the effects of self-esteem on procedural justice. Another interesting avenue that needs further investigation is regarding one’s desire to understand the procedures, i.e. when one is sceptical and curious about certain procedures, the knowledge of the procedure threatens his/her self- perceptions which affect his/her belief and self-esteem. Schroth and Shah (2000) question if knowledge of the procedures threatens our self-perceptions under certain conditions: do we really want to know the procedures used? This is particularly relevant in cases of negative outcomes, when ignorance of the procedures enables people to make self-serving external attributions. Schroth and Shah (2000) also proposed that future studies could examine conditions under which individuals actively seek procedural information. The above theory is very much relevant to the FPS study because as one perceives an unfair procedure, this person will produce negative feelings about himself/herself and alternatively these negative feelings will induce negative work performance. Further, the accumulation of negative informal information about certain procedures and policies, plus the ignorance and lack of knowledge about those policies by certain stakeholders, can add to the negative self- perceptions by the person, his peers and his wider external grouping and network.

In many institutions, the study of organisational justice explores employee perceptions of the fairness of organisational decisions and decision-making procedures. P a g e | 74

Gilliland & Chan proposed four instruments for measuring justice in the procedures of an organisation: the issue of validity in the use of generic versus domain specific scales, the validity of inferences drawn from laboratory research versus filed studies, the levels of analysis issues relevant to the valid measurement of justice constructs, and finally, the assessment of relationships between justice and other constructs of interest. Their study suggested how these instruments might be used to critique the dominant theories and methods used to investigate justice in organisations (2001: p143).

In view of the perceptions of Gilliland and Chan’s (2001) findings, fairness can be divided into two (2) major categories: fairness of outcomes (distributive justice) and fairness of processes (procedural justice). An additional type of justice was later added to the two justice constructs as researchers realised that people also consider interpersonal treatment to be a form of justice and this is termed ‘interactional justice’. A limitation stated in Gilliland and Chan’s (2001) work is the need to examine justice perceptions in the context of diversity training, since diversity training provides an excellent context to investigate several justice issues. The justice variables or instruments developed by Gilliland and Chan (2001) has been incorporated to the FPS theoretical framework (refer to Figure 2.1 at the end of the chapter) because its four instruments are particularly important to the advancement of justice research and they are: individuals’ subjective perceptions /self-serving biases (i.e. perceptions of distributive justice); test content and methods/selection tools and how they are implemented

(i.e. perceptions of procedural justice); quality of interpersonal treatment received (i.e. perceptions of interactional justice) and adequacy of information conveyed (i.e. perceptions of informational justice.

It is highly likely that employee perceptions of justice in the FPS are formed through either their expectation or experience of HR selection policies and implementation. In developing their theory of organisational justice, Bell, Ryan and Wiechman (2004) examined P a g e | 75

the association between organisational justice and applicant perceptions in an HR selection environment. Specifically, they examined and tested three sources of expectations – direct experience, indirect experience and other beliefs. Their research design was formulated around three underlying assumptions. The first was that justice expectations in a selection context are derived from and greatly influenced by an applicant’s past experience in the same institution, taking into account his/her personal characteristics and the environmental influence. The next assumption was around the idea that justice expectations in an HR selection context are derived from and greatly influenced by a person’s existing beliefs about an organisation, taking into account his/her personal characteristics and the environmental influence. The final assumption was that justice expectations in an HR selection context are greatly influenced by an applicant’s indirect experience in the same institution, taking into account his/her personal characteristics and the environmental influence. These assumptions have been taken into account for the FPS research. According to Bell et al. (2004), the correlation between the respective personal and contextual factors and antecedent behaviour will induce a justice expectation measure which will then produce the relevant outcome behaviour.

The relevance of Bell, et al.’s (2004) research on the proposed Fiji study is substantial for the following reasons: firstly, an applicant’s past experience in HR selection decisions in one of the public service organisations in Fiji may influence his or her current justice perception level. Further, a person’s extant beliefs about an organisation, taking into account his/her personal characteristics and the environmental influence, are a strong determinant of justice perception and this is crucial for the FPS study. Lastly, an applicant’s indirect experience in the same organisation, taking into account his/her personal characteristics and the environmental influence, for example, he or she may not be formally advised as to why he/she was not selected for the job, has significant implications on his/her justice perceptions. P a g e | 76

For many job applicants, the selection process is the first official contact they make with the organisation. Chan and Schmitt (2004), in their seven-area construct-oriented agenda for research on applicant reactions to selection procedures, identified justice principles as one of the four major determinants of applicant reactions. Hence, the perceived fairness of the selection process might be expected to affect an applicant’s perceptions of the organisation as a whole. In essence, applicants often share their experience and justice perceptions of HR selection to peers in view of certain memorable experiences which exposed the rationality of their comprehension.

Organisational justice theory was concerned exclusively with distributive justice, i.e. the perceived fairness of outcomes (Lazar, Zinger & Lachterman, 2007). Further, workers were also concerned not only with fairness of the distributive aspects of organisational decisions, but with the fairness or justice of the procedures associated with and leading these decisions. A key significance of Lazar et al.’s (2007) research on the FPS survey is that they found workers were concerned not only with fairness of the distributive aspects of managers’ decisions, but with the fairness or justice of the procedures associated with making these decisions. Hence, procedural justice is crucial to the study on the FPS because public service managers, applicants and current employees should all be concerned with the fairness of not only distributive aspects of manager’s decisions, but also with the fairness or justice of the procedures underpinning managers’ decisions.

In view of this part of the chapter, the following sub-research question was formulated to substantiate the contribution of the specific literature on justice and employee discretions within FPS, depicted as Case Study Three and presented in Chapter 7:

Sub-research question five (a):

Do employees perceive justice and/or injustice with the HR selection decisions of the FPS? P a g e | 77

2.8 Conclusion

All in all, this chapter strongly acknowledges that conducive HR selection decisions for the

FPS could be configured after stakeholders carefully deliberate on key issues such as: the application of a relevant HRM and HR selection theoretical framework, consideration of

Fiji’s social environment, acknowledgement of FPS employees’ justice perceptions on HR selection decisions and policies, and the state of Fiji’s political and economic environment.

The following are the sub-research questions developed from the literature review to focus the data collection:

Sub-research question one:

What are the HRM challenges facing the FPS and what are the implications on HR selection decisions?

Sub-research question two:

What are the critical HR selection strategy challenges facing the FPS in view of HR selection decisions?

Sub-research question three:

Do the HR selection policies of the FPS contribute to the concepts of fit and flexibility to develop a highly qualified HR pool for the FPS?

Sub-research question four (a):

Could the key relations of the dominant social groups in Fiji that build and maintain social capital influence HR selection decisions within the FPS?

Sub-research question four (b):

What are the key characteristics of instrumental and affective ties within the FPS that have the potential to influence HR selection decisions? P a g e | 78

Sub-research question four (c):

Do the demographics of employees in the FPS ministries examined correlate with a level of sustained social capital and social networking?

Sub-research question five (a):

Do employees perceive justice and/or injustice with the HR selection decisions of the FPS?

In answering the central questions that drive the thesis, the chapters have been organised to correspond to the literature review in view of the above-mentioned sub-research questions. In this regard, the following chapter will exemplify the methods employed in gathering the raw data, as well as in analysing the data as driven by the stated sub-research questions. Specifically, the chapter describes the research methods used for this thesis and presents a frame of analysis of the outcome of structured survey and archival data analysis.

P a g e | 79

FIGURE 2.1 – THEORETICAL FRAMEWORK FIJI CASE MODEL

Group Heterogeneity (Control - A) 1. Age 2. KSA

3. Ethnicity

Instrumental Ties (X1)

1. School 2. Political 3. Religious Employee perceptions of justice of selection system & Employee reactions to perceptions Affective Ties (X2) 1. Selection 1. Community ties 2. Pay 2. Extended family 3. Turnover 3. Friendship 4. Tardiness

Perceptions of Justice (Moderating Variables) 1. Fairness of HR selection processes and how well they are implemented in the organization (Distributive Justice)

2. Fairness of current HR selection policies and how they are implemented (Procedural) 3. How respectful and sincere HR managers are when enacting HR selection procedures (Interpersonal) 4. The extent to which HR selection procedures are honestly and adequately explained by HR managers

(Informational)

P a g e | 80

CHAPTER 3

METHODS

P a g e | 81

Table of Contents

3.1 Introduction 3.2 Research questions, units of analysis and data 3.3 Response rate 3.4 The embedded case study 3.4.1 Case study organisation sample 3.5 Data collection methods 3.6 Design and administration of data collection methods 3.7 Qualitative data analysis 3.7.1 Qualitative data analysis methods 3.7.1.1 Pattern Matching 3.7.1.2 Explanation building 3.7.1.3 Time Series Analysis 3.8 Quantitative data analysis 3.8.1 Survey Data analysis 3.8.2 Method 1: Descriptive analysis 3.8.3 Method 2: Factor Analysis 3.8.4 Rules of Thumb for retaining factors 3.8.4.1 Rules of Thumb 1: Kaiser Guttman Criterion 3.8.4.2 Rules of Thumb 2: Gorsuch’s subject Scree Test 3.8.5 Rotation 3.8.5.1 Retaining factor loadings greater than an arbitrary threshold 3.9 Ethics 3.10 Conclusion

P a g e | 82

3.1 Introduction

This chapter describes the research methods used for this thesis and presents a frame of analysis of the outcome of structured survey and archival data analysis. In his study, Yin confirmed that having well designed methodology is the logical sequence that connects the empirical data to a study’s initial research questions and ultimately to its conclusions (1989).

In this regard, this chapter initially covers justification of the methodology in terms of the research problem and literature review. In particular, it highlights an awareness of the strengths and weakness of the positivist and phenomenological paradigms as a basis for discussing the methodological constructs of the survey (Perry and Coote, 1994).

The chapter also explains the unit of analysis and subjects and sources of data collected in the research. Within the analysis proper, the following methodologies were employed. For Case Study One (Chapter 5) and under a qualitative framework, explanation building was employed and this was complemented by time series analysis. In Case Study

Two (Chapter 6) under a qualitative framework where data triangulation was employed, multiple-level analysis was used. The levels included: the FPS organisation as a whole, ministries and departments in ministries and individual employees of the FPS. For Case

Study Three (Chapter 7) and through a multiple-level analysis, descriptive and factor analysis were used to examine sub-research questions of the study. This is in line with Anbari’s

(2002) work, where he found that descriptive statistics helps us boil down raw statistics into an interpretable digestible form. Further, Chapter 7 also uses factor analysis, acknowledging

Landau and Everitt (2004), to investigate the underlying structure of the association between the 10 items per questionnaire in the 10 questionnaire survey for the FPS. The multiple-level analysis used was Yin’s (2004) framework of analysis.

P a g e | 83

3.2 Research questions, units of analysis and data analysis

The following sub-research questions developed from the literature review (Chapter 2) in association with the Fiji context narrative (Chapter 4) have been organised to be examined as

Case Studies One, Two and Three in Chapters 5, 6 and 7. The organisation of the sub- research questions and corresponding chapters are presented in Table 3.1.

Table 3.1 Sub-research questions and chapters

Chapter 5 Chapter 6 Chapter 7

Sub-research One, Two and Four (a), (b) and (c) Five (a)

questions Three

Sub-research question one:

What are the HRM challenges facing the FPS and what are the implications on HR selection decisions?

Sub-research question two:

What are the critical HR selection strategy challenges facing FPS in view of HR selection decisions?

Sub-research question three:

Do the HR selection policies of the FPS contribute to the concepts of fit and flexibility to develop a highly qualified HR pool for the FPS?

P a g e | 84

Sub-research question four (a):

Could the key relations of the dominant social groups in Fiji that build and maintain social capital influence HR selection decisions within the FPS?

Sub-research question four (b):

What are the key characteristics of instrumental and affective ties within the FPS that have the potential to influence HR selection decisions?

Sub-research question four (c):

Do the demographics of employees in the FPS ministries examined correlate with a level of sustained social capital and social networking?

Sub-research question five (a):

Do employees perceive justice and/or injustice with the HR selection decisions of the FPS?

As shown in Table 3.1, the chapters have been organised to correspond with the relevant research questions.

For the qualitative framework employed and exemplified by Yin (1984), the objective is to analyse the case data by building an explanation about the case. In this regard, the objective is to explain a phenomenon by stipulating a set of causal links. This is a preferred and relevant framework for analysing qualitative data. Using the explanation building model, a crucial question arises and that is: how did the causal links reflect critical insights into

HRM focus? For example, the current system of governance for the FPS was formed in view of many influences from the Fiji social context over the years; hence the iterative nature of causal links is demonstrated through explanation building following a timeline.

Further, explanation building was completed by time series analysis, which is an analytical framework that portrays a chronology, which is a frequently used technique in case P a g e | 85

studies. Yin (1984) asserted that for analysis of chronological events, the analytical goal is to compare the chronology with that predicted by some explanatory theory. The chronology and qualitative data assessed within the qualitative framework depict the origin of the formation of the government of Fiji, and critical events that led to its development and the modification of HR selection policies and practices. Notable events include the cession of Fiji in 1874, the independence of Fiji in 1970, the beginning of the coup era in 1987, and the annexation of

Fiji as a republic.

For the quantitative analysis, a mixed method was employed which will explore both a descriptive and an inferential analysis. Initially, descriptive statistics are distinguished from inferential statistics (or inductive statistics), in that descriptive statistics aim to summarise a sample, rather than use the data to learn about the population that the sample of data is thought to represent. Hence, under descriptive statistics, the analysis began by examination of relevant summary statistics and a variety of graphical illustrations.

3.3 Response rate

In the study, the researcher distributed 100 questionnaires randomly for each ministry and the response rate for each ministry (see Table 3.2) was recorded as follows: Ministry of Works,

Transport and Public Utilities – 75 per cent; Ministry of Lands and Mineral Resources – 54 per cent, Ministry of Agriculture – 40 per cent. A total of 169 usable responses resulted from this survey, with a total response rate of 56 per cent. The average rate of individual participation in the study within groups responding was 70 per cent of the membership (sd. =

0.18). The high response rate was probably due to the formal process of acquiring grant approval through the ethics requirement of the survey, where departmental heads’ approval was granted prior to the survey being conducted. While it was not possible to compare the P a g e | 86

sample frame for this study with the entire population of the FPS, which according to the latest census in 2011 stands at around 17,819, analysis of the non-respondents indicated no difference in terms of composition, size or location compared to ministries that did participate in the survey. Zikmund stated that response rate is defined as the number of questionnaires returned or completed, divided by the total number of eligible people who were contacted or asked to participate in a survey (2003:p215).

Table 3.2 Ministry response rate

Ministry No. of No. of Response rate questionnaires questionnaires distributed returned and completed Works, Transport, Energy, Communication and Public 100 75 75% Utilities

Lands and Mineral Resources 100 60 60%

Agriculture 100 40 40%

Descriptive analysis helps us boil down raw statistics into an interpretable digestible form (Anbari, 2002) and this was relevant in this analysis because it ascertains whether instrumental ties variables do exist and, most importantly, confirms the magnitude of their presence and significance in terms of testing our hypothesis.

The objective of the chapter is to present a frame of analysis to empirically examine archival narratives and survey data with regard to the central and sub research questions of the thesis. Specifically, the findings will describe the general distributional properties of the P a g e | 87

data in the survey, and any unusual observations (outliers) or patterns that may cause problems for later analysis to be carried out on the data.

3.4 The Embedded Case Study

In this thesis, the study of a single case (see Fig 3.1) has enabled the researcher to investigate a phenomenon in depth, enabling a rich description and revealing its deep structure (Cavaye,

1996). The purpose of a single case study can be description, discovery or testing. A full and rich description of a rare phenomenon contributes to knowledge (Yin, 1989). Further, single case studies enable theory building by developing and refining concepts (Pettigrew, 1985), and can also be used for theory testing by confirming or disconfirming theory (Markus,

1989). The investigation of a single case can be designed so that more than one unit of analysis is embedded within the case, as evident in the FPS study, where a structured survey and case study analysis using both quantitative and qualitative analysis are employed. On that note, case study evidence can involve multiple types of data and data collection techniques.

In addition to including both qualitative and quantitative data, the evidence can cover two different levels of analysis: a main unit (the case) and a subunit (embedded within the overall case) (Yin, 2004). For instance, in the FPS study, the case itself (FPS) has provided the setting, but within that setting several instances of the phenomenon are present, hence the three FPS institutions examined (see Figure 3.1). For this phenomenon and ideology, contrasting observations from several units of analysis within one case can create and highlight theoretical constructs (Yin, 1989).

This chapter also demonstrates how case studies can include advanced (or very advanced) quantitative analysis, and in doing so it reinforces the definition that a case study can rely on qualitative or quantitative data, hence calling upon any number of analytical P a g e | 88

techniques (Yin, 2004). In this regard, development of theory is a central activity in organisational research, as shown in the FPS research. Traditionally, authors have developed theories by combining observations from previous literature, common sense and experience.

However, the tie to actual data has often been tenuous (Perrow, 1986, Pfeffer, 1982). This issue initially challenged the constructs of this thesis, however the works of distinguished scholars and researchers like Yin (1984, 2004), Eisenhardt (1989), Cavaye (1996), Perrow

(1986), Pfeffer (1982) and Harrington (2001) enabled the author to synthesise the conclusions based on the data collected and analysed.

The case study is a research strategy which focuses on understanding the dynamics present within single settings. The definition could be contrasted with other renditions of the case study method, which tend to limit case studies to either qualitative data or a subset of field techniques; hence, as proven through testing its constructs, it has tenaciously withstood all challenges with extremely well. Case studies can involve either single or multiple cases, and numerous levels of analysis (Yin, 1984). For example, Harris and Sutton (1986) studied eight organisations on the brink of collapse; Bettenhausen and Murnighan (1986) focussed on the emergence of norms in 19 laboratory groups; and Leonard-Barton (1988) tracked the progress of 10 innovation projects.

Moreover, case studies typically combine data collection methods such as archives, interviews, questionnaires and observations. The evidence may be qualitative (e.g., words), quantitative (e.g., 534 numbers), or both. For example, Sutton and Callahan (1987) rely exclusively on qualitative data in their study of bankruptcy in Silicon Valley; Mintzberg and

McHugh (1985) use qualitative data supplemented by frequency counts in their work on the

National Film Board of Canada; and Eisenhardt and Bourgeois (1988) combine quantitative data from questionnaires with qualitative evidence from interviews and observations. Finally, case studies can be used to accomplish various aims: to provide description (Kidder, 1982), P a g e | 89

test theory (Pinfield, 1986, Anderson, 1983), or generate theory (e.g., Gersick, 1988, Harris &

Sutton, 1986).

Yin (1989, 2004), who is a strong proponent of the deductive use of case research, confirmed Cavaye’s (1996) finding and asserted that case research methodology starts with the construction of specific theoretical propositions derived from a theoretical model. He further argued that during data collection, information is gathered on the individual variables indicated by the model. At the analysis stage, the factual conditions and relationships are compared to the theoretical ones. However, using the principles of analytical generalisation, the early model is tested and may be modified according to case findings. This deductive approach is also advocated by Yin (1989), and Benbasat et al. (1987). The use of case research for deductive purposes represents a non-traditional way of testing theory. Cavaye further stated that” since case research has not been traditionally used for testing theory, it does not make the case strategy any less appropriate for this purpose than, for instance, survey strategies” (1996: p 235). In this regard, a research strategy aiming at validation or theory testing should be able to confirm or disconfirm existing theory. All in all, since deductive logic can be used in both case and survey research, and the appropriateness of employing a case research strategy for theory validation is unquestionable.

Lastly, Yin (2009) stated that all empirical research studies including case studies have a story to tell which must have a beginning, end and middle. In this regard, an analytical strategy will be the researcher’s guide in crafting a narrative, and will also help to use tools and make manipulations more effectively and efficiently. Yin (1984) confirmed that there are four strategies involved, and these include relying on theoretical propositions, developing a case description, using both qualitative and quantitative data and examining rival explanation.

The analysis of data interchangeably used the three (3) Fiji ministries for the qualitative and quantitative analysis. Some ministries were quite flexible in the dissemination P a g e | 90

of the required data, while others were a bit rigid. The Ministry of Agriculture, for example, gave comprehensive data sets including productivity proxy variables, while the other two (2) ministries did not. The reasons for the inconsistency were various, mainly the unavailability of staff and the current political climate in Fiji. Some staff members were reluctant to disclose information even though the FPSC and the office of the Prime Minister had granted research approval. Therefore, the Ministry of Agriculture data was often used where appropriate as a case sample within the case study analysis in view of its reliability and comprehensiveness.

3.4.1 Case study organisation sample

There are three ministries in the FPS survey case study sample (see Figures 3.1).

Figure 3.1 – Fiji Public Service embedded case study design

Case Study Organisations

Fiji Context Social Capital of the Social Network (Social Environment)

Individual

Ministry of Ministry of Works ad Lands and Public Utilities Ministry of Mineral (Operational) Agriculture Resources and Primary (Operational) Industries (Operational)

Fiji Public Service (Strategic) P a g e | 91

The embedded case study approach was also preferred for this research because it is an effective research strategy in examining past and contemporary events (Yin, 2009). In the process, Schramm (1971) reiterated that the essence of a case study is that it tries to illuminate a decision or set of decisions, including why they were taken, how they were implemented, and with what result, hence the illumination of those decisions are important to the study on the FPS HR selection system. Yin (2004) also demonstrated how case studies can even include rather advanced quantitative analysis, and in doing so reinforce the definition that a case study can rely on qualitative or quantitative data, hence calling upon any number of analytical techniques. This is evident in the thesis, as the entire subject for the Fiji research is a single case study analysing the impact of the social context on the functional areas of HRM and productivity, particularly HR selection in the public service.

3.5 Data collection methods

For the qualitative data, archival data and publicly available documents on the historical, operational, societal and HRM aspects of the FPS and the government of Fiji as a whole were gathered. This involved collecting historical documents from the Fiji archives and other reputable learning institutions, including the National Library of Australia, The University of

New South Wales@The Australian Defence Force Academy and the NSW State Library.

Other publicly available documents were gathered from the three ministries in Fiji, the FPSC,

Fiji Archives, the Tongan Public Service Commission, the Western Samoan Public Service

Commission and the Solomon Islands Public Service Commission.

For the quantitative data, a survey was designed with structured questions. The survey was administered by the researcher in each ministry by arrangement with a civil servant P a g e | 92

contact person. For every ministry, the appointed contact person then handed out the surveys and I returned to pick them up at a later date.

Survey sample: I surveyed three Fiji Government ministries classified under the membership of the FPS, an umbrella institution for public services in Fiji. The three (3) government ministries surveyed provide specialised services in the area of Lands and Mineral

Resources; Agriculture, and Works and Public Utilities (Energy, Roads, Communication and

Transport). They provide vital and essential services to the Fiji public through the maintenance and management of works, transport and public utilities; the maximum utilisation of all state land that also enables the exploration and mining of mineral resources; and the enhancement of food production and income security through the agricultural sector.

3.6 Design and administration of data collection methods

Survey administration: After the questionnaire was prepared, it was sent to my supervisors for review and comment. After minor modifications as per their comments, the questionnaire was endorsed by the Human Research Ethics Advisory Panel of [email protected] questionnaire was then taken to Fiji and distributed randomly to employees of the Ministry of

Works and Public Utilities, Ministry of Lands and Mineral Resources and Ministry of

Agriculture and Primary Industries.

The researcher then adopted the interactive procedure of questionnaire administration whereby group administration of the questionnaires was implemented, and the delivery of questionnaires was standardised, so that it was exactly the same for each respondent. To achieve the objective of standardisation, the researcher organised meetings with the

Permanent Secretaries and Chief Executive Officers (CEOs) of the respective ministries in

Fiji, where the objectives of the survey and other issues pertaining to the research were P a g e | 93

comprehensively explained. In the process, the questionnaires were disseminated within the respective ministries to the respondents from a centralised dissemination point, and the

Administration and Corporate Department was given this task. Further, in group administration, a standardised format was important to ensure that, as far as possible, the responses were based upon the same cause (question) and effect (response) relationship. Page and Meyer (2000) stated that administering questionnaires in this way is particularly useful for collecting quantitative data in association with qualitative data. They further stated that it should really be used when researching sensitive issues that cannot be discussed in group settings, or when prompting for further responses is required. The stated questionnaire administration methodology was well received by respondents given Fiji’s current political, economic and social context. Hence, it ensured that the researcher had more flexibility about the meaning or intent of the research items, controlling for prompts.

Considering that the key theoretical foundation of the thesis emulated from

Harrington’s (2001) theoretical model, the sample was systematically selected in view of the importance of the depth of public services rendered by the selected Fiji ministries. Each government ministry received a packet containing survey instruments, with two major parts.

The first one was purposely designed to glean individual-level survey information or biodata.

The second part was constructed to gather data from individual civil servants, their individual discretion on employee performance proxies, given their perception of network ties in their respective organisations. Individual survey participants each filled out a 15-page survey, including 100 multiple choices on a 7 point Likert scale about their instrumental ties, affective ties and perceptions of justice. In the process, 100 questions were developed for 10 subsections of the questionnaire from the following causal relationships: instrumental ties

(school, political, work) and affective ties (communities ties, extended family, friendship) P a g e | 94

correlated against productivity, perceptions of justice on the HR selection systems taking into account distributional, procedural and interactive justice.

The survey instrument: As stated, the survey methodology involving a structured questionnaire was used to test the central research question and respective sub-research questions for the different stages of the study. Specifically, the study analysed the hiring process in the FPS using the stated mixed method approach: a structured survey, which was followed by an analytical narrative where archival data pertaining to the HR selection processes of the FPS were meticulously gathered and examined. Page and Meyer (2000) ascertained that the term ‘archival’ refers to any historical records and documentation held by companies, libraries, specialist institutions and so on, that may or may not be within the public domain.

The questionnaires and research instruments were then subjected to a pilot study (see

Appendix 9.6) aimed at obtaining an insight into the real conditions of the actual study, given the current era of political instability in the Fiji Islands. In the process, the prime objectives of the pilot were to assess the initial reactions of the public servants and ascertain what elements of the research instruments were mostly provocative and sensitive given the current instability. This would enable the researcher to anticipate and adjust to potential problems during the full-scale research and ascertain the validity of the survey instruments by modifying the current instruments so as to ensure that they were measuring what they were supposed to measure and not something else (Maiyaki & Mokhtar, 2011). Maiyaki and

Mokhtar (2011) explored a small sample data to determine customer behavioural responses in

Nigerian retail banking, where the context of the survey guided them to examine instrument validity, reliability and, subsequently, data normality.

Survey Design: To increase reliability, the design of the questionnaire for the survey used pre-existing scales and the following measures are highly evident in the design: P a g e | 95

perceptions of justice; distributive justice; procedural justice; interactional justice; and informational justice. For network ties: instrumental ties (school, political, work) and affective ties (community, extended family and friendship) are the key variables in the construct. For perceptions of justice/injustice: the extent to which selection procedures are honestly and adequately explained by FPS HR managers, perceptions on fairness of HR selection procedures and policies, and how they are implemented, and the impacts of the perceptions of justices/injustices on turnover rate, tardiness and absenteeism.

As depicted in Figure 2.1, the final variable is group heterogeneity, and the constructs are: KSAs, age and ethnicity. These have been incorporated in the theoretical framework for the study, and the assumptions are that employee perceptions of justice would then trigger the application of group heterogeneity variables (ethnicity, age and KSAs) through employee discretions to affect the exploitation of social capital within the FPS.

Price and Mueller’s (1986) model was used to measure distributive justice. They developed this construct on a 7 point Likert scale, with the scores ranging from one (1) = strongly disagree to seven (7) = strongly agree. It was confirmed that previous studies that used this scale had a reliability of 0.90. For procedural justice, the construct used a scale developed by Niehoff and Moorman (1993). It is a 7 point scale and the scores on the scale range from one (1) = strongly disagree to seven (7) = strongly agree. Previous studies that used this scale also had a reliability of 0.90. Measuring interactional justice or measuring quality of interpersonal treatment received, the questionnaire used a 7 item scale developed by Moorman (1991), with the scale ranging from one (1) = strongly disagree to seven (7) = strongly agree. Previous studies that used this scale had a reliability of 0.98. Measuring employee performance proxies, employee discretions were used as the key measure, with the following subcategories as the specific measures within the constructs. These are pay, HR policies, turnover rate and tardiness. P a g e | 96

The above is in line with the findings by Bacharach (1991) when he asserted that theories cannot be tested without measurement. Simply put, measurement is the assignment of numerals to objects, events or concepts (Kerlinger & Lee, 2000). In this regard, measurement serves to operationalise the constructs in a theory so that they can be subjected to empirical testing. Indeed, Colquitt and Shaw (2001) confirmed this by theorising that even the most interesting and powerful theories are useless if their component constructs are poorly measured. This reinforces Zhang and Shaw’s (2012) theory on the importance of clarity in the measurement constructs.

Pilot Survey: A pre-analysis of likely demographics of the population in the FPS identified six basic types of network ties involved in HR selection into the FPS – three kinds of instrumental ties, formed through work, politics and school; and three kinds of affective ties, based on friendship, family relationship and community ties, such as being neighbours, and being members of the same villages, provinces or settlements. The FPS is also characterised by compositional heterogeneity, for example, gender analysis indicates that in the permanent establishment of the service 49 per cent are women and 51 per cent are men

(see Table 6.24).There is also wide variation on characteristics such as age and occupational groups (see Table 6.25). For example, the age distribution by gender analysis has indicated that for the 31 – 40 age group, 49 per cent are women and 51 per cent are men. Finally, the

FPS makes an appropriate sample for social capital research due to variation in members’ motives, particularly with regard to task orientation. On the one hand, the FPS provides an opportunity for individuals who wish to establish themselves and upskill and develop their competencies of public service skills in general.

P a g e | 97

3.7 Qualitative data analysis

Qualitative data analysis based on explanation building was implanted in the survey and as exemplified by Yin (1984), the objective was to analyse the case data by building an explanation about the case. Secondly, pattern matching was also employed to highlight emerging patterns as researcher link data to propositions. Further, explanation building was completed by time series analysis, which is an analytical framework that portrays a chronology, a frequently used technique in case studies. Yin (1984) asserted that for analysis of chronological events, the analytical goal is to compare the chronology with that predicted by some explanatory theory.

3.7.1 Qualitative data analysis methods

In this part of the analysis, the three dominant analytical techniques for the inductive approach were employed and they are: pattern matching, explanation building and time series analysis (Yin, 1984).

3.7.1.1 Pattern matching

In this framework, an empirical based pattern is compared with a predicted one. Even though linking the data to propositions and the criteria for interpreting the findings are the least developed aspects in case studies (Yin, 1984); Campbell (1975) described pattern matching as a useful technique for linking data to the propositions. He asserted that, “pattern matching is a situation where several pieces of information from the same case may be related to some theoretical proposition” (1975: p 182). Campbell (1975) coined the term ‘pattern identification’ as a characteristic of qualitative analysis which he identified as holistic (i.e. analysing the pattern) rather than atomistic (i.e. analysing its constituents). However, Yin

(1984) defined pattern matching as the most desirable analytical strategy in case study research, and discussed pattern matching in two perspectives. Firstly, it could be an analytical P a g e | 98

strategy to deduce the pattern in “a non-equivalent dependent variable design, in which the initially predicted value must be found for each element of a pattern of dependent variables”

(Yin, 1984). He also confirmed that pattern matching could be desirable in “a non-equivalent independent variable design” as well. In this scenario, the researcher should formulate different expected patterns of independent variables, each based on a different and mutually exclusive (rival) theory and the concern of the case study would be to determine which of the rival patterns has the largest overlap with the observed one. As depicted, the analysis exposed

‘pattern codes’ in the FPS study which exemplify an explanatory or inferential code, ones that identify an emergent theme, configuration or explanation (Yin, 2009, Miles &

Huberman, 1994).

3.7.1.2 Explanation building

For this framework and as exemplified by Yin (1984), the objective was to analyse the case data by building an explanation about the case. In this regard, the objective was to explain a phenomenon by stipulating a set of causal links about it. This is a preferred and relevant framework for analysing narratives. Using the explanation building model, a crucial question arose and that is: how did the causal links reflect critical insights into public policy processes? For example, the current system of governance for the FPS was formed in relation so many influences from the Fiji social context over the years; hence the iterative nature of explanation building is outlined in the analysis proper. In an earlier study, Yin contended that explanation building is an explanatory phenomenon consisting of: “(a) an accurate rendition of the facts of the case, (b) some consideration of alternative explanations of these facts, and

(c) a conclusion based on the single explanation that appears most congruent with the facts”

(1984: 24)

P a g e | 99

3.7.1.3 Time series analysis

Lastly, a time series framework using the Campbell (1969) model will help to lay the foundation for the conclusion of the case study analysis. Hence a chronology, which is frequently used in case studies and time series analysis, is used. Yin (1984) asserted that for analysis of chronological events, the analytical goal is to compare the chronology with that predicted by some explanatory theory.

3.8 Quantitative data analysis

For the quantitative analysis, a methodology was employed which will explore both a descriptive and an inferential analysis. The analysis began by examination of relevant summary statistics and a variety of graphical displays as descriptive analysis was employed.

This was conducted to provide simple summaries about the sample of the FPS survey and about the observations that have been made within the analysis. Further, factor analysis was employed with a view to reduce the complexities and unearth the key underlying themes that correlate the data distribution to the sub-research questions and central research question.

3.8.1 Survey data analysis

Acknowledging Landau and Everitt (2004), and under a quantitative framework, the exploratory factor analysis method was used to investigate the underlying structure of the association between the 10 items per questionnaire in the 10 questionnaire survey.

Specifically, the analysis investigated whether there is a relatively simple underlying structure that produces the observed associations among the 10 questionnaire items. The 10 questionnaires measure FPS workers; perception on networking ties and justice/injustice on

HR selection processes and policies. P a g e | 100

Further, descriptive and factor analysis were used to test the central research questions through the sub-research questions that drive the thesis statement of the study: “Social capital through network ties influence HR selection policies and decisions within the Fiji

Public Service”.

The semi-structured approach has been supported by Acquaah and Eshun who stated that measuring managerial networking relationships involves focusing on relational ties and trying to measure the soft nature of personal and social networking relationships, hence, it may be difficult using a structured objectivist approach (2010). They further stated that future researchers might take heed of the above point and use some unstructured subjective approach or a mixed method to probe the unknown. Further, the analysis of gathering information under a qualitative paradigm is an ongoing process as new ideas will emerge as the research begins to unfold. Hence, new developments will continually be noted and strategies reassessed to incorporate those new findings into the theory. Bryman (2008) reinforced the proposed method by stating that even though exponents of the case study research design often favour qualitative methods, such as participant observation and unstructured interviewing, researchers however use the case study methodology in both quantitative and qualitative research.

The adopted methodology was selected for the FPS study because it allows a large sample to be collected from different organisations, thereby enhancing the generalisability of the results. Further, the selection of the sample from different FPS ministries also ensures organisational sub-units are independent of each other, and therefore avoids the problem of the confounding effect of networking associated with the technical characteristics of HR selection systems of any particular institution.

P a g e | 101

3.8.2 Method 1: Descriptive Analysis

Identifying the social conditions of its use, this discourse acknowledges conventional theories on statistical analysis including Anbari (2002) in ascertaining that descriptive statistics helps us boil down raw statistics into an interpretable digestible form. There are several ways to do this, but the starting point is to count the number of times each score occurs, hence the number of times a score occurs in a set of data is the score’s frequency. The first step in trying to understand and analyse the raw data in the Fiji study survey was to bring to light two (2) important aspects buried within the raw data. These are: which scores occurred, and how often did they occur? These questions are answered simultaneously by looking at the frequency of each score, and this is always the first step in understanding score from a study.

Hereafter, it is only logical that firstly a descriptive analysis and frequency distribution of the study on the three (3) Fiji government ministries be presented.

3.8.3 Method 2: Factor Analysis

In further analysing the FPS survey data through a descriptive statistical lens, I will explore whether there is a relatively simple underlying structure that produces the observed associations among the 10 questionnaire items of the survey. In perusing the various analytical techniques, I confirmed use of exploratory factor analysis to investigate the underlying structure of the associations between the 10 questionnaire items. Basically, factor analysis is an approach to summarising and uncovering any pattern in a set of multivariate data, essentially by reducing the complexity of the data (Landau & Everitt, 2004). Landau and Everitt found that factor analysis is concerned with whether the covariances or correlations between a set of observed variables can be explained in terms of a smaller number of unobservable constructs known as latent variables or common factors. This finding confirms that the correlation between each pair of measured (manifest) variables arises because of their mutual associations with the common factors. P a g e | 102

In their study, Preacher and MacCallum (2003) found that exploratory factor analysis is a method of indentifying unobservable latent variables (LVs). LVs as opposed to observed variables are those that are not directly observed but rather, are inferred through a mathematical model from other variables that are observed and directly measured.

The two major stages in the application of factor analysis in the FPS survey study as confirmed by Landau & Everitt include:

“Determining the number of common factors needed to adequately describe the

correlation between the observed variables, and estimating how each factor is related

to each observed variable (i.e. estimating the factor loadings).

Specifically, exploratory factor analysis is concerned with whether the correlations

between a set of independent observed variables (X1, X2...... ,Xn) can be explained

in terms of a smaller number of unobservable latent variables or common factors (f1,

f2, ...... fk), where k < q and hopefully much less (Landau & Everitt, 2004).

Formal model linking manifest and latent variables is simply that of multiple

regression with each observed variable being regressed on the common factors. The

regression coefficients in the model are known as “factor loadings” and the random

error terms as “specific variates” since they now represent that part of an observed

variable not accounted for by the “common factors”.

Trying to simplify the initial solution by the process known as factor rotation”

(2004: p 288).

P a g e | 103

The measured variables and the common factors (LVs) are formally linked through multiple regressions with each observed variable being regressed on the common factors. In mathematical terms, the factor analysis can be written as follows:

= + …+ +

= + …+ +

= + …+ +

The equations above can be written more concisely as:

x = + u

Where

x = { , , f = , u=

(Landau & Everitt, 2004: p 287)

P a g e | 104

Further, the analytical procedures of factor analysis to be applied in this analysis are exemplified in Table 3.3.

Table 3.3 Analytical procedures for factor analysis

Stages Analytical Procedure Specifics and Status

A Grouping objects The survey instruments for the FPS study (questionnaire) has been subjected to respondents and raw data collected.

B Determining how many groups In factor analysis, it was the question of are needed how many factors are needed to explain the variability in a set of items

C Determining relationships The object is to derive a component between groups correlation matrix which shows the correlation between the extracted factors

D Outliers Factor analysis has variables with low communalities, low factor loadings and low correlation with other items

E Other forms of structure Factor analysis is specifically designed for pulling out latent factors or common factors that are assumed to have given rise to observed variables

(Source: Landau & Everitt, 2004)

In the ensuing analysis for the FPS survey, the distinction between principle component analysis (PCA) and exploratory factor analysis (EFA) was carefully scrutinised to ensure that the most conducive analytical method for 10 items on a 7 point Likert scale survey be employed. The utilities behind PCA and EFA lie in data reduction. However, the P a g e | 105

components in PCA are conceptually different from factors in EFA. PCA and EFA may seem superficially similar, but they are very different, and problems may arise when one attempts to use PCA as a substitute for approximation for EFA (Preacher & MacCallum, 2003).

Hence, in the FPS survey analysis, EFA was chosen ahead of PCA in view of Preacher and

MacCallum’s (2003) findings because PCA does not explicitly model error variance which renders substantive interpretation of components problematic. They also found that EFA is not intended to explain optimal amounts of variance; however, the EFA’s success is gauged by how well the model helps the researcher understand the sources of common variation underlying observed data. Hence, for the FPS survey, the EFA will help the researcher to understand the sources of common variation within the data and the rationale behind such variation.

Another key aspect of EFA is deciding on the number of factors to retain. The objective of this key aspect is to identify the most useful component or factors that account for more variance than could be expected by chance (Preacher & MacCallum, 2003:p 21). In other words, these factors are underlying themes that summarise the FPS employees’ perceptions of network ties and justice. Determining the most effective analytical technique for retaining factors, conventional theorists including Preacher and MacCallum (2003), ascertained that because there is no single, fail-safe criterion to use in arriving at this crucial decision, researchers usually rely on various psychometric criteria and rules of thumb.

However, to derive the most suitable methodology for retaining factors to be considered as the summarising lynchpins for the FPS perceptions as per the survey, several rules of thumbs have to be analysed. These included the Kaiser-Guttman criterion and

Gorsuch’s (1983) subjective scree test. Section 3.8.3 summarises the methodology employed to decide the most obvious choice.

P a g e | 106

3.8.4 Rules of thumb for retaining factors

3.8.4.1 Rule of Thumb 1 –Kaiser-Guttman Criterion

In his study, Guttman (1954) developed the ‘weakest lower bound’ for the number of factors, and ascertained that if the common factor model holds exactly in the population, then the number of eigenvalues of the unreduced population correlation matrix that are greater than

1.0 will be a lower bound for the number of factors. In other words and as shown in Figure

3.2, the first rule of thumb is to retain as many factors (or components) as there are eigenvalues of the unreduced sample correlation matrix greater than 1.0.

Figure 3.2 – Scree plot

However, Preacher and MacCallum (2003) objected, not to the validity of the weakest lower bound, but to its application in empirical studies. Firstly, they found that the weakest lower bound applies to the population correlation matrix and assumes that the model holds exactly in the population with m factors. However, in practice, the population correlation P a g e | 107

matrix is not available and the model will not hold exactly. Hence, application of the rule to a sample correlation matrix under conditions of imperfect model fit represents circumstances under which the theoretical foundations of the rule are no longer applicable. Another objection by Preacher and MacCallum (2003) for the first rule of thumb was that the Kaiser-

Guttman criterion is appropriately applied to eigenvalues of the unreduced correlation matrix rather than to those of the reduced correlation matrix. Hence, in practice, the criterion is often misapplied to eigenvalues of a reduced correlation matrix.

An additional objection was made by Gorsuch (1983) when he found that many researchers interpret the Kaiser criterion as the actual number of factors to retain, rather than as a lower bound for the number of factors.

The objections by Preacher and MacCullum (2003) and Gorsuch (1983) were taken on board for the FPS survey factor analysis; hence the Kaiser criterion was not used in the survey.

3.8.4.2 Rule of Thumb 2 – Gorsuch’s subjective Scree Test

Gorsuch’s model, as shown in Figure 3.3, ascertains the retention of many factors as there are eignevalues that fall “before the last large drop on the scree plot. A scree plot is a scatter plot of eignenvalues plotted against their ranks in terms of magnitude” (Preacher & MacCallum,

2003:p 23). The scree test can be applied to either a reduced or unreduced correlation matrix; hence the methodology is conducive to the FPS survey in view of the EFA objective in this survey for a reduced correlation. Further, several studies also found the scree test to result in accurate determination of the number of factors most of the time (Cattell & Vogelmann,

1977; Tzeng, 1992).

P a g e | 108

Figure 3.3 – Scree plot depicting Gorsuch’s (1983) model

(Preacher & MacCullum, 2003)

3.8.5 Rotation

In their study, Preacher and MacCallum (2003) ascertained that a third important decision in factor analysis involves rotation. In his studies, Thurston (1935, 1947) contributed to the factor analysis methodology discipline by recognising that to be interpreted meaningfully factor solutions should be rotated to reflect what he called “simple structure”. Preacher and

MacCallum (2003) ascertained that rotation methods are designed to find an easily interpretable solution from among an infinitely large set of alternatives by finding a solution that exhibits the best simple structure.

Preacher and MacCallum (2003) further confirmed that many rotation methods have been developed over the years, some proving more successful than others. They found that orthogonal rotation methods restrict the factors to be uncorrelated, while oblique methods P a g e | 109

make no such restriction, allowing correlated factors. Kaiser (1958) complemented this finding when he found that one of the most often used orthogonal rotation methods is

Varimax.

However, in examining Tom Swift’s data, Armstrong (1967) highlighted that Swift suspected that the factors underlying the 11 measured variables (MVs) would be statistically independent of each other, and because of this, he used orthogonal rotation. However,

Preacher and MacCallum (2003) objected to this finding and confirmed that this may sound like a reasonable approach, however, it is a strategy that is very difficult to justify in most cases. In other words, Preacher and MacCallum (2003) found that in using orthogonal rotation, the Tom Swift experiment introduced a fundamental contradiction between method and the underlying structure of the data.

Hence, if the researcher does not know how the factors are related to each other, there is no reason to assume that they are completely independent. In view of this issue, Preacher and MacCullum (2003) advised that it is almost always safer to assume that there is no perfect independence, and to use oblique instead of orthogonal rotation. They confirmed

Floyd and Widaman’s (1995) finding that if an optimal simple structure is exhibited by orthogonal factors, an obliquely rotated factor solution will resemble an orthogonal one anyway. In another words, nothing is lost by using oblique rotation. It also offers a further advantage of allowing estimation of factor correlations which is surely a more informative approach than assuming that the factors are completely independent. This is the theoretical basis used in choosing oblique rotation as the most appropriate rotation method for the FPS survey.

P a g e | 110

3.8.5.1 Retaining factor loadings greater than an arbitrary threshold

In their study, Preacher and MacCallum (2003) also reported that it is advisable to report all obtained factor loadings so that readers can make judgements for themselves regarding which loadings are high and which are low. This corresponds with Floyd and Widaman’s (1995) finding that it is advisable to report factor loadings for all variables on all factors, i.e. the research should have been interested in the complete pattern of loadings (low loadings as well as mid-range), and not simply the ones arbitrarily defined as large.

3.9 Ethical considerations

The questionnaires and other intended research instruments were prepared after being thoroughly discussed with my supervisors for review and comments. After minor modifications as per their comments, the questionnaire was then sent to the Human Research

Ethics Advisory Panel of UNSW@ADFA for final assessment and approval of the research.

After another minor modification, the Human Research Ethics Advisory Panel of

UNSW@ADFA endorsed the research instruments in view of the intended research for the

FPS. Their endorsement of the research project reads:

Regarding your project “Human Resource Selection Practices in Fiji's Public

Services”; At its meeting of 10th December, 2010 UNSW@ADFA Human Research

Ethics Advisory Panel was satisfied that this project, is of minimal ethical impact and

meets the requirements as set out in the National Statement on Ethical Conduct in

Human Research. Having taken into account the advice of the Panel, the Deputy

Vice-Chancellor (Research) has approved the project to proceed. The approval

number will be A-10-55, and is valid for twelve months from the date of that meeting.

(Human Research Ethics Advisory Panel of UNSW@ADFA, 2010). P a g e | 111

3.10 Conclusions

As shown in this chapter, the investigation of a single case in the thesis has been purposively designed to accommodate a multiple-level unit of analysis. The embedded constructs within the case as evident in the FPS study depicted a structured survey and case study analysis using both quantitative and qualitative analytical framework. The case study evidence for the

FPS was subjected to multiple types of data collection techniques. In addition to including both qualitative and quantitative data, the evidence covered two levels of analysis: a main unit (the case) and a sub-unit (embedded within the overall case) (Yin, 2004). In this regard, the case itself (FPS) has provided the setting, but within that setting several instances of the phenomenon are present depicting the three FPS institutions examined in the FPS case study.

Finally, and as detailed earlier, the analysis of the raw data involved a mixed methodology approach. For qualitative analysis, it was deemed that an explanatory effects matrix and time series analysis were the best methods considering the character and elements of the Fiji context’s archival data and narrative. On the other hand, descriptive and factor analysis techniques were employed for the quantitative analysis. These were also carefully selected in view of the survey instruments for gathering raw data and the character of questionnaires employed.

P a g e | 112

CHAPTER 4

THE HISTORICAL, POLITICAL AND SOCIAL CONTEXT OF HUMAN RESOURCES SELECTION DECISIONS IN THE FIJI PUBLIC SERVICE

P a g e | 113

Table of Contents

4.1 Introduction

4.2 Historical background of labour (1643 – 1987)

4.3 The role of chiefs in the Fiji Public Service

4.4 Indian immigrants and indenture

4.4.1 End of indenture

4.5 Fiji Independence 1970

4.6 The political context of employment (1987 – 2007)

4.7 The political context and HRM challenges in the Fiji Public Service

4.8 The legislative charter of Fiji and its autonomy

4.9 The role of the Republic of Fiji Military Forces in Fiji’s political context

4.10 Demographic structure of Fiji in 2011 (base year)

4.11 Social networking

4.11.1 Networking ties in Indian social structure

4.11.2 Networking ties in Fijian social structure

4.12 Demographic structure of the Fiji Public Service – 2011

4.13 Conclusion

P a g e | 114

4.1 Introduction

This chapter has been crafted to highlight substantive contributions from the historical, political and social context of HR selection decisions in Fiji. In the process, it has provided literature and evidence to help the reader understand underlying issues behind the central research question of the FPS study:

How does social capital through network ties influence human resource management

selection decisions and policies in the Fiji Public Service?

Specifically, the chapter describes the historical, political and social context in which the FPS operates. It broadly examines the history of labour in Fiji, and the role of the chief system in the development of the FPS. It also outlines the demographics of the population and the FPS in 2011, the base year of the study. Further, it presents the vital streams that encapsulate the network ties of the Fijian and Indian social structures.

In many developing countries firmly bonded through collectivism and networking, like Fiji, the use of social networking can create real or perceived injustices in the operation and management of their bureaucracies, or they can be managed to induce the required efficiency levels. These perceived injustices can escalate to damage relationships between employees and also between employees and employer, hence affecting worker perceptions of the fairness of HR selection decisions. In this regard, social networks may be perceived to either a potential source of capital or could incur unnecessary costs in the administration of

HR selection policies.

This chapter articulates the theoretical space as exemplified by the literature which had substantive relevance to the application and theoretical contribution of HR selection decisions and social capital. Specifically, the study investigates how social network ties impacted HRM policy and practice in the FPS. Of importance are the justice perceptions of P a g e | 115

HR selection decisions of those FPS employees within the case sample institutions. A key substantive outcome of this chapter is that it provides evidence of social capital in the Fiji context. Further, it presents a narrative that exemplifies the formation of social capital for the two major ethnic groups in Fiji. Most importantly it shows the influence of social capital and networking on the FPS and its HR selection decisions.

4.2 Historical background of labour and networking in Fiji

Fiji’s labour background beginning in 1643 is described against a social, economic and political backdrop in this part of the chapter. The Fiji group was sighted by Abel Tasman in

1643 and subsequently visited by Captain James Cook in 1774 (Burns, 1963). The sightings and visitations by these great explorers were crucial for the historical background of labour in

Fiji because they laid the foundations for further visitations from capitalist planters and entrepreneurs. The small number who initially came to settle had their individual network ties in terms of religion, politics, schooling, family, friendship and community (Burns, 1963). The

Methodist missionaries arrived in 1835, followed by their Catholic counterparts in1844. The early missionaries initially set up churches and schools for the planters and entrepreneurs, and later for the wider social environment, which met the demand of their religion and schooling network needs.

The arrival of the early missionaries influenced key social indigenous institutions in

Fiji; the most significant was when Ratu Seru Cakobau, the supreme war lord of Bau and the

Kubuna Confederacy, renounced his ancient gods to assume the Christian faith. The Kubuna

Confederacy consists of a network of leading provinces which have been loyal to the supreme war lord of Bau (Baleiwaqa, 2005). Baleiwaqa (2005) suggested that the early missionaries, including Tongan allies with the help of the white settlers, sided with Ratu Seru Cakobau and P a g e | 116

overcame a strong resistance movement in South East Vitilevu in the Battle of Kaba. The movement wanted to topple Ratu Seru Cakobau and his allies, since they did not want to renounce their ancient gods even though they were traditionally obliged to respect the supreme war lord of Bau (Baleiwaqa, 2005). The outcome of this important conflict had a crucial implication in the history and formation of the colony of Fiji because it led to the introduction and nationalisation of the Christian faith and the introduction of schools and educational institutions initiating preparation of the labour force in early Fiji.

On their arrival in Fiji, planters, entrepreneurs and missionaries took heed of the need to formalise their existence, hence the need to establish networks to protect their interests, assets and polity from the wider Fijian population. Those with connections to Great Britain claimed Captain James Cook’s discovery and made representation to the British crown to formalise that discovery and make Viti (Fiji) a colony of Great Britain. However, they needed to befriend the most powerful war lord in Fiji and his allies. When the protocol was taken care of, the British Consul was established in Levuka town in 1857 (Burns, 1963). In this set up, the legal and political network was formed with a view to formalising the religious and school networks in protecting the interests of the planters, entrepreneurs, the government and certain high chiefs in Fiji.

4.3 The role of chiefs and the Fiji Public Service

The role of chiefs within the FPS was initially and formally recognised as a key element within the instruments of cession on 10 October 1874:

That, pending the making by Her Majesty as aforesaid, of some more permanent

provision for the Government of the said islands, His Excellency Sir Hercules George

Robert Robinson, in pursuance of the powers in him vested, and with the consent and P a g e | 117

at the request of the said Tui Viti and other high Chiefs, the ceding parties hereto,

shall establish such temporary or provisional Government as to him shall seem meet.

(Fiji Government Gazette, No. 1, 1874)

The above confirmed that the Fijian chiefs’ consent and request enabled the setting up of a provisional government in Fiji in 1874, to formally await Queen Victoria’s acceptance of the deed of cession. The provisional government in Fiji adopted the statutes and regulations of the Australian state of New South Wales and its governor, Sir Hercules George Robert

Robinson, was also made the provisional governor of the Colony of Fiji in 1874 (Fiji

Government Gazette, No. 3, 1874). However, on Queen Victoria’s acceptance of the deed of cession, and with the support of the high chiefs of Fiji, the formal administration of the

Colony of Fiji began on 1 September, 1875 (Fiji Government Gazette, No. 1, 1875). Hence, the role of chiefs within the FPS began with the provisional and formal administration of the

Colony of Fiji.

In an interview with a former District Officer in the Fiji colonial administration, Mr

Raymond Baker in 2011, confirmed that the selection of indigenous Fijian candidates to the administrative officer grades of Her Majesty Overseas Cadet Service (HMOCS) were subject to the colonising regime and, where practicable, it had long been the practice for Britain to govern its overseas territories through the existing chiefly hierarchy. Many prominent chiefly figures were given positions of responsibility in the provincial administration, which ran in parallel to the United Kingdom (UK) colonial administration. Burns supported this theory when he stated that the British colonial administration did not have any other alternative but to approach the chiefs of Fiji to be part of the colonial administration (1963). This is also evident in the promulgation of the Fiji Government Gazette No. 1, 1874, where 12 Fijian chiefs were appointed to the FPS to help the Fiji colonial administrators manage the affairs of

Fiji, its citizens and particularly the operations of the FPS. Burns (1963) confirmed the above, P a g e | 118

stating that Sir Hercules Robinson established the interim government in Fiji after the cession after he had organised 12 of the leading chiefs as Rokos (provincial administrators), each in charge of a provincial area (see Appendix 9.7). In that context, the initial network ties that controlled entry into the FPS though the role of chiefs were initiated. The network context was characterised by a combination of British colonialists, chiefly families and other prominent businessmen and women who had strong social, political and economic undertakings in the administration of Fiji.

4.4 Indian immigrants and indenture

In 1875, Governor of the colony of Fiji, Sir Arthur Gordon, pointed out to the white settlers that the supply of Polynesian labour was decreasing and its cost increasing, and suggested that it would be wise to supplement it with Indian coolies (Burns, 1963). After reviewing the proposal in 1877, and after long negotiations with the Government of India, a scheme of indentured labour was agreed. Burns (1963) said that the principal feature of the scheme was that volunteers were recruited by agents in India and, after signing contracts, were transported to Fiji at the expense of the employers to whom they were later assigned. The contract in each case was with the Fiji Government and the labourer bound himself to work where he was directed for a term of five years in return for agreed wages and accommodation. Recruiters in

India drew on the urban migration of people displaced by the commercialisation of agriculture (Morton, 1944). Morton found that in getting workers from India, recruiters used deception and intimidation to secure contracts with individuals of diverse castes. The first immigrant ship arrived from India in May 1879, with 481 labourers on board, and between that date and 1916, when the indentured system was abolished; some 62,000 Indians were taken to Fiji under indenture (Burns, 1963). It is important for this thesis to state that the indentured Indian labours were introduced to Fiji in view of a British colonial government P a g e | 119

policy that the appointed governor and senior administrators in the colonised nations and territories utilise local resources to amass the required annual revenue for the operation of the colonised territory. Hence the colony of Fiji organised and took Indian workers as indentured labourers to work on the sugar plantations so as to boost the state coffers.

Further, Indians were also introduced to Fiji as an essential part of the colonial rulers’ programme to limit the capitalist exploitation of Fijian labour, and brought in an individualistic system of living and way of life. Fijians were not used to living within an individualistic mindset, since their society was based on collectivism through a communal system of living. To preserve the social structure and to protect indigenous Fijians from exploitation by planters, the colonial government decided that the indentured labourers needed to work on European planters’ sugarcane farms which were crucial for economic development (Norton, 1977).

When the colonial government in Fiji supported the European planters’ proposal to plant sugarcane as a commodity that could stabilise the state’s coffers, local Fijians were initially trained through the subordination to their respective chiefs to work on these farms

(Norton, 1977). Norton said that aligning themselves closely to the collectivist model of a sustenance economy within the social structure, these local labourers would often run away from these farms to their respective villages, hence disrupting the operations of the sugarcane farms. These irregularities in farm labour operations, and the complaints by the European planters, set the precedence that left the colonial government in Fiji with no other alternative but to indenture the labour needed to not only maintain, but also sustain the sugar industry in

Fiji.

The European farm owners and the Colonial Sugar Refining (CSR) Company also noticed that the Indian farm labourers had a different work ethic which was conducive to maintaining productivity levels (Burns, 1963). This work ethic was also noticed and P a g e | 120

embraced by the Fiji colonial government which then churned out policies to accommodate the labourers’ wish to staying on in Fiji as planters. These policies include the granting of land leases, granting of marriage laws – for example, a 12-year-old Indian girl could marry a

16-year-old Indian boy – and also granting of marriage licences to Indian priests to conduct marriages according to their own culture (Fiji Blue Book: (1891 – 1937 ). The timing and introduction of this new labour force was also vital in the economic development and survival of the Fiji colonial government, where sugar production was formalised as the mainstay of the economy. Hence, the government collaborated with the CSR Company and sought ways to increase output, given the constraints at hand (Norton, 1977).

4.4.1 End of indenture

The indenture system was halted in 1916 in view of public outrage in the UK and other leading economies of the world. This led to its cancellation and ultimately abolishment on 1

January 1920 (Burns, 1963). After the formal abolition of indenture, small groups of free migrants from Gujarat and Punjab joined the Indian community (Norton, 1977). In view of the economic and trading opportunities provided by the colonial government, free migrants took advantage of the opportunities which added a new dimension to the Indian social structure and system. This added dimension was significant in the political and socio- economic development of the colony of Fiji. Due to their economic prowess, the free migrants established themselves in small to medium business entities, dominated the sugar industry and established networks. They also asserted themselves as leaders of the Indian community because of their economic and social standing within the network, which was acknowledged by the wider Indian community especially the lower castes, who were now free indentured labourers (Scarr, 1984).

It was deemed by all entrepreneurial stakeholders in Fiji at that time that the prosperity of the Fiji colony depended on the collaboration between European capital, Indian P a g e | 121

labour and Fijian land (Norton, 1977). CSR, with the backing of the Fiji Government, sought to increase output by turning its plantation labourers into individual cultivators working on their own 10-acre leases of company land with the assistance of their wives and children.

Norton (1977) further ascertained that because the company estates were inadequate to absorb the growing Indian population and to fulfil Fiji’s potential for sugar production, CSR with the support of the government encouraged Indian farmers to acquire leases of Fijian land on a 30-year contract. Burns (1963) also found that when the indenture system came to an end, it became difficult for CSR to obtain labourers to work on its fields, especially as many of the Indians whose indentures had expired were growing cane and other crops as independent farmers on lands leased from Fijians. Hence, the company decided to give up direct cultivation of their land and to subdivide it into farms of about 10 acres to lease to tenant farmers. It was considered that a farm of this size would allow the tenant to make a living for himself and his family by growing cane, and vegetables and keeping a few farm animals (Burns, 1963). Most of these tenants were Indians and today most of the cane crops are produced by Indian farmers, even though the number of Fijian farmers has been on the rise since the expiry of the land leases initially issued to most Indian farmers.

The significant changes in the social structure of the Indian population transformed their social and economic status from that of a labourer and employee to a producer and owner with a direct and personal interest on his farm (Merrett, 2007). Hence, Indian labour became the cornerstone of the Fijian sugar industry and the mainstay of the Fijian economy.

The end of the indenture system had important implications for the Fiji social context.

In his study, Merrett (2007) contended that the large Indian community that remained were composed of tenant farmers, who sold their cane to a company that pressed for the lowest price. One commentator noted: “from a tenant's perspective, all this was indenture in disguise, CSR’s method of shifting cultivation risks while simultaneously ensuring that cane P a g e | 122

could be bought cheaper than it could be grown on company plantations” (Merrett, 2007:p

248). To safeguard their interests and assets, the Indian farmers formed unions, but in their negotiations with the company, CSR had the upper hand, leading it to insist on terms that occasionally prompted strikes by the tenants (Merrett, 2007). The union ideology had been the key cornerstone that Indians relied on to fight for their rights in Fiji, not only as citizens but, importantly, as planters of the key commodity that had been stabilising the economy. In the process, the Indian population’s social capital in Fiji was again reinforced through unionism which is generally tasked to negotiate wage rates and conditions of employment for its members (Rawson, 1986).

4.5 Fiji Independence – 1970

From the end of indenture to a decade before independence, the Indian population was of the view that because they had been the mainstay of the economy, the colonial government of

Fiji needed to recognise their contribution and grant them civil liberties. Coleman (1988) finding has affirmed that the theory of rationale is critical in understanding social capital.

Specifically, the theory outlines a scenario where each actor has control over certain resources and interests in those resources deduced some sort of social capital depending on the magnitude of the interests of the stakeholder. In the Fiji context, land in Fiji is a resource where the two main actors (Fijians and Indians) have control and interest in; Fijians have almost absolute control through hereditary, whilst Indians have unwavering interests.

Discerning their shaky social status, Indians in Fiji used the sugar cane farmers’ union and other subsidiary network institutions within their social structure to support their plea and pressure the government for equal parliamentary rights. Their prime objective was to agitate constitutional and political reforms that would reinforce the legitimacy of their status in the social structure. The early history of pressure for constitutional reform consisted principally P a g e | 123

of calls by the Indian community for a common roll method of election and by the European community for an unofficial majority (Mara, 1997). In his memoir, Ratu Sir Kamisese Mara

(1997) highlighted that the next political development of interest was the formation of the

Fijian Association to champion the rights of Fijians. Lasaqa (1984) said that in 1963, Fijian leaders formed the Fijian Association which aimed to protect Fijian interests which were not to be comprised for short-term political gain. Primarily, this action was initiated to counter the combined monetary power of the Indian population which the colonial government relied on.

In view of the ever-increasing pressure from the two major ethnic groupings in Fiji for social and political prominence, the British Government, after a constitutional conference held in London in 1970, devised a constitutional reform reconstructing the seating arrangement of the Legislative Council of Fiji (Lasaqa, 1984). All stakeholders, particularly the Indian population of Fiji, perceived this conference as a triumph for Fijian interests because the common roll was rejected (Norton, 1977). The key decision from this conference also determined that Fijian chiefly institutions through the Bose Levu Vakaturaga (Great

Council of Chiefs (GCC)) retained the right to elect two more representatives into the

Legislative Council, further strengthening indigenous Fijian institutions and their aspirations.

However, this result reinforced affective and instrumental ties within the two main ethnic groupings in Fiji and propelled the emergence of party politics where the Indian population rallied behind the National Federation Party while the Fijian population, with the support of other minority grouping, rallied behind the Alliance Party.

Further, “the pressure from the United Nations to colonial masters, especially from its committee on colonialism to grant independence to colonial territories had been mounting for some years” (Lasaqa (1983). He also asserted that this pressure led to colonial powers taking the view that it was better to grant independence too early rather than too late. This meant P a g e | 124

progress towards self-government for Fiji was inevitable and on a Legislative Assembly representation followed by the constitutional conference held in May 1970, the colony of Fiji gained independence from Britain on 10 October, 1970, after nearly a century as a British colony.

4.6 The political context of employment

A prime consequence for Fiji in gaining independence from Britain was the uncontrollable state of political bickering for the two major ethnic groups. In that context, Putman’s argument presents “social capital essentially as the amount of trust available and as the main stock characterizing the political culture of modern societies” (1996a:47). He confirmed that voluntary associations enable a horizontal linking of people produce trust, the norm that causes interpersonal bonding. Putman (1996a) also stressed that social capital is elevated from a feature of individuals and small groups in local communities to a feature of large population aggregates; hence social capital becomes a diagnostic tool for societal political and economic health. As the aspirations of the two dominant ethnic groups and their social structure began to escalate for prominence and control in securing jobs and status within Fiji, the repercussions also infected the FPS. In the process, the influence exerted through the social and political arena in Fiji had the potential to influence HR selection decisions in the

FPS, as elected members of the major ethnic groupings fought for the democratic rights of their electorates.

A significant challenge for FPS managers was political instability, which led to an era of uncertainty in Fiji’s social and economic development. Hence, the reform programmes for

FPS institutions were greatly hampered throughout much of the reform period and particularly the 1987 – 2010 period (Reddy et al. 2004). In 1987, the democratically elected P a g e | 125

government was overthrown by Fiji’s armed forces. The military coup was executed by Fijian extremists on behalf of the Fijian social system and its chiefs, since it was generally considered ‘taboo’ to challenge the leadership and authority of such institutions. The coup d'état of 1987 was generally perceived as a return of political control to the indigenous population as the newly elected government was assessed by perpetrators as being dominated by those of Indian heritage (Sharpham, 2000). The perpetrators of the coup and their allies accused the government of being controlled by networks of the Indo Fijian population for their self-interest and control of the political context of employment. However, this contradicted the findings by Nanda (2001) who confirmed that when the Labour Party

Government was overthrown in 1987, key positions in government were held by Fijians.

However, indigenous Fijian supremacy was an excuse, as the same “propaganda [was] used by Europeans for their own political purposes in the colonial era” (Lawson, 1990). This meant that, although senior positions in government were held by indigenous Fijians, the majority of the government caucus were Indians. Hence, the political context of employment was viewed to be controlled by Indians.

The perceptions for control of the political context of employment in Fiji was also a major determinant for the coup d'état in 2000 and 2006. In May 2000, there was another illegal takeover of the democratically elected government led by Mahendra Chaudary and dominated by Indians. The election result a year earlier and Chaudhry's subsequent appointment as Prime Minister had angered hard-line Fijian nationalists, and the same

‘propaganda cause’ which initiated the 1987 coup d'état was also perceived to be the major determinant this time. The general perceptions of Fijians were that whoever was managing government had the political context of employment in Fiji.

The coup d'état of 2006 had an unusual ‘propaganda cause’, which disrupted the societal and political norms in Fiji. Specifically, Prime Minster Qarase’s government and the P a g e | 126

military started a public fight over government policies and bills, which resulted in the

December 2006 takeover (Ramesh, 2007). Commodore Frank Bainimarama, the Commander of the Republic of Fiji Military Forces (RFMF), overthrew the democratically elected government led by Prime Minister Laisenia Qarase, which he accused of being inefficient, ineffective, unproductive, pro-indigenous Fijians/I Taukei and racist. The military regime under Commodore Bainimarama came to prominence and was backed by the Military

Council who claimed that the democratically elected government was supporting the aspirations of the indigenous Fijians or Kaiviti. The assumption was that the political context of employment was with the Fijian community under a Fijian-backed government. The political instability beginning in December 2006 in Fiji was an outcome and a prime example of when the political context of the employment of government is infiltrated. The vehicle for infiltration was intense competition for social status, power and recognition by the different social groupings and institutions in Fiji. The fierce rivalry exacerbated and snowballed to an uncontrollable extent and was evident in 2006 when the coup d'état perpetrators emerged from within a key institution whose role was to manage the external security situations in Fiji.

The coup d'état perpetrators were senior officers of the RFMF led by its commander, Frank

Bainimarama.

In essence, the unstable and fractured political and social economy in Fiji provided the environment within which differing self-interests propagated. These divergent self- interests activated the established networks to affect the basis of entry into the labour market, particularly the FPS probing gains, advantages and protection for the respective major social groupings. In the process, senior public servants, policymakers and key public figures took advantage of the political and social processes to aggregate the required government policy support for their respective networks. The multiplier effect of this social network agitation P a g e | 127

posed a limitation on the effectiveness of public service delivery, which has been challenging

FPS managers ever since independence in 1970.

In Fiji, economic policy is largely driven by the various stakeholders within the social environment, hence moving the economy in a particular direction requires a large degree of consensus among the stakeholders, particularly the major social groupings. In essence, multiple networking originally formed the basis of the major social groups, entwining the very core of their existence. In his study, Putman (1996a) confirmed that the lack of strong family and community ties has significance bearing on intergenerational decline of social capital in the United States of America. This finding is quite significant for the Fiji context where the strong ethnic basis for Fijians and Indians has led to a solid social capital base that provides the platform for networks to blossom continuously. These networks include religious, political, school, family, friendship and community groups. The military coup d'état was classified as a consequence of conflicts primarily between the two main ethnic groups in Fiji. Sutherland stated that the military coup in Fiji was also attributed to a class struggle between chiefs, tribes, eastern and western indigenous Fijians, and urban and rural folk (1992). However, he failed to ascertain that it was the discernment of future insecurity and mismanagement of resources, particularly land from Fijians, as being the key determinant of the illegal takeover of the democratically elected governments in 1987 and 2000. Further, the perceptions of shift in belief in the political context of employment was also a key determinant of the coup d'état. The belief of the majority of the Fijian people was that their aspirations and the security that reinforced the framework of their respected institutions were now challenged, and this was considered a ‘taboo’ and an insult to their existence and credibility. Because of this, the perpetrators of the military coups of 1987 and 2000 were backed by prominent Fijian chiefs, institutions like the GCC, the 14 provinces (yasana) and their respective sub-units (tikina). It was also a struggle for ownership of resources, P a g e | 128

especially land, as most Indian farmers’ tenancy contracts (99-year leases) were coming to an end. Hence, the survival and livelihood of both dominant ethnic groupings depended on the mercy of the government of the day. In view of this, the perception of Indians was that any

Indian dominated political party, for instance the Labour Party under Mahendra Chaudary, would come to their rescue. The same theory could also be applied to the perceptions of

Fijians in view of their status, security and belongingness.

Finally, the influences on key institutions prominent in the formation of governments in Fiji are critical in shaping the social order and political context of employment. For example, the chiefly system and its associated institutions have been influenced by key immigrant groupings or their friends and allies. Burns (1963) said that the influence of

Charles Savage on the future of Fiji was considerable. He was the first to use a firearm in tribal warfare and was the leader of the small band of whites who increased the military power of Ratu Seru Cakobau, the supreme war lord of Bau, and made him the most powerful ruler in Fiji. Further, the Alliance Party was formed when Ratu Sir Kamisese Mara and other prominent Fiji chiefs colluded with prominent Fiji businessman and women to support the

Alliance Party financially (Mara, 1997). The political context of employment in Fiji resurfaced when Mr Mahendra Chaudary formed government through the Fiji Labour Party after colluding with some prominent Fijian and Indian key institutions and personalities.

Later, in 2006, it resurfaced again only in a different form, as the well-established Fijian or

Itaukei institutions and the traditional belief on the political context of employment was publicly challenged through the coup d'état manifesto of Commander Frank Bainimarama.

4.7 Political context and HRM challenges in the Fiji Public Service

Conspiracy theory has suggested that, in the Fiji context, collusion between the government of the day and prominent Fiji businessman and women induced a political context of P a g e | 129

employment that automatically favoured the businessman and women who were the major employers, particularly in the private sector. This was evident when a new government came in and political appointments were made, especially to senior and middle level designations.

By convention, different social values have impacted different HR systems in different countries. A system where political responsiveness and personal rule are prevalent tends to be characterised by political appointments. In Fiji, the Prime Minister and leader of government influences HR selection in the appointment of senior managers in government and key civil servant designation (see Appendix 9.2). He also influences key appointments for most government-controlled corporatised institutions. Even though sceptics may classify such management techniques as undemocratic, unconstitutional or even illegal, it is obvious that he employs such techniques to protect his cause and justify the management of state institutions. .

4.8 The legislative charter of Fiji and its autonomy

Firstly, the thesis needs to highlight and examine the character of the legislative charter of

Fiji under which the current military government in Fiji and the FPS are operating. This is important because the reliability of the legislative context will provide the rigour that reinforces public service effectiveness and efficiency, justifying policies that the military regime has chosen after seven years in office. It is assumed that this robustness will induce the required level of stability that is politically, socially and legally correct for the function of the FPS.

In April 2009, the 1997 Amendment Constitution of Fiji was supposedly abrogated by the then President of Fiji, Ratu Josefa Iloilo. However, sceptics asserted that under the same set of rules that define Fiji’s system of government and describe its institutional structure, he P a g e | 130

did not have the power to abrogate those fundamental principles and established precedents according to which the Republic of Fiji is governed. Hence, his actions to de-constitute the rules and regulations could be legally challenged. This thesis has maintained that if the

President of the Republic of Fiji was not legally empowered to de-constitute the supreme law of the land, then that may tarnish and challenge the astute and political, social and legal correctness for those established legal precedents and subsidiary legislation by which the current regime operates. There is obviously a legal connectivity between a constitution which is the supreme law of the land and its subsidiary legislation which translates the major legal framework into applicable rules and regulations. Hence, if the supreme law of the land is technically illegal, then the subsidiary legislation may also be deemed illegal.

The subsidiary legislation (Public Service Act of 1999) covers the functions of the

FPS. Hence, is the FPS constitutionally correct and valid, given that the legislative charter from which the current regime derives its authority to act from (the 1997 Amendment

Constitution of Fiji) was abrogated? In essence, to abrogate means to abolish or annul, hence the then President of Fiji used his statutory powers and authority under the 1997 Amendment

Constitution of Fiji to dismantle it, beginning with the preamble of the same legislative charter, which outlines its purpose, aims and justification. However, to abrogate requires a constitutional, pre-condition which is to have a full parliament, and the current regime doesn’t have a full democratically elected parliament hence, the current legal fraternity has resorted to relying on precedence, hence one may only read Government of the Republic of

Fiji Decree No. 22. (www.psc.gov.fj). This substantive finding also has the potential to influence the HR selection decisions of the FPS, as the policies could be deliberately tampered with by the key actors within the social system to attain their course and objectives.

Another key challenge for FPS managers is the continuous interruption of the autonomous element of the judiciary of Fiji, which impacted the consistency and stability of P a g e | 131

the application of Public Service Act of 1999. In view of this, the independence of the key legal institution that managed and asserted formality and control in the operations of public and private sector institutions in Fiji has been continuously tampered with. The judiciary of

Fiji oversees the legislative charter that established the character of the Fiji government and it’s Constitution. Since the coup era, beginning in 1987, the continuous interference has created fear in the minds of investors and workers about the validity of the Constitution with respect to enforcement of rights (Reddy et al. 2004). This is also a major substantive outcome which influences HR selection decisions within the FPSC.

4.9 The role of the Republic of Fiji Military Forces in Fiji’s political context

It is also envisaged that the role of the RFMF be critically examined in this thesis because of its repercussions in influencing HR selection decisions of government. Hence, the narrative is substantive because it covers a key gap in the current literature in view of the fact that the current role of the RFMF in Fiji’s social system and in influencing HR selection decision has never been academically and empirically examined.

The military regime managing the bureaucracy in Fiji controls HR selection for senior and executive appointments within the FPS and other corportised institutions. (see Appendix

9.2). Further, the RFMF is a key institution for indigenous Fijians and the government of Fiji.

Historically it has provided essential military and support service to the British Crown,

NATO and the United Nations over the years. From its inception, the RFMF has been traditionally dominated by indigenous Fijians at around 95 per cent. Indigenous Fijians have also aligned themselves more easily with the RFMF in view of the similarities in institutional functions between the RFMF and autonomous indigenous Fijian institutions. Over the years, P a g e | 132

indigenous Fijians have been protective of their turf and territories, and battles are the only mechanisms for resolving conflicts and disputes.

The origins of the RFMF could be traced to the formation of the Armed Native

Constabulary (ANC) that was established by the Colonial Government of Fiji through an alliance between Ratu Seru Cakobau the leading war lord in Fiji at that time and Governor

Gordon, the first Governor of the British Colony of Fiji (www.rfmf.mil.fj). In essence its origin was associated with the Fiji Police Forces, and at that time there was one combined internal and external security institution through the Colonial Government whose primary objective was to subdue and eradicate all forms of opposition. The legality for law enforcement and the quelling of internal disturbances was established by two ordinances. The first was for the regulation of the Police Force and the second for the regulation of an ANC force in 1876. Both ordinances commenced with the governor’s authority to appoint the commanding officer of each force, a Superintendent of Police in the case of the former, and a

Commandant in the case of the ANC. However, both positions during Fiji’s first 12 years as a colony were held by one man, the aide-de-camp (ADC) to the Governor. Further, there were different classifications of designations in the Fiji Police Force: the ANC, known in Fijian as

Sotia; the regular police in Levuka and Suva; the rural police reporting to District

Commissioners; and the village police responsible to the village headman (Turaga ni Koro)

(www.police.gov.fj).

The formal role of the RFMF could be legally traced to the RFMF Act [Cap 81], which is a preamble and declaration by the legislature of the reasons for the passage of the statute. It aids in the interpretation of any ambiguities within the 1970 Constitution and later the 1997 Constitution of Fiji in reference to the operations and management of the RFMF.

The RFMF Act [Cap 81] outlines the formal role of the RFMF, which is to provide for the establishment, maintenance and regulation of military forces for purposes incidental thereto. P a g e | 133

Generally, the RFMF Act [Cap 81] stipulates that the RFMF shall be charged with the defence of Fiji, with the maintenance of order and with such other duties as may from time to time be defined by the Minister.

The RFMF has been indirectly involved to secure stability and peace during the political upheavals in 1987, 2000 and 2006. Reddy et al., (2004) ascertained that state institutions within any government system have to remain intact irrespective of either the selection into power of new governments under the democratic rule of law, or seizure of democratically elected governments, as evident in Fiji. In view of the latter, the RFMF was involved in the four (4) military coup d'état in Fiji, which were a source of insecurity, fear and demotivation for all stakeholders. In the context of this thesis, the most important stakeholder that needs the backing of a firm stable government is the FPSC. It is ascertained that when public servants discern insecurity and injustice, they will of course seek the protection and support of their individual network within the social group and rally behind such social institutions to remove the sources of their insecurity and eradicate injustice. They may also exercise their discretion to counter any illegalities by being the source of the grapevine feeding the numerous blogs that have being established by social media as a counter mechanism. Current members of the FPS have the advantage of being insiders.

As depicted in the RFMF Act [Cap 81], the RFMF shall be charged with the defence of Fiji, with the maintenance of order and with such other duties as may from time to time be defined by the Minister (www.rfmf.mil.fj). Kotobalavu (2012), however, ascertained that the

RFMF intrusion in politics and government has been a founding principle of instability and bad governance. It has impacted the robustness of the FPS and prompted negative publicity for stakeholders to assume that the FPS is politically, socially and legally aberrant, hence being branded as ineffective, inefficient and unproductive. It is deemed that in the process, the HR selection decisions are interfered with. P a g e | 134

Further, it is stipulated that under the RFMF Act [Cap 81] the RFMF shall be under the supreme command of the President and, through the Minister, under the command of the

Commander who shall be appointed by the President upon the advice of the Prime Minister and who shall be responsible to the Minister for the proper expenditure of all public moneys appropriated for the service thereof. The Minister may at any time and for any of the purposes set out in subsection (2) order that the Forces or any part thereof shall be employed out of Fiji; provided that no officer or soldier of the Forces shall be liable to serve or proceed on duty without his consent out of Fiji. In view of this, Kotobalavu (2012) critiqued the formality of the RFMF by insisting that long before Commodore Bainimarama’s seizure of power in December 2006, the RFMF had installed itself in the office of the President of Fiji, took effective control of its operations and arrogated to itself the role of advisor to the

President of Fiji, displacing the Prime Minister and Cabinet from their constitutional role. In the process, the RFMF fully exploited the concurrent role of the President and Commander in

Chief of the RFMF, and the separate functions between the head of state and head of government to totally isolate the operations and management of the President’s office. This finding confirmed that the RFMF has ignored constitutional stipulation that it should operate under the overall policy oversight of government and must adhere to the bureaucratic directions annexed within the administrative function of the executive through the Minister responsible for the RFMF. This was evident during the pre-2006 military coup when the

Commander of the RFMF began to publicly defame the political and professional character of both the then democratically elected Prime Minister, Mr Laisenia Qarase, and the then

Minister for Home Affairs, responsible for the RFMF, Mr Josefa Vosanibola. He was not reprimanded or disciplined for this insubordination. This finding has confirmed that in ignoring constitutional stipulation, the RFMF may also influence HR selection decisions for corporate level appointments in governments. P a g e | 135

4.10 Demographic structure of Fiji in 2011 (base year)

The base year of the research, 2011, is significant for the thesis because it was five years since the current military regime took control of the governance of the Republic of Fiji. The diversity in Fiji’s social groups was even more evident in view of the repercussions of the overthrow of the democratically elected government, as different groupings sided with their major social network ties. The projection for the 2011 population, based on the 2007 Fiji census of population and housing, revealed that the different categories of races were classified into eight subgroups: Chinese, European, I’Taukei, Indians, Part European,

Rotuman, Other Pacific Islanders and Others. Increasingly, the demographic classification for races recognises the heterogeneity of the Fiji population. The reclassification was also in line with the current military regime’s strategy to combat racial overtones and discrimination, particularly changing the indigenous Fijian classification from Fijian to I’Taukei.

In a meeting of the Legislative Council in 1966, the name for the nationals of Fiji was debated, and then leader of government and Chief Minister, Ratu Sir Kamisese Mara, proposed that the indigenous race should be called Taukei (a proud title meaning owners of the land). However, after consultation with the GCC, the word ‘Fijian’ was adopted as a formal classification for the indigenous population. The reason was that it would be easy to distinguish between the races (Mara, 1997). However, the projection for the 2011 population based on the 2007 census officially announced that all Fiji citizens, irrespective of ethnicity, are deemed to be ‘Fijians’. In this regard, a policy initiative for the interim regime in Fiji was for Fiji to move away from an ethnically based society towards an integrated one.

Specifically, the current regime in Fiji had segregated its citizens’ classification on the basis of race and ethnicity. This exercise has indirectly reinforced the basis of networking of the major ethnic classes. The census data highlighted (see Table 4.1) that indigenous Fijians

(I’Taukei) comprised 56.8 per cent of the total population, and Indians 37.5 per cent. P a g e | 136

http://www.statsfiji.gov.fj/. An important observation for this census is that the segregation of the major races had been downgraded to three (3) major categories, Fijians, Indians and others, even though the population sizes grew across the board. As is evident, Fijians and

Indians make up approximately 94.3 per cent of the population of Fiji, hence their influence on the labour market, particularly the FPS, is highly significant. This finding is also important, as the labour market base in the study of the thesis. The data also revealed that

Chinese, Europeans, Part Europeans, Rotumans, Pacific Islanders and others comprised 5.7 per cent of the whole population of Fiji.

Table 4.1: 2007 Census of Population and Housing – Fiji

Population size and growth by ethnicity and geographic sector Geographic Sector Ethnic Group Population Size P1996 (Nr) P2007 (Nr) Total Fiji All 775,077 837,271 Fijians 393,575 475,739 Indians 338,818 313,798 Others 42,684 47,734 Rural Sector All 415,582 412,425 Fijians 232,240 264,235 Indians 170,783 135,918 Others 12,559 12,272 Urban Sector All 359,495 424,846 Fijians 161,335 211,504 Indians 168,035 177,880 Others (http://www.statsfiji.gov.fj/

4.11 Social networking in Fiji

Social networks are either directly or indirectly formed as a result of the continuous association, interaction and development of specific social structural streams. Coleman

(1988) stressed that a prescriptive norm within a collectively that constitutes an especially important form of social capital is the norm that one should forgo self-interest and act in the P a g e | 137

interests of the collectivity. For the FPS study, the specific streams include: ethnicity, gender, provincialism, settlements, politics, school, family and friends. In addition, interpersonal influence and interpersonal selection are vital elements of the networking theory (Acquaah &

Eshun, 2010), and are considered crucial in this thesis because past experiences of social networks can result in new FPS employees bringing established network ties to the organisation. Once established within the FPS, these ties exert themselves and, similar to cell populations, go through a particular type of exponential growth phase called ‘doubling’ and snowball overtime because of pre-contracted commitments that are often difficult to detach from. These contracts have been indirectly formulated within the social system. Thus, each generation of network ties produce social capital that influences different functions of an organisation like the FPS, depending on the interests of the networking group. Further, as the networking groups align themselves to the different organisational function of interests, collusion often happens as the groups strive for support and eventually survival. One of the key outcomes which is considered a crucial social capital variable produced from collusion is organisational politics.

However, in modern Fiji, it is also interesting to note that some members from the two major races have crossed the racial divide, distorting the norms and status quo, and exemplifying a diversified social grouping. Further, issues like interracial marriages have been prevalent for all races, hence shaping a culture that exemplifies globalisation and the new age for Fiji. Consequently, in creating a roadmap for democracy and sustainable socio- economic development, the current regime in Fiji under Frank Bainimarama has sought to build a national identity through a common name to eliminate racial discrimination. In this regard, the objective was to develop a common identity and build social cohesion. The social cohesion key elements include: civic and multicultural education, establishment of a commission for healing, reconciliation, truth and justice, promulgation of an anti- P a g e | 138

discrimination act, teaching vernacular languages, and comparative studies of religion (Fiji

Government, 2007).

The key to achieving social cohesion in Fiji rests entirely on how the two main dominant social groups sacrifice and allow some degree of their status quo to be tampered with. In other words, they need to give space to each other. The outcome would minimise the racial divide to exemplify optimism towards a diversified social grouping and impact the justice perceptions of FPS employees’ public service policies, particularly in HR selection.

Specifically, this chapter examined the origin of those dominant social groups and the association between their interests and influence on HR selection decisions in the FPS. Even though Schuler stated that “organisations are faced with increased social pressure to behave in a socially responsible manner which should not offset the need for HR selection managers to be meticulous in selecting applicants with the relevant set of competencies for each designation” (2000:p 240).

4.11.1 Network ties in Indian social structure

The majority of Indian immigrants were agriculturalist from different parts of India, and a report on the 1956 census of Fiji indicated the numbers of Indian households were classified according to languages spoken. They included Hindustani, Hindi, Tamil, Urdu, Gujarat,

Telegu, Gurmukhi, Malayalam and other languages not listed (Fiji Blue Book: (1891 –

1937),). This initial classification listed eight different cultures and traditions where each culture had its own network ties in relation to religion, political beliefs, school, community ties, family and friendship ties.

Over time and up to the post-independence era, the diversity within the Indian population of Fiji was so vast that it began to create social dysfunction among the Indian population in view of competition for social, economic and political prominence (Norton, P a g e | 139

1977). The competition for prominence within the Indian social structure emanated from deeply engrained culture and traditions which had been developed over centuries. The immigrants brought these diverse cultures and associated network ties from the Indian sub- continent and, as shown in the census of 1956, the different languages that the immigrants spoke and associated themselves with resembled their respective castes, ethnicity and status within the wider Indian social network (Fiji Blue Book:(1891 – 1937), ). As evident in Fiji today, the two major ethnic groups within the Indian social structure are Hindus and

Muslims. However, there are other minor ethnic subgroups which affiliate themselves to the two major groups. The diverse nature of the Indian social systems in Fiji provided leeway for differing religious, political and economic interests. These social beliefs encapsulated different degrees of networking activated from miscellaneous push and pull factors within the

Indian social structure. The push and pull factors induced conflicts and internal bickering.

The conflicts within the Indian population have also escalated since the arrival of the first immigrant ship from India, as these minor ethnic subgroups aligned themselves to their cocoon network and social system.

However, it is also important to note that today; the majority of Indo Fijians have excelled and progressed in the public and private sectors in Fiji which can be accredited to their own competencies and hard work and commitment. Further, it is evident that they also control the business sector of the Fiji economy, where approximately 90 per cent of the small, medium and mega size shops and businesses are owned by them. One could only walk through the streets of the major cities and townships in Fiji to ascertain this finding. The other

10 per cent are owned by others, including people of Chinese origin. While establishing themselves in urban dwellings, citizens of Fiji with Indian descendants have affiliated and committed themselves to the respective settlements and housing estates through religious, P a g e | 140

political, school, provincial and wider extended family networks prevalent in the Indo Fijian social system.

4.11.2 Network ties in Fijian social structure

The Fijian people have always lived in kin groups, which were generally autonomous, with tenuous, political connection with each other (Ravuvu, 1988). In the process, mutual suspicion and fear generally kept them apart, which provided an environment where there is a sense of guardedness in approaching and interacting with one another regarding issues essential for survival. The autonomy was reinforced over time by religious and political networks, where the credibility of such networks in view of the significance of their success, particularly in gaining military prominence, led to the attainment of competitive advantage within the Fijian social structure. This was the initial basis for the emergence of network ties prior to Christianity and the interaction with the white settlers and their way of life. The introduction and adoption of Christianity modified how the indigenous Fijian people worshipped, instilling forgiveness and trust to eradicate mutual suspicion and fear of the unknown.

The Fijian kin groups also varied in size, but were generally fairly small and membership was fluid, allowing dissenters to break away and form their own groups, establishing themselves in various localities – not too close to be harassed and not too distant to receive help from their kin when needed (Ravuvu, 1991). Migration trends were evident, especially when clans and kin groups were harassed and pressured by other kin groups due to competition for survival and subsistence. As is evident today in Fiji, the migratory trends of the kin groups, clans and sub-units happened continuously, and migration paths were always from the interior to the coastal territories. For example, the formation of the 14 provinces

(yasana) in Fiji and three confederacies (Matanitu) was initiated by the colonial P a g e | 141

administrators to formally control and organise the management of indigenous Fijians affairs for administration purposes ((Fiji Government Gazette, No. 1, 1874).

Securing a good education is paramount in the modern Fijian social system where competency building and upskilling is the key for migration from the rural villages to the urban areas and also for migration to other countries. The common reasons for migration were securing a better livelihood and quality of life. Consequently, this has broadened to a new spectrum nowadays, as those Fijians who are accredited with their own specialised competencies have now secured jobs in urban areas in Fiji, purchased land and other properties, and established themselves in settlements and housing estates. In the process, they have an urban dwelling and have inherited a rural dwelling in their own villages and social structure sub-units. While establishing themselves in an urban dwelling, they have also affiliated with their respective settlements and housing estates through religion, politics, schools, provincial, and the wider extended family networks prevalent in the Fijian social structure. The process has also been highly globalised, where Fijians have now immigrated to countries including Australia, New Zealand, US, UK and other European countries, and the same process of establishment through connections and networks continues. As already stated, irrespective of where a Fijian chooses to abide, he is always a member of a kin group and has a piece of land traditionally allotted for survival and subsistence. This allocation has been formally recognised by government during the post colonisation era. In this regard,

Fijians are registered in the Native Lands & Fisheries Commission (NLFC), which was established in 1880 primarily to register native lands, ownership of such lands, and to classify customary roles and migration records of communal units. http://www.fijianaffairs.gov.fj/NLFC.html

The introduction of foreign fighting weapons encouraged and strengthened the formation of special elite groupings within the social structure, with the chiefs and their P a g e | 142

allies. Hence, the most powerful chief would have amassed the most updated fighting arsenal.

In addition, the introduction of a new system of worship challenged the religious norms in the indigenous Fijian social environment, especially when Christianity principles were adopted by the high chiefs. As is customary, the subordinates of the chiefs had no choice but to follow and worship the new religion, so as not to be publicly branded for insubordination and rebelliousness. Lastly, the colonial government introduced a foreign labour force through the indentured labour system to improve efficiency in sugar production, and most importantly to beef up the Fiji government revenue. This was in line with the colonisation policies of the

British Government, where administrators of the colonised territories were directed by the crown to use the resources available within the territories to manage and operate the colonies.

These distinctive social capital variables have shaped and transformed Fiji’s social system over time and are also prominent in influencing HR selection decisions within the FPS.

4.12 Demographic structure of the Fiji Public Service – 2011

In 2011, the base year of the research, the demographic structure of the FPS was significant for this thesis for a couple of reasons. Firstly, the military regime in Fiji had been operating for five years since they took control of the governance of the republic, and stakeholders were sceptical in monitoring the performance of the republic in view of the 11 pillars highlighted in the Strategic Framework for Change and Roadmap for Democracy and Socio-Economic

Development (SFCRDSED). The SFCRDSED was the blueprint proposed by the military regime under Commodore Frank Bainimarama to resurrect a supposedly corrupt government led by the deposed Prime Minister of Fiji, Mr Laisenia Qarase. One of the millennium development goals (MDGs) specified in the SFCRDSED was the elimination of gender disparity in the labour force in Fiji, and promoting and sustaining what is termed ‘gender equality’ in the civil list of 2011 (www.psc.gov.fj). However, this MDG was proposed and P a g e | 143

endorsed by 189 countries including Fiji at the year 2000 Millennium Summit. Demographic figures in relation to the permanent establishment of the FPS in 2011 (see Table 6. 24) recorded that of the 20,886 civil servants, 50.1 per cent were female while 49.9 per cent were male. This reading has achieved the MDG goal set in the year 2000.

Further, section 15 (1) of the State Services Decree 2009 detailed the need for the government to reduce the retirement age for the civil service, Fiji Police Force and Fiji

Prisons Services employees to 55 years. The 2011 Fiji Civil List reported an age distribution analysis (Table 6. 24). As verified, 2.7 per cent of the current public service workforce had exceeded the retirement age of 55 years, but were still being employed by the FPS. The majority of those employed were between the ages of 31 to 40, recording 39.4 per cent. This was followed by those between the ages of 26 to 30 at around 22.3 per cent.

The demographic criteria for ethnicity and race are highlighted in Chapter 10, Section

140 of the 1997 Amendment Constitution of Fiji. It emphasises four important principles for

HR recruitment selection into state services: government policies should be carried out effectively and efficiently and with due economy; appointments and promotions should be on the basis of merit; men and women equally, and the members of all ethnic groups, should have adequate and equal opportunities for training and advancement, and the composition of the state service at all levels should reflect as closely as possible the ethnic composition of the population, taking account, when appropriate, of occupational preferences. However, under

State Services Decree 2009, an important principle was omitted, and that is for members of all ethnic groups to have equal opportunities in training and advancement and also for selection into state services.

P a g e | 144

4.13 Conclusion

The historical, political and social environments have gradually shaped the FPS and subsequently precipitated those networking ties that have influenced HR selection decisions in the FPS. These network ties have been reinforced by the influence of planters, entrepreneurs and missionaries who formalised their existence by forming networks with the local chiefs and also among themselves to protect their interests and assert the legitimacy of their institutions. These influences, together with the new alliances and networking, grounded the future of the colony of Fiji beginning with the urge and motivations of the supreme war lord to cede Fiji on the invitation of Britain. A key feature of those factors that influenced HR selection decisions in the FPS was the introduction of indentured labourers from India who brought different work ethics and network ties which have over time been integrated into the work culture and norms of the private and public sectors in Fiji, particularly the FPS.

The Fiji context chapters provide the necessary complementary literature for readers to understand the social context of Fiji and how social capital was manufactured over time within the historical, political and social environments that have shaped Fiji’s state institutions. Being used conjointly with the literature review in Chapter 2, it provides a solid backdrop, understanding and basis for critical analysis.

In the next chapter, the analysis assesses the HRM challenges facing the FPS and its implications on HR selection decisions; the critical HR selection strategy challenges facing the FPS in view of HR selection decisions; and how HR selection policies of the FPS contribute to the concepts of fit and flexibility to develop a highly qualified HR pool of employees.

P a g e | 145

CHAPTER 5

THE STRATEGIC APPROACHES OF HUMAN RESOURCES SELECTION WITHIN THE FIJI PUBLIC SERVICE

P a g e | 146

Table of Contents

5.1 Introduction

5.2 Response rate

5.3 Archival data and publicly available documents

5.4 Human resources management in the Fiji Public Service

5.4.1 Mimicking of Western HRM practices as a challenge facing the Fiji Public Service

5.4.2 Management of the new Republic of Fiji

5.4.3 Cultural constraints

5.4.4 Pressure for public service reform from developing economies and donor agencies

5.4.5 Changing the mindset of Fiji public servants and effective management of the Fiji Public Service

5.4.5.1 Fijians’ perceptions of social capital

5.4.5.2 Indians’ perceptions of social capital

5.5 Human resources selection and the Fiji Public Service

5.6 The strategic approach of Human Resources Management and the Fiji Public Service

5.6.1 Choices about particular HR policies that best achieve organisational fit

5.6.2 Building theoretical frameworks to guide policy formulation for the FPS

5.6.3 Clarity in HR policies impacting HR selection decisions for the FPS

5.7 Conclusions

P a g e | 147

5.1 Introduction

The theoretical framework and analysis introduced in this chapter set the parameters to critically analyse the HR selection processes in the FPS and examine pieces of legislation that have constitutional significance for the Fiji government, and in particular the FPS, with respect to HR selection decisions. Further, the analysis for this section is using Yin’s (1984,

2004) model of theory testing so as to meticulously examine the impact of social capital through network ties on HR selection decision in the FPS.

As shown in Chapter 2, the literature review conducted for the study was aligned with the research questions of the thesis to identify the research gaps that need to be empirically examined through research questions and analysis of qualitative and quantitative data of the survey. Hence, the rationale for this chapter was to present case evidence on Fiji and the FPS from the context of the literature and the ensuing narrative that exemplifies the concept’s fit and flexibility of HR selection approaches within the FPS. This ascertained the justification that the research gaps highlighted are significant enough to present substantive findings. A critical aspect of this chapter was to ascertain the robustness of the survey instruments in qualitative analysis. Further, the response rate for the survey depicted a handsome return and good representation of the sample in the population studied for the FPS.

In this aspect of the analysis, critical research questions regarding the HR selection strategies of the FPS were assessed. It was also ascertained that rational decision-making is critical for the implementation of HR selection policy choices that best achieve organisational fit. Further, the social, political and economic environment, particularly the current magnitude of social capital for the dominant races, should be continually assessed. These environmental concerns may prompt FPS mangers to be cautious and not too aggressive in the choice and implementation of HR selection policies. Hence, they should always consider P a g e | 148

the environmental preconditions, before formulating and implementing HR selection policy choices.

For the qualitative framework employed, and as exemplified by Yin (1984), the objective was to analyse the case data by building an explanation about the case. In this regard, the objective was to explain a phenomenon by stipulating a set of causal links. This is a preferred and relevant framework for analysing qualitative data. In using the explanation building model, a crucial question arises: how did the causal links reflect critical insights into

HRM focus? For example, the current stature of the system of governance for the FPS was formed in view of various influences from the Fiji social context over the years; hence the iterative nature of causal links was demonstrated through explanation building following a timeline.

Further, explanation building was complemented by time series analysis, which is an analytical framework that portrays a chronology, which is a frequently used technique in case studies. Yin (1984) asserted that for analysis of chronological events, the analytical goal is to compare the chronology with that predicted by some explanatory theory.

The chronology and qualitative data assessed within the qualitative framework depicts the origin of the formation of the government of Fiji, and critical events that led to its development and the modification of HR selection policies and practices. Notable events include the cession of Fiji in 1874; the independence of Fiji in 1970, the beginning of the coup era in 1987 and the annexation of Fiji as a republic (see Figure 5.1).

P a g e | 149

5.2 Response Rate

In the study, the researcher randomly distributed 100 questionnaires for each ministry and the response rate for each ministry (see Table 5.1) was recorded as follows: Ministry of Works,

Transport and Public Utilities – 75 per cent; Ministry of Lands and Mineral Resources – 60 per cent, Ministry of Agriculture – 40 per cent. A total of 169 usable responses resulted from this survey with a total response rate of 56 per cent. The average rate of individual participation in the study within groups responding was 70 per cent of the membership (sd. =

0.18). The high response rate was probably due to the formal ethics requirement of the survey where departmental heads’ approval was granted prior to the survey being conducted. While it was not possible to compare the sample frame for this study with the entire population of the FPS, which according to the latest census in 2011 stands at around 17,819, analysis of the non-respondents indicated no difference in terms of composition, size or location compared to ministries that did participate in the survey. Zikmund (2003) stated that the response rate is defined as the number of questionnaires returned or completed divided by the total number of eligible people who were contacted or asked to participate in a survey.

Table 5.1 Response rate

Ministry No. of No. of Response rate questionnaires questionnaires distributed returned and completed Works, Transport, Energy, Communication and Public 100 75 75% Utilities

Lands and Mineral Resources 100 60 60%

Agriculture 100 40 40%

P a g e | 150

5.3 Archival data and publicly available documents

Acknowledging Yin (1984, 2004), the framework for analysing case studies in this thesis adopted the descriptive approach to identify a type of event that could be quantified and an overall pattern of complexity that was ultimately used in a causal sense. In this regard, and considering heterogeneity, historical, regulatory and affiliation data, the framework of analysis for archival data and publicly available documents used the analytical framework depicted in sections 3.6.1.1 to 3.6.1.3. Further, the collected archival data and publicly available preliminary documents on demographics (group heterogeneity), as defined in the thesis, included the threshold policies on HR selection gathered from the FPSC, the laws of

Fiji which include the; General Orders of Fiji (2011), State Services Decree 2009 (Decree no. 6), the Public Service Act 1999, Fiji Public Service Regulations 1999, Constitution of Fiji

(1970, 1990 and 1997) and the Fiji Civil List which outline heterogeneity variables for all public servants within the organisations under study, the minutes of the GCC meetings from its inception in 1875, and the minutes of the Legislative Council of Fiji during the colonial era.

P a g e | 151

Table 5.2 Archival data and publicly available documents

Publicly Available Documents

1. Fiji Civil List 2. General Orders of Fiji, 2011 3. State Services Decree, 2009 (Decree No. 6) 4. Public Services Act, 1999 5. Constitution of Fiji, 1875, 1970, 1990, 1997

Archival Data

1. GCC Meeting Minutes, from1875 2. Fiji Legislative Council Meeting Minutes 3. Learning Institutions: Origins and Background 4. The Social and Political background of Fiji 5. A New Religion and System of Worship

6. Financial Constraints for the Colony of Fiji

7. Census of Fiji: 1882, 1911, 2007

5.4 Human resources management in the Fiji Public Service

Sub-research Question One:

What are the HRM challenges facing the FPS and what are the implications on HR selection decisions?

The theory on HRM challenges facing the public service for a developing economy like Fiji was exemplified in Chapter 2 of the thesis, highlighting three critical approaches, and these are presented in Figure 5.2. In activating the analytical processes to best answer the question, a qualitative data analysis based on explanation building was used. In this regard, P a g e | 152

the Fiji context chapter was used to highlight the substantive contribution in view of the gaps highlighted in the current literature.

Further, the qualitative analysis employed Yin’s (1984) method where the objective is to analyse the case data by building an explanation about the case. The objective was to explain a phenomenon by stipulating a set of causal links. This is a preferred and relevant framework for analysing narratives, where in this part of the thesis, narratives of the Fiji context were used in conjunction with the survey data to answer the sub-research questions.

The same data were also used for sub-research questions two and three.

However, the substantive contribution of the chapter emanates from Figure 5.1 where key chronological narratives of Fiji gradually influenced HR selection decisions within the

FPS. The ensuing literature has been presented within the Fiji context chapter and narratives.

As exemplified in Figure 5.1 and discussed in the Fiji context chapter (Chapter 4), the critical events not only challenged HRM within the FPS, but also shaped HR selection decisions through the reinforcement of established instrumental (school, religion, politics) and affective ties (family, friendship, community).

P a g e | 153

Figure 5.1 Critical events in Fiji that impact HRM in the FPS

Cakobau Government – Pre Colonisation (1867 – 1874) Dominance of guns and religion Arrival of Europeans

Colony of Fiji – Post HRM Colonisation (1874 – 1970) in the Arrival of Indians FPS

Independence government of Fiji: Post Independence, (1970 – 1987) Economic dominance of Europeans and Indians

Republic of Fiji: Coup d’état era 1987 – 2006 Fiji Military Forces influences on FPS

An example would be the formation of Ratu Seru Cakobau’s Government in Levuka in 1867 which was due to him networking with the early settlers and missionaries (Burns,

1963). Through his strategic alliance with the early settlers, religious and political networks were formed. Further, as settlers’ businesses and farms led to the expansion of trade within the community, key institutions like schools also emerged to take schooling to the wider P a g e | 154

population. The school network in Fiji initially emerged through the religious institutions

(Tavola, 1992). Ratu Seru Cakobau’s networking with key economic players in Fiji, and also with other high chiefs in Fiji including Ma’afu of the Lau group, Tui Cakau of the Northern part of Fiji, Roko Tui Dreketi and others (Burns, 1963) led to a consensus for the cessation of

Fiji to Great Britain (Norton, 2002 ). The networking within the key players also had a critical influence on HR selection decisions when the initial government line-up was formalised in 1874. The first government line up consisted of 31 per cent Fijians and 69 per cent Europeans, and a substantive finding from this thesis is that within that line up, key instrumental and affective networking dominated HR selection decisions, as the leading chiefs of Fiji were related to each other in one way or another (Fiji Government Gazette, No.

1, 1875, See appendix 9.7).

As shown in Figure 5.1 and within the last 27 years, the RFMF has been influencing

HR selection decisions within the FPS. The substantive finding is shown in Appendix 9.2, where the key designations for state and government were controlled and influenced by the

RFMF. In 2011, almost 90 per cent of key designations were controlled and influenced by the

RFMF. This demonstrates a considerable degree of influence that poses to discredit the fine legislation for HR selection within the FPS as outlined under State Services Decree 2009 and the 1997 Constitution of Fiji.

Further, Figure 5. 2 present the HRM challenges normally faced by developing countries, including Fiji. These challenges have been exacerbated through the influence of networking ties on the FPS HR selection decisions, and the political and social upheavals from 1987 which have challenged sturdy HRM, particularly HR selection decisions.

In acknowledging that the current literature on HR selection and social networks has never highlighted the influence of networking ties on the FPS, the application of that literature, as exemplified in Chapter 2, on the FPS is considered substantive in its own right. P a g e | 155

These critical issues are illustrated as HRM challenges through a theoretical framework evident in Figure 5. 2.

Figure 5.2 – HRM challenges facing Fiji Public Service

Fiji Public Service

Mimicking of Pressure for public Western HRM service reform practices Changing the mindset from donors (Mellahi & Frynas, of Fiji public servants (Cassar & Bezzina, 2003) (Hewitt et al., 2002) 2005)

5.4.1 Mimicking of Western HRM practices as a challenge facing the FPS

In this aspect of the analysis, it has been found that a major HRM challenge for the FPS stems from continued use of Western HRM practices brought in to reinforce the operations of the FPS. This can be traced back to when Fiji was ceded to the British Crown in 1874, and the regulatory framework of the state of New South Wales was used in Fiji in the interim

(Fiji Government Gazette, No.1, 1874). During this period, the New South Wales Governor,

Sir Hercules Robinson, was also assigned as the interim Governor for the Colony of Fiji. The importance of the discussion here is that the initiation of the application of Westernised forms of governance in 1874 has been considered a norm. This was also the case when the governance structure evolved from a colonial state through independence to a republic. P a g e | 156

The Western HRM models, which are alien to the functionality of the FPS, were not brought in as a result of comprehensive needs assessment initiated by the FPS backed by a theoretical framework to guide the developmental aspect of ministries and departments within the FPS. Rather, the rationale for their implementation could be aligned with how

Mellahi and Frynas (2003)’s findings inform the discussions that in view of their fragility, developing countries had no alternative but to accept aid from donor agencies, particularly from developed economies. However, as shown in this thesis, the changes were always conducted on a reactive rather than a proactive basis, and the major ‘push factor’ was the change in government. Consequently, the forces behind these changes emanated through the obligation of democratically elected governments to their electorates which indirectly accumulated unnecessary mileage that pressurised the bureaucracy and state institutions. The significant determinants as to why the FPS has to mimic Western HR practices are depicted in Table 5.3.

5.4.2 Management of the new Republic of Fiji

After the constitutional conference in London in 1970, Fiji became a fully sovereign and independent nation within the Commonwealth on 10 October 1970. Fiji was a parliamentary democracy for 17 years after independence, and during that period Ratu Sir Kamisese Mara and the Alliance Party dominated political life. In that era, the Fijian and Indo-Fijian communities were roughly balanced numerically, but political power rested in Fijian hands.

However, the 1987 election brought a shift in the political landscape in favour of Indians when Dr Timoci Bavadra’s Coalition Government installed a Cabinet with a substantial ethnic Indian representation, and indigenous Fijians played on fear of domination by the

Indo-Fijian community (Commonwealth Secretariat, 2006). The result was a series of military coups (1987, 2000 and 2006) and a social situation which has remained troubled and fractured. The most significant event was the deposition of Elizabeth II as Queen of Fiji and P a g e | 157

proclamation of the Republic of Fiji on 7 October 1987. Fiji became a republic in line with political conditions in 1987 as key stakeholders of government including the military formed alliances and network.

The developmental status of the HRM framework within the FPS was also not ready to embark into a ‘republic era’; hence attempting to mimic Western and ‘ready-made’ HRM practices was the only way out. The evidence collected through the narratives of this study found that the then military commander, Commodore Frank Bainimarama, with the help of his senior officers persuaded the then Governor General, Ratu Sir Kamisese Mara, to relinquish ties with the British crown, hence initiating the processes for formalising the

Republic of Fiji. In conducting a study on the FPS reforms, Sarker and Pathak (2003) found that the FPS was not ready because social, political and economic objectives were often conflicting. Thus, the aggressive determination of the social and political networking institutions to attain and assume control of the management of the state and its affairs always took a lead over sound economic management.

The Republic of Fiji then used its existing institutions, including the FPSC, to formulate a regulatory framework to support infrastructural development, adapting existing regulations to meet the executive requirements of the republic. The executive authority for the FPS was executed through the promulgation of Short Title 8(1) of the 1990 Constitution of Fiji:

All existing laws shall have effect on and after the appointed day as if they had been

made in pursuance of the Constitution and shall be construed with such modifications,

adaptation, qualification and exception as may be necessary to bring them into

conformity with the Constitution and this decree. P a g e | 158

As a result of the political developments of 1987 and the birth of the Republic of Fiji, the immense pressure to continue and uphold efficiency and effectiveness to operate the FPS was eminent. Considering the situation, managers in the FPS desperately searched for new ready-to-implement management concepts. This was not done because their previous practices and policies had become obsolete and inadequate in the new corporate and public sector environment, but was out of necessity in order to manage and operate the public service in the new republic.

5.4.3 Cultural constraints

Mimicking Western HRM practices in the Fiji social context was not easy, as cultural constraints often infiltrate with the management of the public services. However given the socio economic and legal environment in 1987, it was the ideal policy initiative. . This was because of the inherent difficulty in transferring HRM practices into a work culture consisting of public servants made up predominantly of Fijians who were used to their communal family-oriented framework and Indians with their individualistic cultural set up

(Nayacakalou, 1975). The political and social environment of Fiji, particularly after the developments in 1987, reinforced segregation of the mainstream races, building a nationalistic culture which infiltrated the HR selection processes of the FPS.

. As shown in the census of Fiji in 1987 (see Table 6.3, p203), the main ethnic groupings are Fijians and Indians. Other ethnic groupings include Rotumans, Part Europeans,

Chinese, other Pacific Islanders and other naturalised Fiji citizens. As recorded by the Fiji

Bureau of Statistics, Fijians recorded 56.8 per cent of the total population of Fiji; Indians recorded 37.5 per cent, while people of other ethnicity recorded a meagre 5.7 per cent. The dominance of the major ethnic groupings have also been recorded in the current population of the FPS, where Fijians recorded 62 per cent in 2003, 63 per cent in 2004, 64 per cent in 2005 P a g e | 159

and 63 per cent in 2011. On the other hand, Indians recorded 34 per cent in 2003, 33 per cent in 2004, 33 per cent in 2005 and 32 per cent in 2011 (Fiji Civil List, 2011).

The impact of mimicking Western HRM practices as a challenge on HRM within the

FPS could also be attributed to the immense pressure for survival of the Republic of Fiji and its public service. As depicted within the analysis (Table 5.3), mimicking Western HR practices as a policy choice has been a challenge for the FPS, but crucial in getting rid of the mismanagement and in stabilising the operations in view of political, social and economic uncertainties. In exercising the powers conferred on it by section 173 (1) of the Constitution and section 15 of the Public Service Act 1999, and with the agreement of the Prime Minister, the FPSC formulated regulations in line with the Public Service (General) Regulations 1999 to confirm the ‘take up’ of HRM in the midst of instability. The regulations include: appointments, promotions and transfers, retirement, retrenchment and termination, code of conduct and FPSC values. The regulations also cover annual reports, corporate plans and performance improvements, and the appointments of senior executives by the FPSC

(www.psc.gov.fj). However, Public Service Act 1999, Public Service (General) Regulations

1999 and State Services Decree 2009 may still need to be subjected to and scrutinised against crucial HRM functions/work practices including: job analysis, HR planning, recruitment and selection, performance appraisal, HR development, career planning and development, employee motivation, pay, rewards and compensation, industrial relations and employee health and safety, so that the HRM ‘take up’ “is not only efficient and effective, but theoretically sound” (Stone, 2013).

As depicted in Table 5.3, there are three key forces in the external environment of the

FPS that impede the take up of Western HRM. These forces provided the leeway for mimicking Western HR practices, and included the challenges in the management of a new republic (the Republic of Fiji) and its system of governance; the cultural constraints in view P a g e | 160

of the clashes between the ethnic and organisational cultures; and the immense pressure for survival and the need to operate the state and the bureaucracy to render essential services.

These external environment sub-factors emanate from the social environment and, as exemplified in Table 5.3, they make up the circumstantial elements from which the leeway for mimicking Western HR practices as the only policy choice emerges. The Fiji context chapter provided the narratives, for example, during the post-independence era and the economic dominance of Europeans and Indians, ethnic tensions and cultural conflicts have developed circumstances where the FPS needs to mimic crucial HR practices as a policy outcome. Further, in view of the instability, the management of a new republic and immense pressure for survival was paramount; hence the need to adopt critical HR practices.

These key forces have also been challenging the implementation of HRM and effective selection practices within the FPS. However, their existence may enable the formulation of crucial HR selection theories from which HR and selection policies can be formulated. This would enable a fit with the embedded case framework for the FPS.

P a g e | 161

Table 5.3 Mimicking Western HR practices and the FPS

Circumstances Leeway Mimicking Western HR practices as a policy choice (Outcome)

Management of a new 1. Regulatory framework Positive republic 2. Infrastructural development Cultural constraints 1. Fijian communal Positive family-oriented framework 2. Indian individualistic cultural set up Immense pressure for 1. Political instability Positive survival 2. Brain drain 3. Economic fragility

The implications on HR selection decisions within the FPS highlighted a wide

disparity between the contemporary and recommended HR selection framework versus the

local competency base for FPS workers. In other words, the HR selection infrastructural

developmental stage of the FPS could not contain the fibre and rigour of the contemporary

and modernised HR selection recommended by donor countries and their agencies. The

outcome confirmed the Bair and Meshoulam (1989) study, where they proposed that an

internal fit encompasses aligning a firm’s HR practices, mutually reinforcing each other. In

view of the adaptation of an alien and recommended HR selection framework, there would be

no internal fit, and it would be difficult to accurately assess how the distinctive HR strategies

are contributing to the business outcome for the FPS.

P a g e | 162

5.4.4 Pressure for public service reform from developing economies and donor

agencies

In this analysis, it has been found that one of the key issues that have challenged HR selection decisions within the FPS stemmed from pressure for public service reform from developing economies and donor agencies. For the FPS, pressures for reform usually emanated from stakeholder claims that the public service units were operating at unacceptably low levels of efficiency and effectiveness. These stakeholders included internal private sector institutions and external financiers, including certain leading developed economies and their financiers.

Some of the reasons identified for the low levels of efficiency and effectiveness included a lack of accountability, over-staffing, corruption, cronyism, etc. (Department of Public

Enterprises, 1998). Additionally, other external agents including donor agencies particularly the IMF, World Bank and the Asian Development Bank have been exerting pressures for reforms and played key role in public service reforms in Fiji (Department of Public

Enterprises Report, 1998).

Political and economic instability in Fiji also exacerbated the situation, constraining public service reform initiatives. In this regard, stakeholders had a key role in evolving HR legislation, particularly HR selection and decisions (see Figure 5.1). As discussed earlier, the political and economic instability in Fiji provided the context for the emerging HR challenges, as illustrated in Figure 5. 2. This pressure-cooker situation catapulted Fiji civil service managers into implementing reforms dictated by ‘change agents’ and their reforms included the application of Western HRM practices, which is not conducive to embracing the merits of public service reform. They had no other choice. However, one has to wonder whether they were actually borrowing or whether they were being forced by the ‘change agents’ to adopt public service reforms and HRM irrespective of the ramifications. The SPI countries’ public service reform patterns all had the same unique framework whereby the P a g e | 163

donor agencies dictated and indirectly controlled the reform processes. In Fiji, for example, at the time of the FPSC’s establishment, Western democracies including Australia and New

Zealand were leading the ‘charge’ for efficiency, expediency, and managerial autonomy through decentralisation of the public services (Fiji Government Gazette No.1, 1874).

However, Fiji neither had the resources nor the flexibility to transform its public services quickly or to keep pace with global trends in public management. The scenario was exacerbated in the period 1987 – 2006, (see Figure 5.1) with unnecessary intervention in its democratic processes and systems of governance. The outcome made Fiji over-reliant on the developed economies within Oceania and the developed world who asserted their status as the ‘life-saver’ of Pacific Island micro states. This over-reliance created a norm over time in these SPI countries, including Fiji, with assertions from reputable institutions like the United

Nations, World Bank and others who classified them as developing or under-developed. As is evident in many SPI countries, including Fiji, donor agencies for developed nations including

AUSAID, USAID, World Bank, Asian Development Bank etc. have been permanently set up to manage this ‘aid’ in its many forms, from infrastructural to HR development. Further, the public service reforms were often initiated through a tripartite arrangement where political mileage was often prioritised against the rationality of incorporating the desired economic objective (Sarker and Pathak, 2003).

In their study, Vosikata et al. (2004) reported that the constraints and challenges for

FPS managers did not begin during the Bainimarama era post 2006. Over the years, the public sector reforms have been incomplete and largely ineffective and Fiji’s economy has not reacted and performed as expected due to a number of factors. These factors include incomplete complementary reforms where the effectiveness of reform in one area often requires reform in other areas. For instance, the reduction of tariff rates requires the restructuring of the domestic tax system. More importantly, for the successful adoption of P a g e | 164

trade reform, property rights (particularly to land) must be well defined and secure, and labour markets should be flexible, otherwise new investments will not be encouraged to take advantage of the prices made possible by the reduction in import restraints. The outcome would likely be loss of jobs and loss of capital as well as increased antipathy towards trade liberalisation, and the Fiji government would feel the pinch in terms of its operating expenditure. Vosikata et al. (2004) also found that privatisation of state-owned enterprises, particularly essential services, has not gone ahead or where they have gone ahead have led to high-cost, public/private monopolies. Therefore, firms continue to face high costs for these essential services.

5.4.5 Changing the mindset of Fiji public servants and effective management of the

Fiji Public Service

The domination of culture over HRM ideology and the tensions between the competing social capital groups within the FPS have resulted in FPS employees clinging to their own social group mindset. A critical lesson for the FPS in view of this important HRM ideology is that public servants must change to embrace the contemporary though innovations, in order to offset the beggar mentality that had been detrimental for the past few decades. In this regard, public servants’ mindset needs to be changed to fully embrace newer ideologies, good governance and, for the purpose of this paper, HRM concepts that improve and sustain worker performance. In this analysis, it has been found that the major issue contributing to the current mindset of Fiji public servants stems from the social environment, through what is termed in this thesis as ‘social capital’ (see Figure 5.3).

P a g e | 165

Figure 5.3 – Changing the mindset of Fiji public servants

Fiji Public Service HR Selection Process Fiji Public Service HR (Traditional) Change Selection Process eeeeee (Modernised)

Fiji Public Servants Fiji Public Servants (Traditional mindset) Change (Modernised mindset)

Fijian/Indian/Others Change SCI eeee Indian Fijian SCI

In view of the research conducted in this study, the finding can be justified from two perspectives: the belief and traditional obligations of Fijians that social capital is inherited and must be owned for life (Ravuvu, 1987) and the ontology regarding economic prowess from Indians that social capital is acquired through savings and hard work, hence their identity must be formally recognised by the state legalising ownership and status. When entering the FPS, public servants from these two dominant ethnic groups transfer these beliefs into the FPS through their ethnic groups’ social capital ideologies, hence mindsets are preconditioned.

P a g e | 166

5.4.5.1 Fijians’ perceptions of social capital

Ever since colonisation, indigenous Fijians, or I’Taukei believe that Fiji belongs to them

(Ravuvu, 1988 ). They also believe that the government belongs to them; hence they feel that they need to control the constitution of the land, the management of the bureaucracy, the public service and, of course, the HR selection processes and decisions (Burns, 1963). These perceptions are deeply ingrained within their mindset and have been revered in their various kinship institutions, including the GCC, with the view that as custodian of the Islands of Fiji, it is obligatory to preserve the kinship (see Figure 5.3) as it is necessary for retaining social status through social capital (Fiji Blue Book: 1891 – 1937).

One of the first census was held in Fiji in 1882 (Table 5.4), and it released figures which highlighted that the different races in Fiji at that time were generally classified either as whites or coloured. Further, the immigrants from distinctive Asian subcontinents were generally classified as Asians, and these included Indians and Chinese. However, within the generalised category, subcategories were then formulated to depict the racial composition of the then colony of Fiji. As is evident, Fijians comprised nearly 90 per cent of the total population of Fiji in 1882. In the latest census in Fiji (projection for 2011 population, based on 2007 census) Fijians still maintain that dominance at 56.8 per cent in 2011.

P a g e | 167

Table 5.4 Census of Population – Fiji, 1882

Whites Coloured Total

Europeans 2898 2898

Halfcastes 776 776

Polynesians 6540 6540

Asians 1487 1487

Rotumans 2142 2142

Fijians 115977 115977

Others 156 156

Total 2898 127078 129976

(Fiji blue book (1876 – 1890).

Norton confirmed that ever since 1897, the colonial government wanted to see the

Fijians ‘standing on their own feet’ (2002: p137). However, the solid foundations of the

Fijian’s social organisation and their mindset could not be uprooted overnight, hence requiring smooth transitional reforms. This would enable Fijians to understand the rationale in paradigm shifts of mindset and institutional structures in view of the operations of profit- oriented entities and containment of costs. Specifically, it would involve a shift from the dominancy of the GCC in matters relating to the management of the FPS, thus a shift towards an efficient methodology of governing the public service and state institutions.

This finding challenges a key operational objective of the current military regime in

Fiji to destabilise the supreme and leading institution of the Fijian people, the GCC (Bose

Levu Vakaturaga), which from its inception was a constitutional body in the Republic of the

Fiji Islands and was established by the British colonial rulers as an advisory body in 1875, a year after Fiji was ceded to the UK. The decision was taken following consultations with chiefs, who advised Sir Arthur Gordon, the governor of the colony of Fiji on how best to govern the colony's indigenous population (Fiji Royal Gazette, No. 1, 1874). P a g e | 168

In his study, Norton described the GCC as follows: it “embodied the privileged relationship of trust and protection established between the Fijians and the British” (1977: p

10). It is also important to note that the GCC being established in 1875 under the governorship of Sir Arthur Gordon portrayed a solid foundation connecting the numerous kinship units which are part of the Fijian social structure. These kinship units (see Figure 5.4) provide the basis for the subdivisions (mataqali) and primary divisions (A i Tokatoka). A primary division (A I tokatoka) is made up of several extended family units, whilst a mataqali comprises of several primary division (A I tokatoka). Further several mataqali will then form a yavusa (division). However, the number of yavusa will depend on the strength and traditional power base of the vanua (major division). Several yavusa (division) are incorporated to form a vanua (major division). Further, several major divisions provide the format for a province (yasana). Hence, a couple of vanua will then form a province (yasana).

For example, the province of Tailevu in Fiji is made up of the following vanua: Bau, Buretu,

Verata, Nausori, Namara, Dravo, Namata, Namalata, Tai, Vugalei, Taivugalei, Sawakasa,

Namena, Dawasamu, Nakelo, Nuku, Tokatoka, Naloto, Wailotua, Nasautoka, Nayavu and

Nailega. Hence, these kinship, primary division, subdivisions and division provide the basic premise for each province (yasana), which connects the 14 provinces of Fiji to the three confederacies (matanitu) in Fiji. Hence, this is illustrated as a big family with an intricate network model.

P a g e | 169

Figure 5.4 – Kinship model of nuclear and extended Fijian family unit

NAME NAME NAME NAME

Tukaqu Tubuqu Tukaqu Tubuqu (Paternal (Paternal Grandmother) (Maternal (Maternal Grandmother) Grandfather) Grandfather)

NAME NAME NAME NAME NAME NAME

Father’s Father’s Father-Tamaqu Mother-Tinaq Mother’s Mother’s Younger Elder Brother Elder Elder Brother Sister (Uncle) Sister (Uncle) (Auntie) (Auntie)

Child 1 Child 2 Child 3 Child 4

The character of Fiji’s social organisation has been evolving, as demonstrated in the changes in the mainstream legislation which show that chiefs have tended to be concerned with their personal advancement and aggrandisement rather than with the welfare of the people (Ravuvu, 1987). In the same social structure, there has been a silent exchange of vows, where the people are obligated and are honoured to serve their chiefs from the differing social structural standings they hold. The chiefs are entrusted to provide protection and concern for the wellbeing of subjects to gain respect and legitimacy for the chiefly designation they hold. According to Coleman’s(1988) model, if A does something for B and trusts B to reciprocate in the future, this establishes an expectation in A and an obligation on the part of B. This obligation can be conceived as a credit slip held by A for performance by

B. If A holds a large number of these credit slips, for a number of persons with whom A has relations, then the analogy to financial capital is direct. These credit slips constitute a large P a g e | 170

body of credit that A can call in if necessary, unless, of course, the placement of trust has been unwise, and these are bad debts that will not be repaid.

The different social structural standings are hereditary, where the tasks and roles of each subdivision (mataqali) have been and will be performed by the chiefs for as long as the social organisation is legitimate.

An important observation by the researcher, who is part of a Fijian kinship structure himself, revealed that the attachment of the Fijians to their social organisation reinforces collectivism and communalism where it is an honour to serve the chiefs and elders, and remuneration and paid employment is non-existent. However, the colonial government’s plight and change of policies in 1897 (see Figure 5.1) induced an individualistic lifestyle and introduced paid employment to the Fijians, which impacted their social orientation. Over the years, this has driven the need for ordinary Fijians to acquire some education and competencies for selection into well-paid employment in not only the public but the private sector. In some instances, ordinary Fijians have overtaken their chiefs academically and professionally. The first two Fijians who graduated with PhDs, for example, are commoners:

Dr Isireli Lasaqa graduated from ANU in 1967, and Dr Rusiate Nayacakalou graduated from the University of London with first class honours in Anthropology. Acknowledging Dr

Nayacakalou’s contribution to leadership in Fiji, the former High Chief, Prime Minister and

President of Fiji, Ratu Sir Kamisese Mara acknowledged: “Dr Nayacakalou was not a chief, but he demonstrated that any man in Fiji today who has the ability, the integrity and the energy can make an outstanding contribution in any field or endeavour either within the nation or internationally” (Nayacakalou, 1975:p 3).

This finding accentuated that the transition to Westernised labour markets has transformed the social status of ordinary Fijians and changed the wealth and competency P a g e | 171

disparity within the Fijian social system. The non-chiefly Fijian populations, who have effectively aligned themselves with contemporary ideologies in training and development and are upskilling themselves, have indirectly amended the kinship structure in the process. It is evident that the context of the social capital of the Fijians and their employment opportunities were evolving as a result of the changes and modernisation of the Fijian kinship structure.

Further, ordinary Fijians, who gained exceptional qualifications and professional status, have not only overtaken their chiefs, but established prominent careers in the public and private sectors. This is also a key substantive finding for this research since within a Fijian social structure, comparing the achievement of commoners with their chiefs was originally considered a taboo, and never documented, either socially or academically. In 2011, for example, all chief executive officer (CEO) positions within the FPS were held by people who were either non-chiefs, or those from other ethnic groupings (Fiji Civil List, 2011).

5.4.5.2 Indians’ perceptions of social capital

The Indians were introduced to Fiji as an essential part of the colonial ruler’s program to limit the capitalist exploitation of Fijian labour, and with their arrival brought an individualistic system of living and way of life. The Indian community in Fiji were classified as British subjects and were backed by their political status and economic prowess; they expected to enjoy rights equal to those enjoyed by other British subjects in the colony including the indigenous Fijians. This intention was broadly encapsulated in Lord Salisbury’s dispatch

(Lal, 2008). Further, this demand for equality was at the core of the political debate at

Legislative Council sessions, and was prominent during the 1965, 1966 and 1970 constitutional conference in London, as Fiji was preparing for self-government in 1970.

These significant changes in the political environment of Fiji provided the basis for the social capital identity (SCI) for Indians, hence, from the end of indenture to a decade before independence, they were of the view that because they had been the mainstay of the P a g e | 172

economy, the Republic of Fiji needed to recognise their contribution and grant them relative civil liberties.

However, in discerning unequal treatment since Fijians were also clinging to their social capital through heredity, domination in government and owning the majority of land, the Indians used other institutional networks, for example the sugar cane farmers union and its subsidiary network institutions, to veto their plea and pressure the bureaucracy for equal parliamentary representation, agitating for constitutional and political reforms. The Indian community in Fiji have always believed that they should be accorded economic and political rights equal to other communities in the colony to substantiate their economic standing

(Norton, 1977).

Further, it was deemed by all economic stakeholders in Fiji at that time that the prosperity of the Fijian colony depended on the collaboration between the “European capital,

Indian labour and Fijian land” (Norton, 1977:p 13) (see Figure 5.5). Over the years, this adage has evolved in view of the changes in mainstream legislation and the social and political environment in Fiji, and is modified in this thesis to read: the prosperity of the

Republic of Fiji depended on the collaboration between the “Indian capital, Indian/Fijian labour and Fijian land”. As evident in the analysis, Indians have taken over the stronghold of early European capital ownership, while Fijians have clawed their way to be prominent in the labour market of Fiji, in view of many Fijians attaining credible qualifications and professional status. Through their savings and hard work, Indians were able to buy land and attain prominent and small to medium businesses in Fiji, and so replace the dominance of the early European settlers. However, ownership and security of land still rests with the Fijians and are managed by a separate institution, the iTaukei Lands Trust Board. The majority of

Fiji’s business community is of Indian ethnicity, however, the number of indigenous Fijian business people is growing (Doing Business in Fiji: 2012 Country Commercial Guide for US P a g e | 173

Companies). However, in view of the current political climate in Fiji and the sensitivity of social capital and its impact, further research may be undertaken in due course on the specifics of ethnic business and competency analysis.

The Indians were introduced to Fiji through the indentured labour system as an essential part of the colonial ruler’s program to limit the capitalistic exploitation of Fijian labour (Norton, 1977). In building their social capital, they believed that their contributions through the indentured labourer system was crucial for the economic development of Fiji, particularly as they paid taxes to the colonial government to fund the overall operation of state institutions in the colony. Further, their work ethic, which was totally in contrast to the communally based subsistence model of Fijians, was more conducive to maintaining productivity levels, particularly in sugar cane farms.

The European farm owners and the CSR at that time also noticed the different work ethic of the Indians. It was also noticed and embraced by the Fiji Colonial Government which then churned out policies to accommodate the labourers’ wish to stay in Fiji as planters

(Norton, 1977). As stated, this work ethic also reinforced the Indians’ SCI, and was subsequently transferred into the FPS when they were appointed to jobs.

P a g e | 174

Figure 5.5 – Ownership of capital and labour as the basis of social capital

The Colony of Fiji The Republic of Fiji Pre/post 1970 Change Pre/post 1987 eeee

Capital: European Capital: Change Indian

Labour: Labour: Indian Change Indian/Fijian

Land: Land: Fijian Change Fijian

In view of sub-research question one: the critical events in Fiji that impact HRM in the FPS have substantiated the critical HRM challenges as illustrated in Figure 5.2 and continue to shape the theoretical framework for HRM in the FPS.

P a g e | 175

5.5 Human resources selection and the Fiji Public Service

Sub-research question two:

What are the critical HR selection strategy challenges facing the FPS in view of HR selection decisions?

This theory on HR selection, as exemplified in Chapter 2 of the thesis, highlighted a few critical aspects of HR selection as one of the key functions in HRM. Firstly, it discussed

HR selection as a stand-alone theory in HRM, and the confirmation that the quality of HR selection decisions depends on the accuracy and completeness of the information gathered about applicants (Gatewood & Field, 1987). Secondly, it presents the importance of maintaining a pool of qualified applicants, since it will ensure that the general competency bases for selected employees are highly regarded and competitive (Rynes, 1992). Lastly, and as confirmed by Gatewood et al. (2011), selecting the most proficient candidate would enable these workers to efficiently perform their tasks, which would then enable organisations to meet and strategically align stated goals to critical business outcomes.

In considering the above HR selection theories, which are conducive to effective and non-biased FPS HR selection decisions, this analysis used an explanatory framework (see

Figure 5.6) to highlight the critical HR selection challenges facing the FPS. The qualitative analysis framework employed is Yin’s (1984) method, where the objective is to explain a phenomenon by stipulating a set of causal links. As previously stated, this is acceptable for analysing narratives and survey data for the FPS study.

The HR selection challenges were assumed to be constraining the attainment of an autonomous HR selection institution within the FPS, and an efficient HR database system to accurately collect and maintain critical information about current and future employees of the

FPS. On this note, the study acknowledges the Fiji civil list and section 5.2 of the FPSC P a g e | 176

manual which covers preparation for FPS employees’ biodata and kinship relations. This policy indicates that information for current employees includes: current positions, date of birth/age, EDP/FNPF number (superannuation identifications), ministry/department/company, qualifications, work experience and annual confidential report grading/references and testimonials (www.psc.gov.fj).

This analysis revolves around Figure 5.6 and through an explanatory effects matrix informs the reader that the HR selection strategy challenges facing the FPS over the years have being exacerbated by the promulgation of mainstream legislation which has always supported ethnic- and raced-based HR selection into the FPS. The determinants of these mainstream legislation changes have otherwise been responsible for distorting the merits, disrupting FPS autonomy and shattering the rationality of HR selection decisions. These were evident in the HR selection policies for the FPS when Fiji became independent (1970

Constitution of Fiji) and when Prime Minister Rabuka came into power after the first military coup (1990 Constitution of Fiji). For example, the 1970 Constitution of Fiji exemplified ethnic- and raced-based HR selection into the FPS as follows:

Section 105 (9) (a), (b) of the 1970 Constitution of Fiji:

In selecting candidates for entry into the public service the Public Service

Commission shall –

(a) give preference, other things being equal, to local candidates who, in its opinion,

are suitably qualified and shall not select persons who are not citizens of Fiji

except to the extent that the Prime Minister has agreed that such persons may be

selected; and P a g e | 177

(b) ensure that, so far as possible, each community in Fiji receives fair treatment in

the number and distribution of offices to which candidates of that community are

appointed on entry.

On the contrary, Section 127 (11) of the 1990 Constitution of Fiji reads:

In the exercise of its powers under this section and notwithstanding any provision

contained in Chapter II of this Constitution, the Public Service Commission, subject

to the provision of subsection (12) hereunder, shall ensure that each level of each

department in the public service shall comprise not less than fifty percent Fijians and

Rotumans and not less than forty percent of the members of other communities.

Further, the regulations was again changed through the 1997 Constitution of Fiji as a result of members of other races disputing that HR selection policies enshrined in the 1990

Constitution of Fiji which favoured Fijians and Rotumans only. Further, when Commodore

Frank Bainimarama took control of the democratically elected government in 2006, the regulations were again perceived to have been changed, even though the legitimacy of that process is still being challenged. This thesis has also examined those fundamental principles or established precedents that regulate HR selection policies. It has been found that the ‘push’ for regulatory change was emanating from the SCI power base for the two dominant races, and formally transferred across to influence the bureaucracy through the democratic processes and parliamentary representation.

P a g e | 178

Figure 5.6 – Explanatory effects matrix: critical HR selection strategy challenges Fiji Public Service

Antecedents:

Year: 1. Competition and conflicts 1. 1970 for SCI (capital and labour) 2. 1987 at severity level. 3. 1997 2. First military coup d'état 4. 2006 3. Internal and external pressure to remove bias HR selection regulations in 1990 Constitution. 4. Commander of Fiji Military for “some obvious Outcomes reasons” has been publicly 1. Constitutional conference in ridiculing the democratically elected government of PM London and Fiji gaining independence. Laisenia Qarase. 2. Deposition of Elizabeth II as Queen of Fiji, hence proclamation of the Republic of Fiji on 7 October 1987. 3. Appointment of a three member Constitutional Review Commission. 4. On 5 December 2006, the commander of the RFMF assumed executive authority and effectively Ensuring HR Selection incapacitated the democratically Policies elected government 1. Section 105 (9) b of the 1970 Constitution of Fiji 2. Section 127 (11) of the 1990 Constitution of Fiji 3. Section 140 ( b, d) of the 1997 Constitution of Fiji 4. State Services Decree (2009) – Part 6, Section 9

The outcome of a series of important events in Fiji’s social calendar induced a paradigm shift in the established precedents by which the Islands of Fiji has been governed; particularly in HR selection policies for the FPS (see Figure 5.6). P a g e | 179

For part 1 of the explanatory effects matrix, 1970 was the base year, and the antecedent was competition and conflict for social capital and labour between the two dominant races reaching an acute level. As explained in the matrix, the outcome was when prominent leaders of the dominant races together with the guidance of the colonial government conferred through a constitutional conference in London to formulate the regulatory framework for Fiji’s independence from the British Crown.

The ensuing HR selection policy as stipulated under Section 105 (9) b of the 1970

Constitution of Fiji reads: “In selecting candidates for entry into the Public Service, the PSC shall ensure that, so far as possible, each community in Fiji receives fair treatment in the number and distribution of officers to which candidates of that community are appointed on entry.” The findings of this matrix clarified that this piece of regulation encouraged the dominant races to cling unto their social capital as they entered the FPS, in view of “fair treatment in the number and distribution of officers”. Further, the regulation does not encourage HR selection on merit. Lastly, it shaped FPS workers’ mindset to be ethnicity based and encouraged segregation.

For part 2 of the explanatory effects matrix, 1987 was the base year, and the antecedent was when Military Officer Sitiveni Rabuka staged a military coup on the democratically elected Labour Government led by Prime Minister Timoci Bavadra. The outcome was the deposition of Elizabeth II as Queen of Fiji and proclamation of the Republic of Fiji on 7 October 1987. The ensuing HR selection policies were “stipulated under Section

127 (11) of the 1990 Constitution of Fiji as follows: In the exercise of its powers under this section, and not withstanding any provision contained in the Chapter II of this Constitution, the Public Service Commission, subject to the provision of subsection (12) hereunder, shall ensure that each level of each department in the Public Service shall comprise not less than fifty percent Fijians and Rotumans and not less than forty percent of the member of other P a g e | 180

communities.” This piece of regulation also widened the disparity gap in the fairness of HR selection policies for the FPS, and encouraged the dominant races and others to cling to their social capital groupings as they entered the FPS., It also increased ethnic tensions, hence the aggressive grasp for social capital (Cottrell & Ghai, 2004).

Indians believed that they have been unfairly victimised, and their survival depended entirely on their economic prowess to legitimise their social capital at all cost. Fijians on the other hand believed that as long as they controlled the government their social capital was safe and wouldn’t be threatened. As argued by Appana (1998), a tripartite policy initiation blunder by the Alliance Party Government of Ratu Sir Kamisese Mara led to the most prominent ethnic segregation and a divisive situation between the two dominant ethnic groupings, where Indians supported the FLP/NFP coalition, while Fijians supported the

Alliance Party of Ratu Sir Kamisese Mara, a prominent chief in Fiji. These then migrated from the private sector to the public sector, particularly the public service reinforcing in- depth ethnic segregation on FPS workers mindset, where the general inclination was to support one’s own ethnic grouping irrespective.

For part 3 of the explanatory effects matrix, 1997 was the base year and the antecedent was the discriminatory provisions of the 1990 Constitution of Fiji, which reserved the office of Prime Minister, the majority of members for the House of Representatives, and the 50: 40 per cent ratio in HR selection in favour of Fijians and Rotumans, plus the pressure from internal and external stakeholders of government. The outcome of this antecedent led to the appointment of a three-member Constitutional Review Commission in 1995 to immediately review the 1990 Constitution of Fiji. Their work produced Section 140 (b, d) of the 1997 Constitution of Fiji which modified the HR selection policy to read:

The recruitment of persons to a state service, the promotion of persons within a state

service and the management of a state service must be based on the following P a g e | 181

principles: (b) appointments and promotions should be on the basis of merit; (d) the

composition of the state service at all levels should reflect as closely as possible the

ethnic composition of the population, taking account, when appropriate, of

occupational preferences.

This piece of regulation has certain elements of ambiguity for example: (d). However, even though it encourages the dominant races and others to cling to their SCI as they enter the

FPS, it also provides a leeway for the principle of merit for HRM through recruitment and selection.

Finally, in part 4 of the explanatory effects matrix, 2006 was the base year when the

Commander of Fiji Military for ‘some obvious reasons’ began publicly ridiculing the democratically elected government of Prime Minister Laisenia Qarase. He was also publicly perceived to have been going against the administrative authority of the Ministry of Home

Affairs, which was seen as insubordination (Fiji Times, 2 November, 2006). Further, many

Indians disliked the ethnically exclusive policies of the Soqosoqo Duavata ni Lewenivanua

(SDL) political party led by Laisenia Qarase and thought they had been robbed of their

‘democratic’ choice and started passive resistance in the form of letters to the editor, blogs, and the grapevine. The precursor event of 5 December, 2006, led the commander of the

RFMF to assume executive authority and effectively incapacitate the democratically elected government of Prime Minister Laisenia Qarase. He is still in power to date. The ensuing HR selection regulation was enshrined as State Services Decree (2009) – Part 6, as follows:

The recruitment of persons to a state service, the promotion of persons within a state

service and the management of a state service must be based on the following

principles:(a) government policies should be carried out effectively and efficiently

and with due economy; (b) appointments and promotions should be on the basis of P a g e | 182

merit and equal opportunity; (c) men and women should have adequate and equal

opportunities for training and advancement.

The most distinctive difference between 1997 Constitution of Fiji and State Services Decree

(2009) – Part 6 was that the latter did not highlight “the composition of the state service at all levels should reflect as closely as possible the ethnic composition of the population, taking account, when appropriate, of occupational preferences.” In other words, State Services

Decree (2009) – Part 6 was more explicit and reinforced meritocracy in HR selection and decisions within the FPS.

Many stakeholders and commentators on the management of state services and good governance, including New Zealand, Australia, the UK, US and the majority of the Western world, perceived the actions of the Commander of the RFMF as illegal. Thus, the current state services mechanisms set up by his regime are still perceived by those commentators as constitutionally illegal (Kotobalavu, 2012). In view of the above, the State Services Decree

(2009) – Part 6, Section 9 could also be perceived as illegal, hence challenging the institutional legitimacy of the FPS.

However, it is by far the best piece of legislation to date, as there is no evidence of discrimination and it ensures a flow of employees irrespective of gender, race and age. It also provides leeway for meritocracy in HR recruitment and selection in the FPS. It should also be assessed for its justice elements. Hence, in critically assessing the fairness and justice aspects, three types of perceptions of fairness are evident. These are distributive, procedural and interactional fairness. An analysis of this piece of HR legislation against justice theories follows. For example, Section 15 (1, 2, 3, 4) of State Services Decree (2009) – Part 6 stipulates the retirement age of the FPS as follows:

15 (1) Notwithstanding anything contained in any written law, the retirement age in the public service shall be 55 years. P a g e | 183

(2) The retirement age of 55 years in subsection (1) above, shall extend to the civil service, Fiji Police Force and the Fiji Prisons Service. (3) Any person employed in the civil service, Fiji Police Force and the Fiji Prisons Service, who is already over the age of 55 years, shall retire on 30 April 2009. (4) Nothing in this section prevents, upon retirement, the continued employment of a person on a fixed term contractual employment, if his or her services are required in the Government.

As stipulated, the retirement age of 55 in subsection 15 (2) only extends to members of the FPS, members of the Fiji Police Force and the Fiji Prisons Service, however members of the Fiji Military Forces are exempted, hence are eligible for selection within the Fiji

Military Forces and other ministries and department of the FPS. As evident in Table 6.24 (p

241), a FPS Age Distribution Analysis conducted in 2011 revealed that 2.6 per cent of public servants are over the age of 55, the mandatory retirement age. However, the figures may also include expatriate workers with specialised skills who have been recruited by the current military regime, current Members of Parliament who have also been selected by the regime, and their cronies. As shown in the data, 49 per cent of those still employed are females, while

51 per cent are males. Section 15 (1) of the State Services Decree 2009 details the need for the government to reduce the retirement age for the civil service, Fiji Police Force and Fiji

Prisons Services to 55 years. This highlights that, irrespective of the positives of the State

Services Decree (2009) – Part 6, the application of such legislation is still a big concern for the FPS, as members of different groups, occupational groups in this example, are very keen to maintain their social capital status. In the process, meritocracy has always been challenged, since the FPS has not been properly mandated for full autonomy and insulation from political interference in its application of rules and regulations. This finding was also verified through a study of HRM in India, particularly when in one government project’s success story, one of the critical success factors was the ability to formulate policies that insulate the organisation’s operations from political interference and install full autonomy (Nankervis et al. 2002). In the P a g e | 184

process, the project manager was able to select employees on merit in an environment free from political influence.

5.6 The strategic approach of human resources management (HRM) and the Fiji

Public Service

Sub-research question three:

Do the HR selection policies of the Fiji Public Service contribute to the concepts of fit and flexibility to develop a highly qualified HR pool for the FPS?

The theory on the strategic approach of HRM facing the public service for a third world economy like Fiji was exemplified in Chapter 2 of the thesis, highlighting three important approaches. These are choices about particular HR policies that best achieve organisational fit, building theoretical frameworks to guide policy formulation for the FPS and the importance of clarity in HR policies on selection and worker performance.

To best answer the stated research question, a qualitative data analysis based on explanation building following Yin’s (1984) method was used. The objective is to analyse the case data by building an explanation about the case. In this regard, a phenomenon is discussed by stipulating a set of causal links about it. This is a preferred and relevant framework for analysing narratives, where in this part of the thesis, publicly available documents on the FPS were used in conjunction with the survey data to answer the sub- research questions.

It would be prudent when critically examining this research question to understand that HR selection is not a stand-alone function, hence digressing into HR strategy, performance management and understanding the importance of productivity indicators is crucial. P a g e | 185

5.6.1 Choices about particular HR policies that best achieve organisational fit

It is evident that the most crucial ingredient to ensure that rationality as a precondition for excellent HR policy choice is achieved is political stability. In the FPS context, this important ingredient has been missing for more than two decades. However, choices about the particular HR policies that achieve best organisational strategy, goals and outcomes can be made by taking into account the influences of the external factors such as the economy, politics, legislation and the labour market, as well as internal factors such as strategy and workforce characteristics (see Figure 5.7).

Figure 5.7 – Determinants for FPS HR policy strategic choices

FPS External Environment

Economy

FPS Internal Environment 1. HR regulations Conducive HR 2. Workforce characteristics Policy Strategic 3. FPS infrastructural Choices for the standards FPS

Political Societal

(Adapted from Jackson et al., 1989)

P a g e | 186

This aspect of the analysis followed the conventional ideology ascertained by Jackson et al. (1989) that personnel practices vary as a function of organisational characteristics. The outcome here needed to consider the peculiar situation that the FPS is operating in, an era of political instability which may affect organisational characteristics in the internal and external environments. For the Fiji ministries examined, data were gathered on workforce characteristics, HR regulations, the political climate and societal characteristics. Likewise, when HR selection practices are strategically aligned to the changes in external and internal environmental factors, the outcome of HR practices have not resulted in the desired organisational fit between HR practice and organisational strategy. . Hence in considering the environmental conditions, HR selection practitioners of the FPS may vary strategies along a number of dimensions, therefore considering a number of strategy typologies to apply. The rationale from this ideology is for the FPS to implement HR reform policies relative to the specific economic and political phases of development of Fiji, and not to be too aggressive given the merits and fibre of critical organisational characteristics. This is a key lesson for the

FPS.

As evident from the flow chart depicted in Figure 5.8, a conducive HR policy strategic choice for the FPS needs to consider its internal and external environmental factors.

Firstly, the political climate has been crucial in affecting HR policy strategic choices. The military regime’s control of HR selection for senior and executive appointments within the

FPS has drastically affected HR policy strategic choices and the overall strategic direction of the FPS (see Appendix 9.2). This thesis found that key designations in all ministries and departments including the President, Prime Minister, ministers, Chairman and committee of the PSC, permanent secretaries, Commissioner of Police, Commissioner of Prisons, the

Solicitor General, Director of Public Prosecution, Director of Immigration, Fiji ambassadors to foreign missions, Auditor General, supervisor of election, divisional commissioners, etc. P a g e | 187

are all currently held by military appointed officers or their designates. Hence, a key finding was that through the political instability and military intervention, the FPS was not fully insulated from political interference. In the process, the autonomous framework of HR selection policies has been continuously interfered with.

Secondly, with political instability and military rule, ethnic tensions were rife, hence inducing a fragmented and fractured society. It is evident that the frictions between the dominant races and religions affected the fibre and core of the societal environment. These frictions reinforced the desire to cling to social capital, affecting the internal organisational characteristics of the FPS. In particular, as public servants were motivated to upskill and enhance their workforce characteristics, they still held on to their social and kinship status.

This means that segregation ideologies were still being kept alive. Further, frictions between public servants in regard to their ethnic and religion beliefs could build extremism which would sabotage ideologies and dissipate the infrastructural standards of the FPS. Hence, the political climate in Fiji works against workforce commitment and instead encourages workforce resistance.

Table 5.5 HR selection regulations in the FPS

Year FPS HR Selection Policies 1997 Section 140 (b, d) of the 1997 Constitution of Fiji: The recruitment of persons to a state service, the promotion of persons within a state service and the management of a state service must be based on the following principles: (b) appointments and promotions should be on the basis of merit; (d) the composition of the state service at all levels should reflect as closely as possible the ethnic composition of the population, taking account, when appropriate, of occupational preferences. 2009 State Services Decree (2009) – Part 6: The recruitment of persons to a state service, the promotion of persons within a state service and the management of a state service must be based on the following principles:(a) government policies should be carried out effectively and efficiently and with due economy; (b) appointments and promotions should be on the basis of merit and equal opportunity; (c) men and women should have adequate and equal opportunities for training and advancement.

P a g e | 188

It is evident from Table 5.5 that the current HR regulations are an excellent piece of legislation which discourages segregation and encourages meritocracy. However, the unstable external environment has enabled the military regime to control HR selection within the FPS.

For the FPS, it can be argued that operational or functional HRM practices do exist to ensure that the current regime continues to manage state institutions and the bureaucracy; however the strategic aspects of HRM may not be evident. This is due to economic, political and societal constraints as exemplified in Figure 5.7. The central idea behind the strategic aspects of HRM is that all initiatives involving how people are managed need to be aligned with and in support of the organisation’s overall strategy (Mello, 2011) (see Figure 5.8).

Figure 5.8 – Putting strategy in its place – FPS

FPS Strategic Analysis 1. Industry analysis 2. Customer trends 3. Environmental forecast 4. Competitor analysis 5. Assessment of internal strengths, weakness, resources

Supporting Organisational Mission Arrangements : Fundamental purpose 1. Structure : Values 2. Process

Strategy 3. Symbols The central 4. Functional policies integrated and profiles externally 5. Compensation oriented 6. People concept of 7. Activities Objectives how the FPS : Specific targets will achieve its objectives

(Adapted from Mello, 2011) P a g e | 189

It is evident that strategy formulation for the FPS is not aligned and is mis-configured to strategy implementation which means that the recruitment and placement of people is inefficient. Jackson et al. (1989) found that an organisation’s decision to use some personnel practices in favour of others should be a response to the imperatives of the organisation’s industry, strategy, technology and structure.

5.6.2 Building theoretical frameworks to guide policy formulation in attaining organisational fit for the FPS

Building and understanding well-grounded theoretical frameworks of SHRM in the operating mechanisms of the FPS to guide the deployment and use of HR activities and practices towards the stated goal is also a critical challenge for FPS managers. As per the above context, it is absolutely essential that FPS managers are not only guided by a relevant theoretical framework to incorporate and instil elements of rationality to improve accuracy of decision-making and strategy formulation, but also must understand the preconditions, merits and rationality behind the predictions of the model and its framework (Vosikata and Faifua

2012).

Practitioners and academics can’t ascertain the precedent ideology or the preconditions to base the development of theories to guide the relevant work practices for the

FPS. Hence, one of the key challenges for FPS managers is to streamline the theoretical framework through an institution like a Fiji Public Service Institute of Management (FPSIM) to project the preconditions, merits and rationality behind the formulation of policies and strategic directions. Current policy directions indicate that the strategic approach to HRM does exist in the FPS, even in an environment of political instability. However, it is hard to verify at the moment given the FPS supporting organisational arrangements. A policy initiative from this perspective is that the specific elements of Figure 5.8 should be adopted to formulate an initial HRM theoretical base. P a g e | 190

5.6.3 Clarity in HR policies impacting HR selection decisions for the FPS

One of the rationales of this aspect of the analysis is to ascertain whether the clarity/non- clarity in work performance policies may be influencing the fit of HR selection policies to individual outcome and/or organisational productivity. Critical outcome indicators that depict the proper alignment and fit or otherwise of policies to strategic business outcomes are turnover rate and tardiness. These are productivity indicators and their magnitude resembles the effectiveness of policy application in institutions within the FPS. Table 5.6 presents current work performance policies for the FPS.

Table 5.6 Worker performance policies for the Fiji Public Service

Year Worker Performance Policies

Pre 2009 ACR – Annual Confidential Report: using the following variables to assess worker performance: quality of work, quantity, knowledge required by the job, initiative and resourcefulness, judgement, degree of supervision required, responsibility and ability to communicate.1

Post 2009 PACO – Performance Assessment for Contract Officers: initially analyses workplace assessment using the following variables: punctuality and attendance, interpersonal communication, flexibility, commitment to PSC values and code of conduct, presentation, team relationship. (www.psc.gov.fj)

To ensure that HR policies embrace the concepts of fit and flexibility, FPS managers need to ensure that worker performance standards are clarified to FPS employees and other stakeholders. As of now, the Fiji PSC has set in place worker performance benchmarks for civil servants, however, the clarity of these worker performance instruments need to be critically examined. The Fiji PSC Performance Assessment for Contract Officers (PACO)

P a g e | 191

initially analyses workplace assessment using the following values: punctuality and attendance, interpersonal communication, flexibility, commitment to PSC values and code of conduct, presentation and team relationship. The dimensions are then further reduced and analysed using a five point Likert scale. Those variables for analysis include: professional attributes, supervisory and leadership attributes, skills and abilities, standard and quality of work performance, initiative and resourcefulness, judgment, attitude, responsiveness and qualification, training and competencies.

However, a major challenge for FPS managers in the use of PACO is the analysis for the outcome. Using a five-point Likert scale in gathering the raw data is quite impressive, however the subsequent analytical framework to summarise the appraisal needs to be upgraded

For those who were appointed prior to 2009, their performance was assessed using the

Annual Confidential Report (ACR). The ACRs of public servants are written with a view to judge their performance every year in the areas of their work, conduct, character and capabilities. By convention, the system of writing confidential reports has two main objectives. First and foremost is to improve performance of the workers in their present jobs.

The second is to assess their potentialities and to prepare them, to be filled by the reporting, reviewing and accepting authorities in an objective and impartial manner. The current ACR of the Fiji PSC use the following variables to assess worker performance: quality of work, quantity, knowledge required by the job, initiative and resourcefulness, judgement, degree of supervision required, responsibility and ability to communicate.

As per the above, the study on the FPS critically examined the clarity of worker performance measurement standards, particularly in the institutions under examination. The

ACR and PACO format has been assessed in terms of how they can accurately measure the tasks and roles of appointed officers. Hence, the essence of critiquing a model is to build on P a g e | 192

the founding principles so as to formulate newer models relevant to the current circumstances where organisations are operating with the view to not only meeting but also exceeding customer expectations. These newer models provide the basis for the formulation of organisation policies and regulations to ensure that inputs are efficiently utilised, costs contained and productivity enhanced and sustained.

Further, in view of the intangible aspects of services, worker performance measurement standards for FPS institutions must be clarified if performance needs to be measured accurately. This finding is also crucial in reinvigorating the current performance appraisal mechanisms employed by the FPS, which is the ACR for those appointed prior to

2009, and contracts for those appointed post 2009. These performance measurement methods are used across the board in all ministries and departments of the FPS, irrespective of the job classifications and core competency grading, and may cause confusion in the interpretations of actual worker performance. . . These confusion and ambiguity may influence the application of other HRM sub-functions within the organisations.

In theory, traditional job evaluation based pay ties the worker’s pay to the worth of the job based on the job description, hence the pay is job oriented, while competence based pay ties the worker’s pay to his or her competencies. For the FPS managers, the critical question when it comes to measurement of performance and productivity is: are they measuring performance against the job worth specified in the job description or are they measuring against the competencies of the incumbent? It may also be advisable to incorporate both phases depending on the character of the job content and specifications needed.

The diversity of the service rendered from within the FPS ministries meant that each function needed to develop their own worker performance assessment measures relative only to the core job characteristics of their ministries or departments. Further, the distinctive P a g e | 193

ministerial or departmental worker assessment measures should be aligned against a national framework on worker performance for the FPS.

For example, a critique for PACO and ACR revealed that PACO assesses the punctuality and attendance for contracted officers while ACR does not directly address tardiness issues, even though it may indirectly cover the same issue. The researcher understand that the issue of ‘written and unwritten laws’ may be a key issue here. This needs to be highlighted in this chapter because it has contributed immensely to worker performance and productivity issues for the FPS, reinforcing the misalignment of policies to outcome.

A key Fiji ministry in the survey recorded figures (depicted in Table 5.7) to highlight late arrivals in minutes for their public servants working in the seven divisions particular. As illustrated, this particular ministry recorded approximately one thousand one hundred ninety

(1189.7) hours as lost due to late arrival or tardiness in 2011.

Table 5.7 Late arrival – sample ministry – 2011

1st Quarter 2nd Quarter 3rd Quarter 4th Quarter Total

17,821 15,556 15,024 22,981 71,382

(Ministry of Agriculture, Survey Report, 2011) Figures are in minutes.

In considering the productivity lost factor for this indicator (late arrival) for the ministry, the assumption for the calculation of work performance is to multiply the daily mean total hours lost with the daily mean wage rate for the service sector as published by the

Fiji Bureau of Statistics on June 2012 (see Appendix 9.1).

P a g e | 194

(Daily mean wage rate for the service sector) x (Total hours lost per annum/Total number of days per annum)

27.33 (1189.7/365) = FJ$89 per day worker performance decrease.

Hence for the Ministry of Agriculture, late arrival has contributed to on average approximately FJ$90 per day productivity lost which is a considerable sum. In view of this analysis, it is advisable that ministry revise their HR selection and performance policies to reduce the total hours lost for late arrival, reduce the current daily productivity lost margin at around FJ$90, and most importantly decrease the misalignment margin. The stated productivity loss has to be aligned with the current FPS HR selection regulations, where the assumption is that State Services Decree (2009) – Part 6 has been found in this thesis to be one of the best FPS HR selection policy regulations. However, the shortcomings may be identified in the application of State Services Decree (2009) in various FPS institutions, and the connectivity between the main regulatory framework and subsidiary legislations.

5.7 Conclusions

Confirming the findings in the literature and providing evidence for the research question, the key HRM challenges for the FPS concluded that the FPS doesn’t have a choice but needs to mimic Western ideologies and ready-made public service policy directions. The fragility of

Fiji’s social, political and economic environment aggravated by a period of political instability left FPS policymakers no alternative but to ensure that the state institutions are operational, because they need to survive.

Evidence provided in the literature also confirmed that key HRM challenges come from the belief that social capital reinforces kinship and ethnicity status to control HR selection within the FPS. The finding was justified from two perspectives: the belief and P a g e | 195

traditional obligations of Fijians that social capital and identity are inherited and must be owned for life and the ontology for economic prowess from Indians that social capital is acquired through savings and hard work; hence their identity must be formally recognised by the state by legalising their ownership.

Critical HR selection challenges facing the FPS have been constraining the attainment of an autonomous HR selection and recruitment institution within the FPS, and an efficient

HR database system to accurately collect and maintain critical information about current and future employees of the FPS. These included competition for and conflict over social capital

(capital and labour), the military coup era in Fiji beginning in 1987, internal and external pressure to remove bias HR selection regulations in 1990 Constitution and the internal conflicts between the military and the government of the day.

The key strategic HRM challenges facing the FPS included obtaining and building theoretical frameworks to guide policy formulation in the FPS that promotes fit and flexibility. ; Further, these frameworks would enhance HR policy formulation choices that best achieve organisational objectives and clarity in HR policy application for the FPS that confirm the fit or otherwise. It is also obvious from the context of current HR policies on performance policies for the FPS that the mainstream regulations are crystal clear, however ministerial and departmental regulations are heavily infiltrated. This has caused ambiguities within the context of the policies which may create confusion, duplication and complications in the application of those policies.

P a g e | 196

CHAPTER 6

CHARACTERISTICS OF INSTRUMENTAL AND AFFECTIVE TIES THAT BUILD AND MAINTAIN SOCIAL CAPITAL INFLUENCE ON HUMAN RESOURCES SELECTION DECISIONS WITHIN THE FIJI PUBLIC SERVICE

P a g e | 197

Table of Contents

6.1 Introduction

6.2 Social capital network approach and Human Resource selection in the Fiji Public Service

6.3 Archival data and publicly available documents

6.4 Social capital and the Fiji Public Service

6.4.1 Representations within the distinctive social systems of the dominant social groups

6.4.2 Representations and influence in government and Cabinet

6.5 Networking ties

6.5.1 Instrumental ties and HR selection decisions in the Fiji Public Service

6.5.2 School and religious influences within the Fiji Public Service

6.5.3 Political influences

6.6 Affective ties and HR selection decisions in the Fiji Public Service

6.7 Demographics of the Fiji Public Service and social capital

6.7.1 Representations of dominant ethnic groups within the Fiji Public Service

6.7.2 Age distribution analysis within the Fiji Public Service

6.8 Conclusion

P a g e | 198

6.1 Introduction

The analysis introduced in this chapter exemplifies the critical examination of the embedded case study labelled Case Study Two. It presents evidence of characteristics of instrumental and affective ties for FPS employees that build and maintain social capital influence on HR selection decisions within the FPS. Specifically, the case study presents the answers to the sub-research questions four (a), (b) and (c), as follows:

Sub-research question four (a):

Could the key relations of the dominant social groups in Fiji that build and maintain social capital influence HR selection decisions within the FPS?

Sub-research question four (b):

What are the key characteristics of instrumental and affective ties within the FPS that have the potential to influence HR selection decisions?

Sub-research question four (c):

Do the demographics of employees in the FPS ministries examined correlate with a level of sustained social capital and social networking?

The analysis examined the impact of social capital through network ties on HRM policy and practice in the FPS. Further, the likely influence of the impact of social capital identity on HR selection was also evaluated. Under a qualitative framework, data triangulation was employed where multiple-level analysis was used to answer the above research question. The levels of analysis included: the FPS organisation as a whole, ministries and departments in ministries and individual employees of the FPS.

As presented in Chapter 2, the literature review conducted for the study was aligned with the research questions of the thesis in view of identifying the research gaps that need to be P a g e | 199

empirically examined through research questions and analysis of qualitative and quantitative survey data. Hence, the rationale for this chapter is to present the analysis of the raw data to justify that the research gaps highlighted are significant in the formulation of contemporary

HR selection theories for the FPS. A critical aspect of this chapter is to ascertain the robustness of the survey instruments in qualitative analysis. Further, the response rate for the survey depicted a handsome return and good representation of the sample in the population studied for the FPS.

In this aspect of the analysis, critical research questions regarding the influence of the social and political environment in Fiji towards social capital aggregation was assessed.

Further, the likely influence of the impact of social capital identity on HR selection was also evaluated. It was also ascertained that the rationale for decision-making is critical for the implementation of HR selection policy choices that best achieve organisational fit. Further, the social, political and economic environment, particularly the current magnitude of social capital for the dominant races should be continually assessed. These environmental concerns may prompt FPS mangers to be cautious and not too aggressive in the choice and implementation of HR selection policies. Hence, they should always consider the environmental preconditions before formulating and implementing HR selection policy choices. The analysis also revealed the important influences and correlations of the distinctive social capital variables including: high school name, political affiliation, religious grouping, ethnicity, township/city/rural community and which provinces respondents belong to. These variables have the potential to influence who is selected for the FPS ministries.

P a g e | 200

6.2 Social capital network approach and HR selection in the Fiji Public Service

The objective for this part of the analysis is to present a brief descriptive analysis of the bio- data for survey respondents. The fibre and robustness of SC triggers ownership beliefs that impact selection into the FPS. For Fijians, social capital and affiliation are determined by the tribal system and kinship group. They believe that social capital and identity are inherited and should be treasured for life, and include controlling government and who gets selected into the FPS. On the other hand, Indians’ perception is that social capital are acquired through their economic prowess and hard work accumulating considerable savings which has reinforced Fiji’s economic growth. They believe that their SC must be formally recognised by the state, and it should impact who gets selected into the FPS. The specific variables that are assumed to control and manufacture social capital in this study are: high school name, political affiliation, religious grouping, ethnicity, township/city/rural community and which provinces respondents belong to. The underlying assumption for the analysis is that Fijians have political status and economic status through land ownership, while Indians have economic status through proprietorship of various major, medium and small business ventures, but are struggling for political status. Hence, the variables surveyed demonstrate the strength of the social capital of Fijians and Indians based on their political and economic status.

6.3 Archival data and publicly available documents

Acknowledging Yin (1984, 2004), the framework for analysing case studies was adopted the descriptive approach to identifying a type of event that could be quantified; hence an overall pattern of complexity reduction was used in a causal sense. Considering heterogeneity, historical, regulatory and affiliation data, the framework of analysis for archival data and P a g e | 201

publicly available documents used the analytical framework depicted in sections 3.7.1.1 to

3.7.1.3 (pp 96 – 98). Further and as presented in table 6.1 , the archival data and preliminary documents on demographics (group heterogeneity), as defined in the thesis, include the threshold policies on HR selection gathered from the FPSC, the laws of Fiji including the

General Orders of Fiji (2011), State Services Decree 2009 (Decree no. 6), the Public Service

Act 1999, Fiji Public Service Regulations 1999, Constitution of Fiji (1970, 1990 and 1997) and the Fiji Civil List which outline heterogeneity variables for all public servants in the organisations under study, the minutes of the GCC meetings from its inception in 1875, and the minutes of the Legislative Council of Fiji during the colonial era. Other data include management records on selection, performance assessment, and absenteeism and turnover rates.

Table 6.1 Archival data and publicly available documents

Publicly Available Documents

1. Fiji Civil List 2. General Order of Fiji, 2011 3. State Services Decree 2009 (Decree No. 6) 4. Public Services Act, 1999 5. Constitution of Fiji: 1875, 1970, 1990, 1997

Archival Data

1. GCC Meeting Minutes, w.e.f. 1875 2. Fiji Legislative Council Meeting Minutes 3. Learning Institutions: Origins and Background 4. The Social and Political background of Fiji 5. A New Religion and System of Worship

6. Financial Constraints for the Colony of Fiji

7. Census of Fiji: 1882, 1911, 2007 P a g e | 202

6.4 Social capital and the Fiji Public Service

Sub-research question four (a):

Could the key relations of the dominant social groups in Fiji that build and maintain social capital influence HR selection decisions within the FPS?

In examining the above question, archival data, publicly available documents and survey data were used to highlight the key relations of the dominant social groups – Fijians and Indians – in Fiji. These included their representation in the social context of Fiji and their representation in government and Cabinet. The environment consists of the internal environment (government and Cabinet) and the external environment (representation in the distinctive social systems of the dominant social groupings). These two environmental factors have the potential to influence HR selection decisions within the FPS. The current ethnic population of the FPS provided evidence on this critical issue.

6.4.1 Representation in the distinctive social systems of the dominant social groupings

The general census data managed by the Fiji Bureau of Statistics was a critical source for highlighting the representation of the dominant social groupings in the distinctive social systems. This aspect of the analysis examined one of the early censuses in Fiji conducted in

1882 (see Table 6.2), the representation recorded 89 per cent for Fijians, 5 per cent

Polynesians, 2 per cent Europeans or ‘Whites’ as they were officially classified, and 1 per cent Asians. The Asian population included Indians and Chinese.

P a g e | 203

Table 6.2 Census of Population – Fiji, 1882

Whites Colored Total

Europeans 2898 2898

Halfcastes 776 776

Polynesians 6540 6540

Asians 1487 1487

Rotumans 2142 2142

Fijians 115977 115977

Others 156 156

Total 2898 127078 129976

(Fiji, The blue book (1876 – 1890).

In the latest census recorded by the Fiji Bureau of Statistics was conducted in 2007, and is presented in Table 6.3, the two main social groupings were Fijian and Indian. Fijians were 56.8 per cent of the total population of Fiji and Indians were 37.5 per cent, while people of ‘other ethnicity’ recorded a meagre 5.7 per cent.

Table 6.3 Census of Population and Housing – Fiji, 2007

Population size and growth by ethnicity and geographic sector

Geographic Sector Ethnic Group Population Size P1996 (Nr) P2007 (Nr) Total Fiji All 775,077 837,271 Fijians 393,575 475,739 Indians 338,818 313,798 Others 42,684 47,734 Rural Sector All 415,582 412,425 Fijians 232,240 264,235 Indians 170,783 135,918 Others 12,559 12,272 Urban Sector All 359,495 424,846 Fijians 161,335 211,504 Indians 168,035 177,880 Others (http://www.statsfiji.gov.fj/ P a g e | 204

Considering 1996 as the base year, in the latest census of 2007, it is important to note that the total population of Fiji increased by approximately 8 per cent. Specifically, the Fijian population increased by around 20 per cent, while the Indian population decreased by 7.4 per cent.

However, even though the Fijians have the majority representation in the government and Cabinet of Fiji, in the FPS and in the distinctive social systems of the dominant social groupings in the external environment, the Indians have dominated the business sector of Fiji.

A brief example is presented in Table 6.4, which shows the ethnicity of licensee holders for hotels and guest houses in the Fiji Tourism Industry. This example shows the dominance of

Indians through their social capital grouping in the business community. It should be noted that even though Indians dominated Fiji’s business landscape, there are limited authorities to publicly substantiate this claim because of lack of research. The writer has also verified the above claim with the Registration of Companies in Fiji. As of now, registration of company in Fiji creates a separate legal entity from its owner(s), with the rights and obligations of a company specified in the Companies Act as modified by its own constitutional documents

(Memorandum and Articles of Association) (http://www.investmentfiji.org.fj/). In other words, records do not verify business ownership by ethnicity.

Table 6.4 Ethnicity of licensee holders – Hotels and Guest Houses, Fiji Tourism

Industry, 2005

Year Fijian Indian Others Total

2005 69 73 163 305

(Fiji Parliamentary Paper no. 110 of 2006) P a g e | 205

As shown, Indians controlled around 24 per cent of the total hotel licenses in Fiji in

2005, while Fijians controlled around 22 per cent. Also, the percentage for Fijians may decrease, since a majority of those registered were merely hotel managers and landowners.

Their status as land owners gave them the right to be minor shareholders.

6.4.2 Representation and influence in government and cabinet

During the formation of the colony of Fiji in 1874, there were no political parties or democratic processes for the selection of the government. Hence, the state institutions made up the core functions of government where, the HR selection of public servants was mostly aligned with the business and social interests of those who had already established themselves in the major business centres, particularly Levuka, and those of chiefly status. In other words, the kinship system in Fiji dominated or crowded out the assumed democratic systems of government of Western societies. Within that process, selection of government was historically based on chiefly status rather than election through meritocracy. For example, the ethnic composition of the initial FPS included 28 Europeans, who were either business people or farmers, and 13 Fijians who were high chiefs (see Appendix 9.7). This trend continued and, as exemplified in Table 6.8, Mr Douglas Brown, a farmer, became the

Minister for Agriculture in the 1972 Cabinet line-up (Norton, 1977). The early domination of

Fijians was also evident in key government positions (see Table 6.5), where the names of the

Fijian high chiefs in government are shown. These Fijian high chiefs were mostly employed by the Department of Native Affairs, which was known later as the Ministry of Fijian Affairs.

P a g e | 206

Table 6.5 Department of Native Affairs, 1874

1. Henry Charles Thurston, ESQ; (Stipendiary Magistrate for the South Coast of Viti Levu and Kadavu); 2. Alexander Eastgate, ESQ; (Stipendiary Magistrate for the North Coast of Viti Levu and Yasawa) 3. John Packer Gore Martin, ESQ; (Stipendiary Magistrate for Vanua Levu and Taveuni) 4. Horace Caton Emberson, ESQ; (Stipendiary Magistrate for the Windward Islands, including Moala, Totoya, and Matuku, and to be Deputy Collector of Customs at Lomaloma 5. Maafu (Roko Lau) 6. Tui Cakau (Roko Cakaudrove) 7. Ratu Epeli (Roko Tailevu and Naitasiri) 8. Ratu Savenaca (Roko Central Islands or Lomaiviti) 9. Tui Macuata (Roko Macuata) 10. Tui Bua (Roko Bua) 11. Tui Dreketi (Roko Rewa) 12. Cagi Levu (Roko Kadavu) 13. Ratu Kini (Roko Nadroga) 14. Ratu Isikeli (Roko Ra) 15. Roko Tui Bau (Roko Ba and Yasawa) 16. Matanitobua (Roko Namosi

(Fiji Civil List 1, 1874)

Indications of the representation and influence in government of the dominant ethnic groups also focussed on the composition of government and Cabinet line-up as the Fiji government began to incorporate the democratic processes which were alien to the traditional governance structure that Fijians were used to. The assumption here is that certain segments of the Indian population of Fiji had experienced some form of democracy since India had been under British rule. Table 6.6 presents the Council of Ministers for Fiji in 1967 and is evident, three prominent chiefs in Fiji held high positions, including the position of Chief

Minister. The ethnic representation was 30 per cent Fijians, 50 per cent Europeans and 20 per cent Indians; hence the domination of Europeans and Fijians are evident.

P a g e | 207

Table 6.6 Council of Ministers, 1967

1. Ratu Sir Kamisese Mara – Chief Minister 2. Ratu Penaia Ganilau – Minister for Fijian Affairs 3. Ratu Sir Edward Cakobau – Minister for Commerce 4. Vijay Singh – Minister for Social Services 5. H P Ritchie – Minister for Finance 6. Charles Stinson – Minister for Works 7. Doug Brown – Minister for Natural Resources 8. John Falvey – Minister without Portfolio 9. Justin Lewis – Attorney General 10. Samy Reddy – Assistant Minister of Education (Mara, 1997)

As discussed in Chapter 4, Lasaqa (1984) found that the ever-increasing pressure from the two dominant social groupings in Fiji for social and political prominence led the

British government to hold a constitutional conference in London in 1970, which devised constitutional reform to reconstruct the seating arrangement of the Legislative Council of Fiji

(see Table 6.7).

Table 6.7 House of Representatives Makeup, May 1970

Communal Roll National Roll

Fijian 12 10

Indian 12 10

General 3 5

(Lasaqa, 1983)

Even though the seating arrangement of the Legislative Council of Fiji appeared to be uniform, as seen above, the Fijians still had the upper hand in numbers due to a clause in the constitutional referendum which allows the GCC to effect certain government appointments P a g e | 208

by law. The GCC is an autonomous institution, established in 1876 under the governorship of

Sir Arthur Gordon following the formalisation of regular consultations of high chiefs of Fiji with the government on how best to govern the colony’s indigenous population. Norton

(1977) said that, it embodied the privileged relationship of trust and protection established between the Fijians and the British. Further, Scarr (1984) confirmed that it is an institution where Fijians had substantial control of their own affairs in government, and in fact had the constitutional power to effect certain government appointments by law. For example, one of the major functions of the GCC in the period 1945 to 1962 was to nominate Fijian representations to the Legislative Council (Lasaqa, 1984). In this regard, it reinforced the majority of Fijians in government and Cabinet. Lasaqa (1984) also said that the GCC was empowered in 1970 to nominate eight senators whose major role was to safeguard Fijian interests in parliament. Hence the GCC was instrumental in reinforcing the structure of the

Fijian chiefly system with government, and in the process statutorily charged with the duty of advising government on matters affecting Fijian interests.

Table 6.8 – Fiji Cabinet, 1972

1. Ratu Sir Kamisese Mara – Prime Minister 2. Ratu Sir Edward Cakobau – Deputy Prime Minister 3. Ratu Penaia Ganilau – Minister for Home Affairs 4. Vijay Singh – Minister for Urban Development 5. Ratu George Cakobau – Minister without Portfolio 6. Charles Stinson – Minister for Finance 7. Doug Brown – Minister for Agriculture 8. James Shankar Singh – Minister for Health 9. John Falvey – Attorney General 10. Ratu Wili Toganivalu – Minister of Fijian Affairs 11. Ratu Josua Toganivalu – Minister for Lands 12. Jonati Mavoa – Minister for Labour 13. John Falvey – Attorney General 14. Jone Naisara – Minister for Education 15. Mohammed Khan – Minister for Commerce (Mara, 1997) P a g e | 209

The above table (6.8) presents the first Cabinet for Fiji after the elections under the

1970 Constitution. As is evident, four prominent chiefs in Fiji held high positions in this

Council, and this was the first time that a high chief of Fiji was made Prime Minister. In this

Cabinet line-up, 58 per cent were Fijians, 21 per cent were from the Indian community and

21 per cent were from the European community. An ethnic analysis for the Council of

Ministers 1967 and the Fiji Cabinet line-up in 1972 shows that the Fijian representation increased to 58 per cent while the Indians recorded 21 per cent and increased by about 1 per cent. In later governments, Ratu Sir Kamisese Mara and the Alliance Party (1967 – 1987);

Timoci Bavadra and the Fiji Labour Party (13 April 1987 – 14 May 1987); Ratu Sir Kamisese

Mara (1987 – 1992); Mr Sitiveni Rabuka and the SVT (Soqosoqo ni Vakavulewa ni Taukei)

Party (1992 – 1999); Mr Mahendra Chaudary and the Fiji Labour Party (1999 – 2000); Mr

Laisenia Qarase and the SDL (Soqosoqo Duavata ni Lewanivanua) Party (2001 – 2006); the composition of their Cabinet line-up reflects the dominating social group and ethnicity that their political party manifestos were based on (Ratuva, 2013).

Further, the contextualisation of the political processes is evident as former prime ministers of Fiji used different frameworks which drove their political manifestos, for example, the pro-indigenous policies of Ratu Sir Kamisese Mara’s Alliance Party after independence in 1970; Mr Mahendra Chaudary’s appeasement strategy; and Mr Laisenia

Qarase’s social engineering approach (Ratuva, 2013). Ratuva also found that Commodore

Frank Bainimarama had contradictory tendencies in his indigenous development strategies, where he demonised affirmative action openly, but on the other hand he reinvented it under the rubric of rural development.

P a g e | 210

Lastly, in examining the current Fiji government cabinet line-up in Fiji for 2011 (see

Table 6.9), the ethnic composition are as follows: Fijians recorded 70%, whilst Indians recorded 30%.

Table 6.9 Cabinet, 2011

1. Commodore Frank Bainimarama – Prime Minister; Minister of Finance, Strategic Planning, National Development and Statistics; the Public Service, Peoples Charter for Change and Progress, Minister of Information, National Archives and Library Services of Fiji; Minister of Lands and Mineral Resources; Minister for Intake Affairs, Minister for Sugar Industry, Commander of the Fiji Military Forces 2. Mr Joketani Cokanasiga – Minister of Defence, National Security and Immigration 3. Mr Filipe Bole – Minister of Education, National Heritage, Culture and Arts 4. Mr Inoke Kubuabola – Minister of Foreign Affairs and International Cooperation 5. Dr Jiko Luveni – Minister of Women, Social Welfare and Poverty Alleviation 6. Captain Timoci Natuva – Minister of Works, Transport and Public Utilities (Water and Energy) 7. Mr Aiyaz Sayed – Khaiyum – Attorney General and Minister for Justice, Anti- Corruption, Public Enterprises, Communications, Civil Aviation, Tourism, Local Government, Urban Development, Housing and Environment, Industry and Trade 8. Dr Neil Sharma – Minister for Health 9. Commander Viliame Naupoto – Minister for Youth and Sports 10. Mr Jone Usumate – Minister for Labour, Industrial Relations and Employment 11. Colonel Inia Seruiratu – Minister for Agriculture, Fisheries and Forests, Rural and Maritime Development www.pmoffice.gov.fj; 14.01.14.

The ethnic and social grouping composition stated above reflects the selection criteria of the interim regime, as they had been managing the affairs of Fiji since 2006. Even though one of their key themes was to install a government of national unity which would eradicate all forms of discrimination, the ethnic composition of their Cabinet line-up reflected an ethnic

Fijian majority. This discourse drove the assumption that in managing the affairs of the state and government, social capital and networking is rife hence influencing HR selection and decisions within the FPS.

P a g e | 211

6.5 Networking ties

Sub-research question four (b):

What are the key characteristics of instrumental and affective ties within the FPS that have the potential to influence HR selection decisions?

6.5.1 Instrumental ties and HR selection decisions in the Fiji Public Service

Harrington (2008) stated that instrumental ties have low intrinsic value of relationships outside task settings. She further found that in an instrumental-based society, members would perceive that any relationships outside the work context would have low intrinsic value and considered to be low in significance.

Further, exploratory factor analysis (EFA) was used to investigate whether there was a relatively simple underlying structure that produces the observed associations among the 10 questionnaire items (see p103). The 10 questionnaires measured FPS workers’ perception of networking ties and justice perceptions. The general perception from respondents was that network ties do exist in the FPS institutions and that their influences have affected the justice perception of current employees on HR selection processes and decision-making.

The questionnaire responses from employees in the three ministries, i.e. Ministry of

Works, Ministry of Agriculture and Ministry of Lands, summarised the three case samples from the FPS. The responses were used in the analysis as follows: school network ties in the

Ministry of Agriculture – Fiji; political network ties in the Ministry of Works – Fiji; and religious network ties in the Ministry of Lands – Fiji. Using the three cases exemplified an embedded case study for the FPS which was the basis of the survey and was complemented with archival data research. Hence, the social capital affiliations of the FPS workers were assumed to portray their affiliation where school, religious and political networks were the key influence in HR selection to FPS positions. In essence, the instruments of analysis were P a g e | 212

formulated to ascertain that people were appointed on the basis of school and religious ties rather than on the basis of merit (as is assumed in HR practice).

6.5.2 School and religious influences in the Fiji Public Service

Fiji’s first schools were founded by European missionaries. The foundation of its early formal education system was established with Christian conversion as its base (White, 2003). The arrival of two British Wesleyan missionaries, David Cargill and William Cross, with several

Tongan pastors-teachers in Lakeba in 1835 would prove precipitous in transforming the religious and educational landscape of rural Fiji. Formal education in Fiji began with the

Methodist missionaries quickly learning the Fijian language in order to translate the scriptures. By 1838, a Fijian grammar had been completed and a printing press established.

The missionaries diligently compiled vocabularies and dictionaries and began translating the

Bible and other related materials into the various dialects of Fiji (Tavola, 1992). The relationship between schooling institutions and religion provided the foundation for competency development which was used by government and the public service. In view of the importance of learning institutions to the colonial government of Fiji, and the need to educate the children and citizens of Fiji, the missionaries began establishing the Methodist missions to not only train the lay preachers and mission workers in Christians principles and doctrines, but to also train them as teachers of basic English and Maths, and also for those who were eager to learn or were recommended by their kinship and social network (Tavola,

1992). On the outset Coleman (1988) added if social capital is lacking in both the family and the community, a religiously based school or a boarding school would be most valuable for the child. This finding is evidenced in some fragmented societies in Fiji including urban communities and settlements when rural dwellers move to urban centers for employment and general improvement of living standards. In the process, they meet other dwellers and over time form strong community relations where social capital is formed. P a g e | 213

Further, the relationship between schooling institutions and religion extended to the

Fijian villages, groups of villages (tikina) and provinces (yasana).Thus, by1900, and due to the efforts of the missionaries, there were schools in most Fijian villages offering up to four years of education, and school attendance had become a normal of part Fijian childhood

(Tavola, 1992). The relationship between schooling institutions and religion also extended to the Indian rural and urban settlements.

The first schools for Indians were started by the Methodist Mission and the Marist

Brothers. By 1916, the Anglican Church and Indian religious groups such as the Arya Samaj and Muslims were also operating schools for Indians. The involvement of Indian organisations was significant, as it marked the beginning of a self-generated demand for education which was to gain momentum, as is evident in Fiji today (Tavola, 1992). .

Initially, there was a high literacy rate for Fijians in the vernacular compared with

Indians, who at that time had limited participation in schools. Fijians were significantly ahead in terms of literacy, access to, and attendance in schools. However, Indians recovered very well to out manoeuvre the Fijians, particularly in the post-independence period. This was the case for Hindu religious organisations like the Arya Samaj, and they are now the financial backer for the University of Fiji, the Arya Samaj Nursing School and a medical school. In view of the above, ethnicity also became an important player in the development of competency and schooling institutions in Fiji (Norton, 1977).

Wesleyan mission schools have been credited with single-handedly creating high rates of literacy in the vernacular among Fijians, providing the educational infrastructure for a colonial government that would initially be minimally involved. On the other hand, the

Marist Roman Catholic missions founded the first mission schools to use the language as the medium of instruction that had the most prestige and advantages in the colony of Fiji –

English. White (2003) supported Clammer’s (1976) findings that, while Wesleyan schools P a g e | 214

provided a superficial level of religious knowledge and academic achievement in the vernacular for the Fijian masses, Catholic mission schools earned distinction as institutions promoting relatively high educational standards catering to economically elite and urban students.

It was also understandable that initially non-Fijians and those with the Roman

Catholic faith would assimilate with the Marist schools and mission. However, in today’s diverse culture in Fiji people have shelved the status quo and crossed the floor for exceptional educational achievements, changing the social capital imbalance between the Fijians and

Indians. Ethnicity was the dividing line for choosing which schools students could matriculate at.

Table 6.10 to Table 6.16 present the demographics for the ministries surveyed. The data for the Ministries of Agriculture and Works were used for this analysis because the data were highly reliable, with response rates at around 40 per cent and 75 per cent respectively.

At the individual analysis level, the respondents’ highest education level were analysed to ascertain this competency strength. The readings for the two ministries are recorded in Tables 6.10 and 6.11. A comparative analysis indicates that the Ministry of

Agriculture employees were superior, as a couple of their officers had attained post graduate education. This may be due to the specialised competencies needed for the ministry and its operations within the Fiji context.

P a g e | 215

Table 6.10 - What is the highest level of education you have completed? – Agriculture

Frequency Per cent Valid Per cent Cumulative Percent Valid 1 2 5.0 5.0 5.0 2 7 17.5 17.5 22.5 3 1 2.5 2.5 25.0 4 6 15.0 15.0 40.0 5 14 35.0 35.0 75.0 6 5 12.5 12.5 87.5 7 5 12.5 12.5 100.0

Total 40 100.0 100.0

However, the Ministry of Works was equally impressive with a couple of undergraduate and post-graduate qualifications.

Table 6.11 - What is the highest level of education that you achieved? – Works

Cumulative Per Frequency Per cent Valid Per cent cent

Valid Did not complete HS 3 4.0 4.0 4.0 Complete HS 9 12.0 12.0 16.0 Certificate level 14 18.7 18.7 34.7 Diploma level 38 50.7 50.7 85.3 Degree level 9 12.0 12.0 97.3 Post graduate level 1 1.3 1.3 98.7 9.0 1 1.3 1.3 100.0 Total 75 100.0 100.0

The above reading is verified through Figure 6.1 and Figure 6.2 where for the

Ministry of Agriculture, the majority of employees spent at least 15 years of total schooling.

As depicted, the outcome presented in figure 6.1 reflects the returns for question 7 of the survey questionnaire.

P a g e | 216

Figure 6.1 – Total years of schooling _ Agriculture

On the other hand, for the Ministry of Works, the majority spent 14, 15 and 16 years of total schooling. This also indicated a wider spread in the years of total schooling, hence a greater dispersion for the KSA for its employees. The key issue regarding the larger dispersion is that several departments and line ministries were amalgamated by the current regime under the Ministry of Works and Public Utilities.

P a g e | 217

Figure 6.2 – Total years of schooling – Works

As depicted in Tables 6.12 and 6.13, the gender readings highlighted that there was less disparity, hence an equal representation for this key demographic element. This was in line with the current regime’s push for equal representation of gender.

Table 6.12 - Please indicate your gender – Agriculture

Cumulative Per Frequency Per cent Valid Per cent cent Valid Male 24 60.0 60.0 60.0

Female 16 40.0 40.0 100.0 Total 40 100.0 100.0

P a g e | 218

Table 6.13 - Please indicate your gender – Works

Cumulative Per Frequency Per cent Valid Per cent cent Valid Male 30 40.0 40.0 40.0 Female 45 60.0 60.0 100.0 Total 75 100.0 100.0

In terms of the demographic variable age, for the Ministry of Agriculture, the 26 – 34 age groups dominated at around 35 per cent.

Figure 6.3 – Age – Agriculture

P a g e | 219

For the Ministry of Works, the 26 – 34 age group also dominated at around 41.3 per cent.

Table 6.14 – In what year were you born? – Works

Cumulative Per Frequency Per cent Valid Per cent cent Valid 1.0 6 8.0 8.0 8.0 2.0 31 41.3 41.3 49.3 3.0 20 26.7 26.7 76.0 4.0 13 17.3 17.3 93.3 5.0 5 6.7 6.7 100.0

Total 75 100.0 100.0

The above readings indicated a booming generation X population for both ministries, which is encouraging to note.

Table 6.15 – Please indicate your ethnic group – Agriculture

Cumulative Per Frequency Per cent Valid Per cent cent

Valid Fijian 26 65.0 65.0 65.0 Indo Fijian 13 32.5 32.5 97.5 Part European 1 2.5 2.5 100.0 Total 40 100.0 100.0

In terms of ethnicity (see Tables 6.15 and 6.16), Fijians dominated the reading at around 65 per cent and 73.3 per cent for both ministries.

Table 6.16 - Please indicate your ethnic group – Works

Cumulative Frequency Percent Valid Percent Percent Valid Fijian 55 73.3 73.3 73.3 Indo Fijian 18 24.0 24.0 97.3 Rotuman 2 2.7 2.7 100.0 Total 75 100.0 100.0

The dominancy of Fijians at this individual level is also reflected at the operational and organisational levels, with the organisational level sitting at around 67 per cent. P a g e | 220

Further, at the individual level, the research question has been analysed in terms of a comparative analysis on school, religious and political affiliations for respondents in the three sample ministries. For this analysis, the three different ministries were assigned to represent a key instrumental tie variable. The Ministry of Agriculture was assessed in view of school networking relationships, the Ministry of Lands and Mineral Resources was assigned religious networking, and the Ministry of Works was assigned political networking.

Frequency distribution analysis was also used to supplement and substantiate the descriptive analysis findings using SPSS version 21.

Table 6.17 School network relationships in the Ministry of Agriculture

Descriptive Statistics N Minimum Maximum Mean Std. Deviation Skewness Statistic Statistic Statistic Statistic Statistic Statistic Std. Error q11a. I know many of the 40 1.00 7.00 4.0250 2.21287 .101 .374 people I work with from school q11b. School 40 1.00 7.00 3.2250 1.90125 .579 .374 relationships are good for productivity q11c. School 40 1.00 7.00 3.7750 1.91469 .384 .374 relationships motivate me to get the job done q11d. Working with 40 1.00 7.00 5.3250 1.97922 -.854 .374 people from the same school keeps me from leaving the organisation q11e. I value working 40 1.00 7.00 4.5250 1.97403 -.260 .374 with people who come from the same school q11f. I arrive at work on 40 1.00 7.00 1.9000 1.35495 2.208 .374 time because I like to work hard q11g. I arrive at work on 40 1.00 7.00 4.5250 2.17194 -.217 .374 time because I enjoy the company of people from the same school q11h. Because of school 40 1.00 7.00 4.3000 2.31162 -.058 .374 relationships, I feel that I am playing my part as a team player in this organisation. q11i. I stay in this 40 1.00 7.00 2.1250 1.34331 1.899 .374 organisation because my job activities are personally meaningful to me q11j. I arrive at work on 40 1.00 7.00 1.9000 1.69161 2.205 .374 time because I am confident about my ability to do my job Valid N (listwise) 40 P a g e | 221

The descriptive analysis for school relationships (see Table 6.17) summarises respondents’ perceptions when asked to rate the importance of working with people they went to school with. As is evident from the spread, 40 FPS employees in the Ministry of

Agriculture responded to the seven-point Likert scale survey instruments. The majority of responses at a mean of around 5.3 indicate that current FPS employees sometimes deduced that working with people from the same school kept them from leaving their organisation; hence school affiliation is quite strong within the ministry. The respective standard deviation for this variable sits at around 1.98, which indicates that employees have diverse views on the question. This also confirms that some employees still value working with people they went to school with. Further, the spread generally indicated that knowing fellow workers from school indicated a modest mean at around 4.1, indicating that employees of the Ministry of

Agriculture know fellow workers and that they attended the same schools. In this regard, school relationships as a social capital determinant may not be openly discussed, but there is evidence that it could impact HR selection policies and decisions. The positive skewness of the data determined how strongly the respondents related to each another in terms of school relationships. In this regard, it may be instrumental in influencing HR selection decisions.

The above finding was also complemented through a frequency distribution analysis

(see Figure 6.4). While acknowledging employee perceptions that school relationships are important in their everyday work life, it was also important to find out those schools that dominated the relationships and networking. Queen Victoria School (QVS), Lelean Memorial

School (LMS) and Nasinu Secondary School topped the frequency distribution, totalling around 30 per cent. They were closely followed by ACS and RKS at 15 per cent and Rishikul

High School at 5 per cent. The other 20 secondary schools were evenly distributed at around

2.5 per cent each, totalling 50 per cent. Queen Victoria School, LMS, Nasinu Secondary,

ACS and RKS are predominantly ethnic Fijian schools, while Rishikul High School is P a g e | 222

predominantly Indian. This again verified the dominancy of the Fijian population within the

FPS where a couple of Fijian schools dominated the analysis, followed by an Indian secondary school (Rishikul High School) at around 5 per cent.

Figure 6.4 School – Ministry of Agriculture

The domination of Queen Victoria School could be traced back to the colonial era. Its role as flagship educational institution in Fiji was substantial, especially in the colonial period, as it was formed to groom sons of Fiji high chiefs for future leadership in government and society (Clammer, 1976). Coupled with the Anglican Church, Hindu schooling institutions, Muslim schools, Marist Roman Catholic and Wesleyan mission schools in the P a g e | 223

primary and secondary sectors, education in Fiji gained momentum and blossomed over the years (Tavola, 1992). The secondary school institutions were the breeding grounds that provided HR for the public and private sectors. In the context of this thesis, the FPS has relied on the education system in Fiji for employee intake. Cross-examining the strong school affiliation, as recorded in Figure 6.4, with ethnicity, the analysis found that ethnicity was highly significant within the school population. In this regard, the main determinant of the strong school affiliation recorded in this study was ethnicity. This finding proved that ethnicity, which stems from the social system and networking for Fijians and Indians, was the key ingredient for instrumental and affective networking within the FPS.

P a g e | 224

Table 6.18 Ministry of Lands – Fiji – Instrumental Ties – Religious

Descriptive Statistics N Minimum Maximum Mean Std. Deviation Skewness Statistic Statistic Statistic Statistic Statistic Statistic Std. Error q13a. I know many of the 54 1.00 7.00 4.8704 2.25745 -.436 .325 people I work with because we belong to the same religion q13b. Religious 54 1.00 7.00 3.4630 2.01610 .293 .325 relationships are good for productivity q13c. Religious 54 1.00 7.00 3.8148 2.27434 .148 .325 relationships motivate me to get the job done q13d. Working with 54 1.00 7.00 5.6296 2.02155 -1.220 .325 people from the same religion keeps me from leaving the organisation q13e. I value working 54 1.00 7.00 3.7778 2.22903 .259 .325 with people from the same religion q13f. I arrive at work on 54 1.00 5.00 1.6481 1.11858 2.091 .325 time because the work I do is very important to me q13g. I arrive at work on 54 1.00 7.00 4.3704 2.25954 -.155 .325 time because I enjoy the company of people from the same religion q13h. Because of 54 1.00 7.00 4.0185 2.28606 .035 .325 religious relationships, I feel that I am playing my part as a team player in this organisation q13i. I stay in this 54 1.00 7.00 4.4815 2.36117 -.257 .325 organisation because my religious relationships are personally meaningful to me q13j. I arrive at work on 54 1.00 7.00 3.2222 2.11613 .577 .325 time because the rewards I receive at work are fair Valid N (listwise) 54

The descriptive analysis for this instrumental tie (religion) variable summarises the current spread and respective descriptive statistics as respondents are asked to rate the importance of working with people who belong to the same religious organisation. As evident from the spread, 54 FPS employees in the Ministry of Lands and Mineral Resources responded to the seven-point Likert scale survey instruments (see Table 6.18). P a g e | 225

The majority of responses to the key question (I know many of the people I work with because we belong to the same religion) were recorded at around 4.9, which indicated that employees of the FPS sometimes know the religious affiliation of fellow employees. The respective standard deviation for this variable sits at around 2.3, which indicates that employees have diverse views on the question, and that is understandable for a multi-ethnic society like Fiji which is also multi-religious.

Further, the variable also recorded a negative skewness of 0.436 which highlights that

FPS workers relate to each other weakly on religious affiliation and convictions. This is again understandable because of the sensitiveness of religious affiliation and commitment as a key social capital determinant. In this regard, it is obvious that religious relationships and affiliation as a social capital determinant may not be openly discussed. The sensitivity element is assumed to trigger informal influences on HR selection decisions within the FPS.

However, responses to the question that rates ‘working with people from the same religion keeps me from leaving the organisation’ was highest at around 5.6, which shows that because of its sensitivity considering Fiji’s political climate, most respondents consider religious relationships as insignificant as a key determinant of turnover rate. The respective standard deviation for this variable sits at around 2.0 which indicate that employees had diverse views on the question. This also confirms that some employees still valued working with people with the same religious beliefs.

The above finding was also complemented through a frequency distribution analysis

(see Figure 6.5). Even though respondents to the Fiji survey acknowledged that religious relationships may influence everyday work life, it was also important to find out those distinctive religious organisations that dominated the relationships and networking. Figure

6.5 presents the dominating religious organisations for the sample Fiji ministry. P a g e | 226

As illustrated, the domination of the Methodist church had its roots in the formation of the FPS and cessation of Fiji to the UK when Fiji high chiefs adopted the Christian religion. Respondents in the sample Fiji ministry were further asked to name the religious groups they belong to (see Figure 6.5. For this social network variable, the Methodist church dominated the scores at around 45 per cent, followed by members of the Hindu religion at a modest 30 per cent. Members of the Catholic Church and SDA church recorded 5 per cent each.

Cross-examining the religious affiliation, as recorded in Figure 6.5, with ethnicity, the analysis found that ethnicity is highly significant within the religious population. In this regard, the main determinant of the strong religious affiliation recorded in this study was ethnicity. This finding proved that ethnicity, which stems from the social system and networking for Fijians and Indians, was the key ingredient for instrumental ties within the

FPS.

P a g e | 227

Figure 6.5 Religion – Ministry of Agriculture

6.5.3 Political influences

Before the last century the government of Fiji was probably patriarchal, or consisted of many independent states with little communication, and many had no political connection with each other; mutual dread tended to separate the various tribes (Ravuvu, 1987). Ravuvu confirmed that the variety of dialects spoken, the comparative ignorance of some of the kingdoms about each other, and a kind of independence in some of the smaller divisions of the same state formed the political power basis within each clan . However, with the arrival of Indians and other immigrant communities in Fiji, the struggle for social capital became a reality. Hence, the political bandwagon was associated with the other variables (high school name, political P a g e | 228

affiliation, religious grouping, ethnicity, township/city/rural community and which province).

It was the vehicle that ultimately channelled and asserted the wishes of ethnic groupings into policies, and as discussed, influenced HR selection and decisions within the FPS.

Table 6.19 presents the perceptions of FPS employees of their political affiliation. As exemplified in Appendix A, an alternate-form reliability approach was adopted to streamline the sensitivity elements around political affiliation in Fiji. In a pilot study conducted for the survey, a 95 per cent non-response rate to the political affiliation question gathered from the

FPS pilot study institution hinted that this element of the research instrument was provocative and sensitive given the current political climate in Fiji. Thus, to reinforce and strengthen the reliability of the research instruments for the major survey an alternate-form reliability approach was used.

P a g e | 229

Table 6.19 Ministry of Works Fiji – Instrumental Ties – Political

Descriptive Statistics N Minimum Maximum Mean Std. Deviation Skewness Statistic Statistic Statistic Statistic Statistic Statistic Std. Error q12a. I know many of the 75 2.00 7.00 6.1867 1.51295 -1.817 .277 people I work with because we belong to the same political party q12b. Political 75 1.00 7.00 5.5067 1.88431 -1.231 .277 relationships are good for productivity q12c. Political 75 1.00 7.00 5.9467 1.73122 -1.746 .277 relationships motivate me to get the job done q12d. Working with 75 1.00 7.00 6.2800 1.35128 -2.184 .277 people from the same political party keeps me from leaving the organisation q12e. I value working 75 1.00 7.00 5.5733 1.86122 -1.197 .277 with people from the same political party q12f. I arrive at work on 75 1.00 7.00 1.8933 1.62392 1.966 .277 time because I like to work hard q12g. I arrive at work on 75 1.00 7.00 5.6267 1.79920 -1.034 .277 time because I enjoy the company of people from the same political party q12h. Because of 75 1.00 7.00 5.8267 1.69557 -1.242 .277 political relationships, I felt that I am playing my part as a team player in this organisation q12i. I stay in this 75 1.00 7.00 5.9200 1.60034 -1.411 .277 organisation because my political relationships are personally meaningful to me q12j. I arrive at work on 75 1.00 7.00 3.5467 2.35552 .463 .277 time because my work load is fair Valid N (listwise) 75

The descriptive analysis for this key variable summarises the current spread and respective descriptive statistics as respondents were asked to rate the importance of working with people who belonged to the same political party. As evident from the spread, 75 FPS employees in the Ministry of Works responded to the seven-point Likert scale survey instruments (see Table 6.19). P a g e | 230

For the key question that, ‘they know fellow workers because of political affiliation’, the majority of responses was recorded at a mean of around 6.2. This indicated that FPS employees rarely know the political preferences of fellow workers. The respective standard deviation for this key question sits at around 1.5, which indicates that employees had diverse views on the question. This is quite understandable considering the current political climate in Fiji. In other words, people were very secretive and discreet about sensitive issues like political affiliation. The standard deviation reading also confirmed that perceptions of political preferences were more dispersed, hinting that political relationships were not prominent for FPS employees. The sample case institution here was the Ministry of Works. It is also important to note from the outcome in Table 6.19 that the key question for political affiliation displayed a highly negative skewness of 1.817. The negative skewness of the data determined how weakly the respondents related to each other in terms of their political affiliation, and that is understandable. Finally, in considering political relationships as a social capital determinant, the spread generally indicated that FPS employees may not openly discuss their political preferences, which has the capacity to influence HR selection decisions.

The above finding has also been complemented with a frequency distribution analysis

(see Figure 6.6). Even though respondents in the Fiji survey acknowledged that political relationships may influence everyday work life, it was also important to find out the political affiliations that dominate the relationships and networking of FPS employees. Figure 6.6 presents the perceptions for political affiliation for employees of the sample Fiji ministry. For this social network variable (political affiliation), members of the SDL Party recorded the highest at around 55 per cent, followed by Labour Party at 27.5 per cent. The other parties recorded the remaining 17 per cent. The SDL and CAMV political parties have been aligned with the Fijian ethnic grouping, while the NFP and Labour parties have been aligned with the

Indian ethnic grouping and network ties. P a g e | 231

In examining the political affiliation, as recorded in Figure 6.6, with ethnicity, the analysis found that ethnicity is highly significant within the political affiliation responses of

FPS employees. In this regard, the main determinant of a political affiliation recorded within this study is ethnicity. This finding proved that ethnicity, which stems from the social system and networking for Fijians and Indians, was the key ingredient for instrumental ties within the FPS. Hence, one of the major determinants for the significant Fijian population in the FPS is political affiliation.

Figure 6.6 Political affiliation – Ministry of Works

P a g e | 232

6.6 Affective ties and HR selection decisions in the Fiji Public Service

Descriptive analysis helps to boil down raw statistics into an interpretable, digestible form and was relevant in this analysis because it would ascertain whether affective ties variables do exist and, most importantly, would confirm the magnitude of their presence in terms of the sub-research question. As stated earlier, Harrington (2008) confirmed that affective ties have a high intrinsic value of relationship. She further found that an affective-based society would perceive that any relationships outside the work context have high intrinsic value.

The questionnaire responses will be used in the analysis as follows: community network ties in the Ministry of Agriculture – Fiji; family network ties in the Ministry of

Works – Fiji; and friendship network ties in the Ministry of Lands – Fiji. Using the three cases exemplifies an embedded case study for the FPS which was the basis of the survey and was complemented with archival data research. Hence, the social capital affiliations of the

FPS workers were assumed to portray their affiliation where community ties, family network ties and friendship network ties were the key influence in HR selection to FPS positions.

The descriptive analysis for community relationships on the selected sample ministry is presented in Table 6.20. The current spread and respective descriptive statistics reflect FPS employees’ perceptions when asked to rate the value of community relationships at their work place. As evident from the spread, 40 FPS employees in the Ministry of Agriculture responded to the seven-point Likert scale survey instruments (see Table 6.20). The mean statistical score for each question highlighted diverse readings and the spread ranged from 2.5

(to a considerable degree) to around 6.5 (rarely). Specifically, when employees were asked the question whether they value working with people from different communities, the average response indicated a mild level of 2.5. In other words, the average response indicated that the majority of employees were keen on working with people from different communities. P a g e | 233

Table 6.20 Ministry of Agriculture Fiji – Community Ties

Descriptive Statistics N Minimum Maximum Mean Std. Deviation Skewness Statistic Statistic Statistic Statistic Statistic Statistic Std. Error q14a. I know few of the 40 1.00 7.00 3.4500 2.08720 .448 .374 people I work with because we do belong to different communities q14b. Community 40 1.00 7.00 5.3000 1.78599 -.791 .374 relationships decrease productivity q14c. Community 40 1.00 7.00 6.0750 1.57525 -1.621 .374 relationships discourage me from getting the job done q14d. Working with 40 2.00 7.00 6.0750 1.50874 -1.501 .374 people from the same community may motivate me to leave this organisation q14e. I value working 40 1.00 7.00 2.5500 2.05002 1.156 .374 with people from different communities q14f. I am late to work 40 1.00 7.00 6.0000 1.43223 -1.819 .374 because the work I do is not very important to me q14g. I am late to work 40 2.00 7.00 6.2000 1.39963 -1.734 .374 because I do not enjoy the company of people from the same community q14h. Because of 40 3.00 7.00 5.9750 1.31046 -.887 .374 community relationships, I feel that my part as a team player in this organisation is diminished q14i. I will leave this 40 1.00 7.00 5.9500 1.31948 -1.525 .374 organisation because my community relationships are personally meaningless to me q14j. I will leave this 40 1.00 7.00 6.5000 1.19829 -3.153 .374 organisation because strangers and friends look at me critically Valid N (listwise) 40

However, for the question, ‘I will leave this organisation because strangers and friends look at me critically’, the majority of responses recorded a mean of around 6.5

(rarely). In other words, aligning turnover rate with how fellow workmates from other communities look at each other strangely is weakly rated. The respective standard deviation for this variable sits at around 1.2, which indicated that employees had diverse views for this P a g e | 234

question. This moderate standard deviation reading indicated that values were more dispersed; hence respondents presented different views on the subject matter within the question. It is also important to note from the outcome in Table 6. 20, that this particular outcome has displayed a high negatively skewness value at around -3.2. The negative skewness of the data determined how weakly the respondents related to aligning turnover rate with how fellow workmates from other communities look at each other strangely.

Finally, the spread generally indicated that the question asking whether employees knew few of the people they work with, because they belong to different communities, recorded a modest mean at around 3.5 (occasionally) indicating that employees of the

Ministry of Agriculture knew fellow workers because they belonged to the same communities. The corresponding standard deviation reading for this variable recorded a strong reading of 2.1, meaning that the majority of workers had diverse views for this question. However the majority posited a positive skewness reading at around 0.448, indicating how strongly the respondents related in terms of knowing people from other communities. In this regard, community relationships as a social capital determinant may not be openly discussed, but it is evident that it could impact HR selection policies and decisions.

Finally, the researcher also found that the methodology as exemplified was the most suitable to assess this affective tie network variable.

A social network can be defined as “a set of nodes (e.g., persons, organizations) linked by a set of social relationships (e.g., friendship, transfer of funds, overlapping memberships) of a specified type” (Laumann et al., 1978, p. 458). Acquaah and Eshun stated that the adage, ‘It is not only what you know that affects your performance, but also who you know’ sums up much of the conventional wisdom about the effects of institutional networking relationships (2010:p 669). This adage substantiates the importance of family and P a g e | 235

friendship relationships. The descriptive analysis presented in Table 6.21 summarises the FPS employees’ perception of family ties.

Table 6.21 Ministry of Works Fiji – Family Ties

Descriptive Statistics N Minimum Maximum Mean Std. Deviation Skewness Statistic Statistic Statistic Statistic Statistic Statistic Std. Error q15. I know few of the 75 1.00 34.00 4.6267 4.19386 4.636 .277 people I work with because we belong to different families q15b. Family 75 1.00 7.00 5.0667 2.27402 -.757 .277 relationships decrease productivity q15c. Family 75 1.00 7.00 5.8000 1.81585 -1.433 .277 relationships prevent me from getting the job done q15d. Working with 75 1.00 7.00 5.7067 1.88000 -1.329 .277 people from the same family may motivate me to leave the organisation q15e. I value working 75 1.00 7.00 2.4267 1.99441 1.245 .277 with people from different families q15f. I arrive at work on 75 1.00 7.00 5.4133 2.21233 -1.097 .277 time because I need to be on guard against others q15g. I arrive at work late 75 1.00 7.00 6.5333 1.15470 -3.005 .277 because I do not enjoy the company of people from the same family q15h. Because of family 75 1.00 7.00 6.3733 1.37323 -2.768 .277 relationships, I feel that I am neglecting my part as a team player in this organisation. q15i. I will leave this 75 1.00 77.00 7.4533 8.22700 8.369 .277 organisation because my family relationships are personally meaningless to me. q15j. I will leave this 75 1.00 7.00 6.5600 1.21076 -3.406 .277 organisation because family members are saying bad things about me behind my back Valid N (listwise) 75

The above spread and respective descriptive analysis reflects FPS employees’ perceptions when asked to rate the value of family relationships at their work place. As evident from the spread, 75 FPS employees in the Ministry of Works responded to the seven- point Likert scale survey instruments (see Table 6.21). The mean statistical score for each P a g e | 236

question highlighted diverse readings and the spread ranges from 2.4 (to a considerable degree) to around 6.5 (rarely). When employees were asked whether they value working with people from different families, the average response indicated a mild level of 2.4. In other words, the average response indicated that the majority of employees were keen on working with people from different families.

However, the majority of responses to, ‘I will leave this organization because family members are saying bad things about me behind my back’ recorded a mean of around 6.5

(rarely). In other words, aligning turnover rate with ‘family members are saying bad things about’ is rather weakly rated. The respective standard deviation for this variable sits at around

1.2, which indicated that employees had diverse views for this question. This moderate standard deviation reading indicated that values gathered were more dispersed; hence respondents presented different views on the subject matter because of its sensitivity. This response reading may also be attributed to the current political climate in Fiji.

The FPS employees’ perceptions of the importance of friendship relationships were also assessed as it is a key affective ties determinant. The descriptive analysis presented in

Table 6.22 summarises the FPS employees’ perceptions of friendship ties.

P a g e | 237

Table 6.22 Ministry of Lands – Fiji – Friendship Ties

Descriptive Statistics N Minimum Maximum Mean Std. Deviation Skewness Statistic Statistic Statistic Statistic Statistic Statistic Std. Error q16a. I know few of the 54 1.00 7.00 5.1667 2.05353 -.695 .325 people I work with and they are not my friends q16b. Friendships 54 1.00 7.00 5.9630 1.55361 -1.566 .325 damage productivity q16c. Having friends 54 2.00 7.00 6.0370 1.35929 -1.474 .325 around prevent me from getting the job done q16d. Working with my 54 2.00 76.00 6.9900 9.51082 7.209 .325 friends may motivate me to leave the organisation q16e. I dislike working 54 4.00 7.00 6.6667 .80094 -2.517 .325 with my friends q16f. I arrive at work on 54 1.00 7.00 5.8704 1.82277 -1.567 .325 time because I am not sure that I fully trust friendships at work q16g. I arrive at work on 54 3.00 7.00 6.5926 1.00035 -2.848 .325 time because I dislike working with friends q16h. Because of 54 2.00 7.00 6.5370 1.11106 -2.840 .325 working with my friends, I feel that my part as a team player in this organisation is neglected q16i. I will leave this 54 3.00 7.00 6.6667 .89020 -2.944 .325 organisation because working with friends is personally meaningless to me. q16j. I will leave this 54 3.00 7.00 6.7037 .79217 -3.194 .325 organisation because my friends are saying bad things about me behind my back Valid N (listwise) 54

The above spread and respective descriptive analysis reflects FPS employees’ perceptions of friendship relationships at the work place. As evident from the spread, 54 FPS employees in the Ministry of Works responded to the seven-point Likert scale survey instruments (see Table 6.22). The mean statistical score for each question highlighted diverse readings and the spread ranges from 5.1 (sometimes) to around 6.9 (not at all). When employees were asked, ‘I know few of the people I work with and they are not my friends’, the average response indicated a robust 5.1. In other words, the average response indicated P a g e | 238

that the majority of employees sometimes knew few of the people they worked with and they were their friends at the workplace.

However, when respondents were asked, ‘Working with my friends may motivate me to leave the organisation’, the majority of responses recorded a mean of around 6.9 (not at all). In other words, aligning turnover rate with ‘working with my friends’ is rather weakly rated and almost insignificant. The respective standard deviation for this variable indicated that employees almost have the same views in aligning working with friends and turnover rate; hence respondents were almost unanimous.

Answering sub-research question four (b): What are the key characteristics of instrumental and affective ties within the FPS that have the potential to influence HR selection decisions? For this part of the analysis, a summary of key arguments indicated that for instrumental ties, ethnicity was a key influence for school, religious and political affiliations within the FPS. It was also a critical component for the overwhelming majority of

Fijian FPS employees; 67 per cent in 2011. Figures also confirmed that the prominence of

Fijian FPS employees evident at the individual analysis level transcended into the operational and organisation analysis (the FPS as a whole). In this regard, ethnicity was one of the key characteristics of instrumental ties with the potential to negatively influence HR selection and decisions.

On the other hand, figures for affective ties indicated a majority of employees within the FPS were keen on working with people from different communities and family background. Further, these FPS employees highlighted that working with friends is rather weakly rated and almost insignificant. In this regard, community relationships, family relationships and working with friends are key characteristics of affective ties that have the potential to positively influence HR selection and decisions. In other words, their impact is almost insignificant. This may be due to the current political climate in Fiji. P a g e | 239

6.7 Demographics of the Fiji Public Services and Social Capital

Sub-research question four (c):

Do the demographics of employees in the FPS ministries examined correlate with a level of sustained social capital and social networking?

6.7.1 Representations of dominant ethnic groups within the FPS

Table 6.23 Ethnic population of the FPS – 1874, 2003, 2004, 2005, and 2011

Year Fijian Indian Others Total

1874 13 28 41

(Europeans)

2003 11416 6252 538 18206

2004 11685 6247 554 18486

2005 11427 5928 401 17854

2011 10284 5173 692 16149

(Fiji Civil List, 2003, 2004, 2005, 2011; Fiji Government Gazette No. 5, 1874)

As presented in Table 6.23, Fijians have been the dominating ethnic social group in the FPS. This domination began when Queen Victoria of Great Britain accepted the deed of cession of Fiji, which was supported by the high chiefs of Fiji. Those chiefs were a critical component of the formal administration of the colony of Fiji (See appendix 9.7). Hence, their role and the role of Fijians within the FPS at large began with the provisional and formal P a g e | 240

administration of the colony (Fiji Government Gazette, No. 1, 1875). The domination of

Fijians within the FPS began at 31 per cent in 1874. The ethnic population in the FPS was officially recorded by the FPS until 2005. From 2006, the current regime in Fiji declassified ethnicity as it moved towards a government of national unity, and issues like gender equality took prominence. The researcher purposively gathered the ethnic population figures in the

FPS for those significant years as stated in Table 6. 23. As stated, 1874 was the beginning of the FPS, 2003, 2004 and 2005 were the three consecutive years for this demographic variable to be presented in the FPS civil listings before its demise in 2006, and 2011 was the year of the research for the thesis. The domination of Fijians in the FPS was 62 per cent in 2003, 63 per cent in 2004, 64 per cent in 2005, and 63 per cent in 2011. On the other hand, Indians recorded 34 per cent in 2003, 33 per cent in 2004, 33 per cent in 2005, and 32 per cent in

2011. It is also important to note that the total number of public servants in Fiji decreased by

3.4 per cent from 2004 to 2005, increased by 1.5 per cent from 2003 to 2004 and decreased by 9.5 per cent from 2005 to 2011.

The First Public Service in Fiji was listed in the Civil List 1 of 1874, (Fiji

Government Gazette No. 5, 1874), the state institutions were formalised through a combination of the following government departments; which initiated the FPS and overall administration of the colony. These include: the Executive Department, Judicial Department,

Colonial Secretary’s Office, Postal Department, Police Department, Gaols, Treasury

Department, Customs Department and the Department of Native Affairs. As presented in

Table 6. 23, the racial composition of the initial FPS included 28 Europeans and 13 Fijians. It is evident Europeans were dominating government which was understandable, since Fiji was just being colonised by Great Britain (see appendix 9.7).

Table 6.24 presents the current age distribution analysis for the FPS. This analysis is important in trying to ascertain whether the age variable correlates with a level of sustained P a g e | 241

social capital and social networking within the FPS. As presented in Tables 6.14 and 6.15, a booming generation X population for both ministries was recorded. In other words, the 26 –

34 age group dominated at around 40 per cent for the organisations surveyed. At the same time, the prominence for the Fijian ethnicity was recorded at around 73 per cent for the ministries surveyed.

6.7.2 Age distribution analysis within the FPS

Table 6.24 FPS Age Distribution Analysis – 2011

Male % Female % Total 22 – 25 587 2.8 830 4 1417 26 – 30 2086 9.9 2574 12.4 4660 31 – 40 3954 18.9 4261 20.5 8215 41 – 45 1459 6.9 1117 5.3 2576 46 – 50 1038 4.9 679 3.3 1717 51 – 55 1045 5 719 3.4 1764 >55 276 1.3 261 1.4 537 Total 10445 49.7 10441 50.3 20886 (Source: Fiji Civil List, 2011)

Hence, the age variable can be assumed to correlate highly with the ethnicity variable, hence sustaining Fijian social capital and social networking within the FPS. Figures at around

67 per cent for the whole FPS population have provided evidence for this finding.

Further, the mandatory retirement age for the FPS as stipulated in State Service

Decree 2009 could also be challenged in view of this finding. In spite of the mandatory retirement regulation, 2.7 per cent of the current public service exceeds the retirement age of

55 years, but are still being employed by the FPS. The majority of those currently employed are between the ages of 31 to 40, that is 39.4 per cent. This group is followed by those P a g e | 242

between the ages of 26 to 30 at around 22.3 per cent. This can be construed as a level of sustained social capital and social networking for those supporting the current regime in Fiji.

Further, this figure may also present those competencies and subsequent designations that are either difficult to replace or are scarce within the FPS.

As presented in Table 6.25, the competency records for the FPS in 2011 closely associates the ethnic population of the FPS, the age distribution analysis and the structure of employment and gender analysis for the same period.

Table 6.25 Competency Index – Fiji Public Service – 2011

Education Level Ministry of Ministry of Ministry of Total Primary Land and Works, Industries and Survey Transport, Agriculture Energy and Public Utilities Service Exams 103 88 140 331 Certificate Level 24 18 24 66

Diploma Level 160 28 65 253

Bachelor Degree (inc. Honours) 41 40 21 102

Postgraduate (Certificate, 8 4 4 16 Diploma and Masters)

PhD 0 0 0 0

Total 336 178 254 768

(www.psc.gov.fj)

The assumption here is that the classification for competency for the FPS (Table 6.25) could be strengthened if the FPS strives for fair ethnic representations, and as is evident in

Table 6.23, Fijians have been the dominating ethnic grouping in the FPS. Hence, if regulations are enacted to promote fair ethnic representation, as exemplified in the 1997

Constitution of Fiji, the competency data records for the FPS could be greatly reinforced. In P a g e | 243

this regard, this thesis recommends that certain elements of the1997 Constitution of Fiji should be used when future constitutions are promulgated.

Section 140 (b, d) of the 1997 Constitution of Fiji reads:

The recruitment of persons to a state service, the promotion of persons within a state service and the management of a state service must be based on the following principles:

(b) appointments and promotions should be on the basis of merit;

(d) the composition of the state service at all levels should reflect as closely as possible the ethnic composition of the population, taking account, when appropriate, occupational preferences.

Lastly, and as exemplified in Table 6.26, the FPS has narrowed the disparity gap in gender representation and recorded a robust 50:50 ratio for 2011. However, this reading is also reflected in the accumulated aggregate for ethnic representation in 2011 where 63 per cent are Fijians and 32 per cent are Indo Fijians as presented in Table 6.23. Male workers for the FPS dominated the Police Force, while female workers dominated the nursing and education sectors. All in all this confirm that the inequities are ethnic rather than gender based. In Fiji, the assignment of gender roles is dictated by cultural customs for the major ethnic groups. For example the nursing fraternity is usually dominated by females whilst the police force is dominated by males. However, the ratio is gradually changing in view of adherence to EEO legislations.

P a g e | 244

Table 6.26 Structure of Employment and Gender Analysis – Fiji Public Service

(Fiji Civil List, 2011)

P a g e | 245

6.8 Conclusion

The key characteristics of instrumental ties within the FPS that have the potential to influence

HR selection decisions are quite substantial, as verified by the archival data. Fijian ethnicity is quite prominent in the latest census and commands a dominant social network through their dominant religion (Methodist Church), their preferred political party (SVT), schools (Fijian- dominated schools) and associated provinces. Hence, the correlations between the stated specific variables that control and manufacture Fijian social capital through: high school name, political affiliation, religious grouping, ethnicity and provinces respondents belong to are assumed to control who is selected for the FPS as per this sample government institution.

There has been a major shift in the increase in the Fijian population, 50.7 per cent in 1996 to

57 per cent in 2007. It is also surprising to note that a large number of Fijians have also moved offshore.

At the same time, the Indian social network through their religion, their preferred political party, schools, and associated communities are also notable. The Indian population, however, has slightly decreased due to emigration and the urge for moving to ‘greener pastures’ for improved quality of life, 44 per cent in 1996 to 37 per cent in 2007. However, table 6.3 indicates that in the last census Fijians comprised 57 per cent while Indians made up

37 per cent.

On the other hand, key characteristics of affective ties within the FPS which have the potential to marginally influence HR selection decisions through a sustained level of social capital and social networking include community relationships, family relationships and friendships. However, the analysis concluded that the influence of friendship is almost insignificant and may not exert any significant level of sustained social capital and social networking to influence HR selection decisions. P a g e | 246

CHAPTER 7

FIJI PUBLIC SERVICE EMPLOYEES’ PERCEPTION OF JUSTICE IN HR SELECTION DECISIONS

P a g e | 247

Table of Contents

7.1 Introduction

7.2 Fiji Government ministries under survey

7.3 Response rate

7.4 Analytical method

7.4.1 Data screening

7.4.2 KMO and Bartlett’s test

7.5 Justice and employee discretions in the Fiji Public Service

7.5.1 Descriptive Analysis – Fairness of current HR selection policies in your organisation

7.5.2 Factor analysis – justice and fairness of HR selection policies

7.6 Conclusion

P a g e | 248

7.1 Introduction

The analysis introduced in this chapter exemplifies the critical examination of the embedded case study labelled Case Study Three. It presents evidence of FPS employees’ perceptions of justice in HR selection decisions. Specifically, the case study presents the answers to sub- research questions five (a):

Sub-research question five (a):

Do employees perceive justice and/or injustice with the HR selection decisions of the FPS?

As for Chapter 6, the analytical framework for this chapter used a mixed methodology framework on a multiple-level analysis. Data triangulation was employed through a multiple- level analysis to answer the above sub-research questions. The levels of analysis included the

FPS organisation as a whole, ministries and departments in ministries and individual employees of the FPS.

Further, and as presented in Chapter 2, the literature review for the study was aligned with the research questions of the thesis in order to identify the research gaps that need to be empirically examined through research questions and analysis of qualitative and quantitative data of the survey. The rationale for this chapter is to present the analysis as evidence that the research gaps are significant in the formulation of contemporary HR selection theories. A critical aspect of this chapter is to ascertain the robustness of the survey instruments for a mixed method analysis. As stated earlier, the response rate for the survey (75 per cent for the

Ministry of Works, 40 per cent for the Ministry of Agriculture and 54 per cent for the

Ministry of Lands) depicted a handsome return and good representation of the sample in the population studied for the FPS.

The instruments of analysis used for the quantitative stream are descriptive and factor analysis. In the process, an inferential analysis was carried out to make inferences about the P a g e | 249

population (FPS), from which data are sampled (FPS case ministries). Hence, a key objective of this survey was to measure key hypothetical constructs: justice and/or injustice of HR selection decisions within the FPS. This would involve measuring the justice perceptions of

FPS employees regarding meritocracy in the application of HR selection decisions.

Hypothetical constructs like these are not directly observable. Instead, they can only be inferred or measured indirectly through observed variables, indicators (Kline, 1998). A key substantive contribution of the thesis is that the justice perceptions of current FPS employees were investigated. This element sufficiently covers a research gap in the existing literature which had investigated former employees (Gilliland, S. W., & Chan, D. (2001).).

Inferential, descriptive and narrative analytical systems employed in this chapter provided ample evidence that there is a significant degree of injustice in the FPS in view of government’s failure to deliver on its promises as obliged by law. In other words, the general outcome of the survey analysis in this chapter contradicted existing policies which articulate an effective and efficient methodology in management HR selection and key decisions.

It is also important to report that the survey outcomes revealed critical comments from current FPS employees which compounded and reinforced the substantive outcome for the thesis, considering that the thesis statement reads: ‘Social capital has influenced the FPS

Human Resources (HR) selection system and HR selection decisions’ These comments include: ‘The survey is very interesting to me as it entails how to be productive and how to be treated fairly in an organisation’; ‘It also highlights how the different relationships (cultural, political, family, religion influences our work in the organisation’; ‘It is no wonder, the FPS elite (high achievers) have migrated overseas because of the unfair selection of personnel to man key positions in government’; and lastly, ‘I like the survey, however most of our colleagues have left the organisation to seek greener pastures, and also due to political instability in the country. We do carry out our work for the sake of the country and our P a g e | 250

families, but still feel that we are not compensated fairly against the effort we exert’. In view of the comments, it is evident that diverse network ties have the potential to impact HR selection and work performance.

Within the study, the survey measured Fiji civil servants’ perceptions of instrumental and affective ties using a seven-point scale with scores ranging from one (1) = Very much so, to seven (7) = Not at all. Relevant studies, including Harrington (2001), were used to construct the measurement scales. To measure distributive justice, Price and Mueller’s (1986) theoretical model was used. They developed their construct through a Distributive Justice

Index on a seven-point scale with scores ranging from one (1) = Very much so, to seven (7) =

Not at all. Previous studies that used this scale had a reliability of 0.90. For procedural justice, the constructs were developed using a 15-item scale developed by Niehoff and

Moorman (1993), on a seven-point scale with scores ranging from one (1) = Very much so, to seven (7) = Not at all. Previous studies that used this scale had a reliability of 0.90. The

Interactional Justice construct which measures the quality of interpersonal treatment received, was assessed using a nine-item scale developed by Moorman (1991), ranging from one (1) =

Very much so, to seven (7) = Not at all. Previous studies that used this scale had a reliability of 0.98. Acknowledging the study by Al-Husa et al. (2008), the responses to the seven-point scales in the questionnaire were coded in SPSS 20 for Windows, using 1 for Very much so, and 7 for Not at all.

Lastly, since the survey data from the Ministry of Agriculture was more comprehensive that the other two ministries, its data was used for sub-research questions 5(a) and 5 (b). This was a limitation observed in the study.

P a g e | 251

7.2 Fiji Government ministries under survey

As evident, the three Fiji Government ministries under study are classified under the membership of the FPSC, an autonomous public service umbrella institution. The principal role of the FPSC is stipulated under the State Services Decree 2009, No. 6 as the central agency responsible for the effective management and development of human resources; organisational and systems management; and the continuous improvement of public services.

The three (3) ministries provide public services in the area of Lands and Mineral Resources;

Agriculture, and Works and Public Utilities (Energy, Roads, Communication and Transport).

7.3 Response rate

In the study, the researcher randomly distributed 100 questionnaires for each ministry and the response rate for each ministry (see Table 7.1) was recorded as follows: Ministry of Works,

Transport and Public Utilities – 75 per cent; Ministry of Lands and Mineral Resources – 54 per cent, Ministry of Agriculture – 40 per cent. With a total response rate of 56 per cent, there were 169 usable responses. The average rate of individual participation in groups responding was 70 per cent of the membership (sd. = 0.18). The high response rate was probably due to the formal process of acquiring the FPSC’s approval prior to the survey. While it was not possible to compare the sample frame for this study with the entire population of the FPS, which according to the latest census in 2011 stands at around 17819, analysis of the non- respondents indicated no difference in terms of composition, size or location compared to ministries that did participate. Zikmund et al. (2003) stated that the response rate is defined as the number of questionnaires returned or completed, divided by the total number of eligible people who were contacted or asked to participate in a survey.

P a g e | 252

Table 7.1 Response Rate

Ministry No. of No. of Response rate questionnaires questionnaires distributed returned and completed Works, Transport, Energy, Communication and Public 100 75 75% Utilities

Lands and Mineral Resources 100 60 60%

Agriculture 100 40 40%

7.4 Analytical method

Descriptive statistics was initially employed to summarise a sample, rather than use the data to learn about the population that the sample of data is thought to represent. Hence, under descriptive statistics, the analysis began by examination of relevant summary statistics with tabulations of outcomes.

Further, factor analysis was employed to explore the current data array to ascertain whether there was a relatively simple underlying structure that produced the observed associations among the 10 questionnaire items of the FPS survey. In perusing the various analytical techniques for factor analysis, the researcher decided to use principle axis factoring as the most suitable to investigate the underlying structure of the associations between the 10 questionnaire items. By reducing the complexity of the data, the reduced structures for procedural and informational justice perceptions of FPS employees would then be used collectively to answer the stated sub-research questions. P a g e | 253

Lastly, the narrative process was used to examine existing rules and regulations on

HR selection and decisions within the FPS. This was conducted to provide evidence for principles of justice when current pay and promotion conditions were examined against existing policies.

7.4.1 Data screening

Data screening needed to be performed in this analysis to detect extreme multicollinearity or singularity by creating a correlations matrix for all variables in the survey. This matrix was created to be part of the main factor analysis, and it checked for the category of multicollinearity within the variables of the survey. Field (2005) found that it is important to avoid extreme multicollinearity (i.e. variables that are very highly correlated) and singularity

(variables that are perfectly correlated). He further found that singularity causes problems in factor analysis because it becomes impossible to determine the unique contribution of the variables that are highly correlated to a factor; hence the need to eliminate any variables that don’t correlate with any other variables or that correlate very highly with other variables (R <

0.9) (Field, 2005). Field found that the determinant of the R matrix should be greater than

0.00001, and if it is less than this value, then look through the correlation matrix for variables that correlate very highly (R > 0.8), and consider eliminating one of the variables (or more depending on the extent of the problem) before proceeding.

SPSS output for the correlation matrix as exemplified shows an abridged version of the R – matrix. Table 7.2 summarises the data screening process for the analysis and presents the multicollinearity and singularity determinant for each questionnaire outcome, depicting whether the data are acceptable for further analysis.

P a g e | 254

Table 7.2 Data Screening – Factor Analysis – FPS Survey

Multicollinearity and Status Questionnaire outcome singularity determinant Ministry of Agriculture, Fiji 0.010 Greater than the necessary School Relationships : value of 0.00001, hence ok to Instrumental be used for further analysis Ministry of 0.001 Greater than the necessary Agriculture/Works, Fiji value of 0.00001, hence ok to Political Relationships : be used for further analysis Instrumental Ministry of 0.001 Greater than the necessary Agriculture/Lands, Fiji value of 0.00001, hence ok to Religious Relationships : be used for further analysis Instrumental

Ministry of Agriculture, Fiji 0.011 Greater than the necessary Community Relationships : value of 0.00001, hence ok to Affective be used for further analysis Ministry of Agriculture, Fiji 0.003 Greater than the necessary Family Relationships : value of 0.00001, hence ok to Affective be used for further analysis

Ministry of Agriculture, Fiji 0.001 Greater than the necessary Friendship Relationships : value of 0.00001, hence ok to Affective be used for further analysis

Ministry of Agriculture, Fiji 0.000 Determinant value is less Justice: Fairness of HR than 0.00001, hence should Selection Policies be checked for Multicollinearity and

Singularity issues Re-run: 0.001 Determinant value is less Ministry of Agriculture, Fiji than 0.00001, hence should Justice: Fairness of HR be checked for Selection Policies Multicollinearity and Singularity issues

As depicted, all questionnaires’ outcomes selected for the analysis were found with no evidence of multicollinearity and singularity contamination, with the exception of the P a g e | 255

outcome of question 17. This question requested respondents’ views on the fairness of HR selection policies currently used in the FPS. Because of the irregularity noted, the significance value was scanned for any reading greater than 0.05 and correlation coefficients scanned against any readings greater than 0.9. Field (2005) ascertained that if contamination is found, then it is necessary to eliminate one of the two variables causing the problem.

Hence, the outcome of question 17 was subjected to this process of data screening and the contamination was dealt with to ensure that all variables were acceptable for further analysis.

In this regard, outcomes for questions 17i and 17j were found to be contaminated, and they were eliminated due the severity of contamination. A rerun of the analysis depicted an acceptable multicollinearity and singularity determinant of 0.001 (see Table 7. 2: Re-run).

7.4.2 KMO and Bartlett’s test

The KMO and Bartlett’s test provides additional information about the factorability of the data. Specifically, the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy reports the amount of variance in the data that can be explained by the factors (Allen & Bennett, 2010).

Further, Bartlett's test of sphericity also indicates how suitable data are for factor analysis.

Allen and Bennett (2010) ascertained that if Bartlett’s test is significant (Sig < 0.05), then the data are acceptable for factor analysis. Further, Kaiser (1974) recommends accepting KMO statistic values greater than 0.5, and values below this are deemed unacceptable. He also found that values between 0.5 and 0.7 are mediocre, values between 0.7 and 0.8 are good, values between 0.8 and 0.9 are great and values above 0.9 are superb.

In view of the above, the variables for the FPS survey were then subjected to the

KMO and Bartlett’s test to ascertain their appropriateness for factor analysis. The outcomes for the tests are presented in table 7.3.

P a g e | 256

Table 7.3 KMO and Bartlett’s Test – Factor Analysis

Kaiser-Meyer-Olkin Approx. Chi-Square Bartlett's Test of Measure of Sampling Sphericity Questionnaire outcome Adequacy Ministry of Agriculture, 0.797 160.016 df – 45 Fiji School Sig – 0.000 Relationships : Instrumental Ministry of Works, Fiji 0.806 256.424 df – 45 Political Relationships : Sig – 0.000 Instrumental Ministry of Lands, Fiji 0.770 243.811 df – 45 Religious Relationships Sig – 0.000 : Instrumental Ministry of Agriculture, 0.666 157.449 df – 45 Fiji Community Sig – 0.000 Relationships : Affective Ministry of Works, Fiji 0.699 204.164 df – 45 Family Relationships : Sig – 0.000 Affective Ministry of Agriculture, 0.608 232.194 df – 45 Fiji Friendship Sig – 0.000 Relationships : Affective Ministry of Agriculture, 0.862 266.410 df – 28 Fiji Justice: Fairness of Sig – 0.000 HR Selection Policies Ministry of Works, Fiji 0.862 266.410 df – 28 Justice: Fairness of HR Sig – 0.000 Selection Policies Ministry of Lands, Fiji 0.862 266.410 df – 28 Justice: Fairness of HR Sig – 0.000 Selection Policies

As is evident, the KMO statistic values for all variables ranged from 0.608 to 0.862, depicting an average value of 0.744, which is considered good. Further, the Bartlett’s test of sphericity indicated that all variables were suitable for factor analysis with a significance P a g e | 257

value of less than 0.05. In this regard, a significant test tells us that the correlation coefficient matrix is not an identity matrix; hence there are some relationships between the stated variables in the analysis. For the presented data, Bartlett’s test was highly significant (p <

0.001), and therefore factor analysis was appropriate.

7.5 Justice and employee discretions in the Fiji Public Service

The critical approaches adopted to examine the major topic of the thesis included concentrating on justice and/or injustice with HR selection procedures and decisions.. The prime objectives of this element of the analysis was to formulate significant themes as the underlying structures that summarise the justice perceptions of FPS employees on the impact of network ties on HR selection decisions.

Sub-research question five (a):

Do employees perceive justice and/or injustice with the HR selection decisions of the FPS?

Organisational justice and attribution theory frameworks have been found to be effective in understanding the processes by which applicants perceive and react to selection procedures and decisions. Ployhart and Ryan (1997) found that behavioural intention variables used in their study were chosen because they had been studied in past research and had practical significance to organisations. This theoretical framework was used for the FPS study since the focus is on current employees’ perspectives of past and current HR selection processes. This highlights a key substantive finding for the thesis in explaining why the other categories of justice/injustice inform HR selection decisions. For example, Price and

Mueller’s (1986) distributive justice theories assessed the fairness of HR selection processes.

Further, Niehoff and Moorman’s (1993) theoretical framework assessed the fairness of HR P a g e | 258

selection processes and how well they were implemented in organisations. Moorman (1991) developed his interactional justice constructs to measure the quality of interpersonal treatment. Lastly, Gilliland and Chan’s (2001) focussed on interactional and informational justice.

As stated earlier, the data from the Ministry of Agriculture was used as a sample because of its robustness and comprehensiveness. Further, its officials were very supportive and provided a comprehensive data range including lateness and tardiness reports.

P a g e | 259

7.5.1 Descriptive Analysis – Fairness of current HR selection policies in your

organisation

Table 7.4 Descriptive Analysis – Fairness of current HR selection policies in your organisation

Descriptive Statistics – Sample Ministry (Agriculture)

N Minimum Maximum Mean Std. Deviation Skewness Statistic Statistic Statistic Statistic Statistic Statistic Std. Error q17a. I am treated fairly by 40 1.00 7.00 2.3250 1.54235 1.186 .374 my boss q17b. I am fairly rewarded 40 1.00 7.00 3.1000 1.76577 .635 .374 considering the work that I do in my organisation q17c. I am fairly rewarded 40 1.00 7.00 3.1500 1.91552 .674 .374 taking into account my education and training background q17d. I am fairly rewarded 40 1.00 7.00 3.6250 2.03416 .348 .374 taking into account my work experience background q17e. I am fairly rewarded for 40 1.00 7.00 3.7250 2.03794 .298 .374 the effort that I put into my work q17f. I receive enough 40 1.00 7.00 4.2500 1.77951 .176 .374 reward when I do a good job q17g. I am compensated 40 1.00 7.00 4.5750 1.87955 -.099 .374 fairly for the stresses and strains of my job q17h. My pay level equals 40 1.00 7.00 4.5250 1.97403 -.134 .374 the worth of my knowledge, skills and abilities (KSA) q17i. I feel that the current 40 1.00 7.00 4.3250 2.09257 -.297 .374 HR selection policies in my organisation are unbiased and non-discriminatory q17j. I feel that the current 40 1.00 7.00 3.7250 2.03794 .164 .374 HR selection policies in my organisation have been improving over time Valid N (listwise) 40

P a g e | 260

As evident from Table 7.4, respondents were asked critical questions about whether they perceived justice and/or injustice in the current HR selection decisions of the FPS. The responses indicated how the fairness of current HR selection policies in the FPS correlated with HR selection and decisions. In other words, the discussion was taken from a broader point of view to ensure that the linkages between justice and HR selection were better understood.

The justice and/or injustice elements were directed towards fair treatment by the organisation in regards to designation responsibilities. The descriptive analysis for this key variable summarised the current spread and respective descriptive statistics as respondents were asked to rate the fairness of HR selection policies within the FPS sample ministry – Fiji

Ministry of Agriculture. As evident from the spread, 40 FPS employees in the Ministry of

Agriculture answered the questions on a seven-point Likert scale. The mean statistical score for each question highlighted diverse readings with the spread ranging from 2.3 (to a considerable degree) to around 4.3 (sometimes). Specifically, when employees were asked to rate the question, ‘I am treated fairly by my boss’, the average response indicated a mild level of 2.3. In other words, the average response indicated that to a considerable degree, the majority of employees felt that they were treated fairly by their work managers. However, others felt that they had not been fairly treated at all.

For the question, ‘I feel that the current HR selection policies in my organisation are unbiased and non-discriminatory ’, the majority of responses recorded a mean of around 4.3

(sometimes) which indicated that there were biases and discriminatory issues within the FPS

HR selection decision policies. The respective standard deviation for this variable sits around

2.0, which indicates that FPS employees had the same sort of views for HR selection decision policies. . It is also important to note in Table 7.4, that this particular outcome displayed a moderate negative skewness value at around -0.29. The negative skewness of the data P a g e | 261

determined how moderately the respondents related to such issues in view of the sensitivity of the question considering Fiji’s social and political climate in the year of research (2011).

Other crucial questions of this survey instrument gathering FPS employees’ perceptions on rewards had similar outcomes with a mean at around 3.6

(occasionally/seldom). Even though FPS employee perceptions on reward is not a core issue for the sub-research question under examination, it is however important to briefly highlight their perceptions on HR selection decisions in regards to policies on pay and reward. The questions included: ‘I am fairly rewarded considering the work that I do in my organisation’,

‘I am fairly rewarded taking into account my education and training background’, ‘I am fairly rewarded taking into account my work experience background’, ‘I am fairly rewarded for the effort that I put into my work’; and ‘I receive enough reward when I do a good job’. This indicated that FPS workers have been unjustly treated in terms of fairness of rewards against work effort and competencies.

Finally, when workers were asked the question ‘I feel that the current HR selection policies in my organisation have been improving over time’, the outcome recorded a modest mean at around 3.7 (seldom), indicating that employees were hinting that HR selection policies within the FPS have hardly improved. This is also reflected in the corresponding skewness measure at around 0.16, meaning that the FPS employees positively relate to this finding.

P a g e | 262

Table 7.5 Fairness of HR selection processes and how well they are

implemented in your organisation

Descriptive Statistics Sample Ministry (Agriculture)

Std. N Minimum Maximum Mean Deviation Variance Skewness

Std.

Statistic Statistic Statistic Statistic Statistic Statistic Statistic Error q18a.I feel I was fairly treated during the 40 1.00 7.00 2.6000 1.85085 3.426 1.264 .374 selection procedure q18b. The current HR selection processes employed in my organisation 40 1.00 7.00 2.8500 1.74753 3.054 1.030 .374 encourages a wide representation of all age groupings in Fiji q18c. The HR selection panel in my organisation are made up of highly 40 1.00 7.00 3.1250 1.66699 2.779 .666 .374 competent individuals q18d. During my interview, I was offered 40 1.00 7.00 2.9000 1.87835 3.528 .861 .374 opportunities to express my views q18e. During my interview, I was offered 40 1.00 7.00 3.6500 2.08228 4.336 .386 .374 opportunities to influence outcomes q18f. I feel that the HR selection processes employed in my organisation 40 1.00 7.00 2.7000 1.43581 2.062 1.053 .374 are used consistently with people of all ethnic groupings q18g. I feel that the HR selection methods employed in my organisation 40 1.00 7.00 3.4000 1.89195 3.579 .629 .374 are neutral and free of bias q18h. I feel that the HR selection outcome in my organisation is based on 40 1.00 7.00 3.3000 1.85638 3.446 .751 .374 accurate information q18i. The current HR selection methods employed in my organisation include 40 1.00 7.00 3.2750 2.30926 5.333 .593 .374 mechanisms for appeal q18j. The current HR selection methods employed in my organisation encourage 40 1.00 7.00 2.6750 1.68534 2.840 .949 .374 a wide representation of all ethnic groupings in Fiji

Valid N (listwise) 40

P a g e | 263

As shown in Table 7.5, respondents were asked critical questions regarding the fairness of HR selection processes in the FPS. The FPS employees’ perceptions of the fairness or otherwise of HR selection processes were directed towards the HR selection procedures and competencies of the selection panel.

The descriptive analysis for this key variable summarised the current spread and respective descriptive statistics as respondents were asked to rate the fairness of HR selection policies within the sample ministry - Fiji Ministry of Agriculture. As is evident from the spread, 40 FPS employees in the Ministry of Agriculture fully answered the questions on a seven-point Likert scale. The mean statistical score for each question highlighted diverse readings with the spread ranging from 2.6 (occasionally) to around 3.7 (seldom). Specifically, when employees were asked to rate the question, ‘I feel I was fairly treated during the selection procedure’, the average response indicated a mild level of 2.6 (see table 7.6). In other words, the average response indicated that ‘occasionally’ a considerable number of employees felt that they were treated fairly during the HR selection processes and decisions.

On the other hand, some felt they had not been fairly treated at all.

Table 7.6 Fair treatment during selection procedures

q18a.I feel I was fairly treated during the selection procedure

Cumulative Per Frequency Per cent Valid Per cent cent

Valid Very much so 14 35.0 35.0 35.0 To a considerable degree 11 27.5 27.5 62.5 Occasionally 6 15.0 15.0 77.5 Seldom 3 7.5 7.5 85.0 Sometimes 1 2.5 2.5 87.5 Rarely 2 5.0 5.0 92.5 Not at all 3 7.5 7.5 100.0

Total 40 100.0 100.0

P a g e | 264

When subjected to a frequency distribution analysis, 15 per cent of the total population indicated that ‘occasionally’ they felt that they were treated fairly during HR selection procedures, as compared to ‘very much so’ and ‘seldom’. This reading also reflects the mean statistical reading for the whole population in the survey.

However for the question, ‘during my interview, I was offered opportunities to influence outcomes’, the majority of responses recorded a mean of around 3.7 (seldom) which indicated that applicants were seldomly asked of their opinion during HR selection interviews. The respective standard deviation for this variable sits around 2.0, which indicated that employees had the same views when asked whether their opinions were sought during selection interviews. It is also important to note in Table 7.5, that this particular outcome displayed a moderate positive skewness value at around 0.386. The positive skewness of the data indicated that the FPS employees considered the current HR selection procedures in the FPS as very critical for which they relate to positively

Further, when workers were asked whether ‘the HR selection panel in my organisation are made up of highly competent individuals’, the outcome recorded a mean at around 3.1 (occasionally), hence the competencies of members of the FPS HR selection panel on average needed to be evaluated. In view of the mediocre reading, each ministry and department within the FPS needs to assess the current skills and/or qualification of its selection panel members.

P a g e | 265

Figure 7.1 Competencies of HR selection panel

A frequency analysis depicted in Figure 7.1 supplements the above findings and confirms that ‘occasionally’ employees believed the HR selection panel within the FPS were not up to par with prerequisite qualifications for an HR selection panel.

This critical outcome also aligned with another question which assessed whether ‘the current HR selection methods employed in FPS organisation encourage a wide representation of all ethnic groupings in Fiji’. The outcome for this question recorded a reading of 2.6 which means that ‘occasionally’ the current HR selection methods employed encourage representation of all ethnic groupings in Fiji. This could also be interpreted that the current

HR selection methods don’t encourage representation of all ethnic groupings in Fiji.

However, as reported earlier in the thesis, the current regulations encourage a non-bias and P a g e | 266

fair ethnic representation. This would mean that the root cause of the problem may lie within the application of those fine regulations, and also be linked to the discretionary power of FPS managers.

7.5.2 Factor analysis – justice and fairness of HR selection policies

The following analysis was also carried out to supplement the descriptive analysis carried out earlier for the research question: Do employees perceive justice and/or injustice with the HR selection decisions of the FPS?

. The expectation within the analysis are not just about HR selection, but tailored to include justice and fairness of HR selection policies. In other words, the discussion takes a broader point of view to ensure that the links between justice, fairness and HR selection policies and decisions are better understood.

In addition to the descriptive analysis a factor analysis was conducted to explore the current data on the justice and fairness of HR selection policies within the FPS. The exercise was conducted to ascertain whether there is a relatively simple underlying structure in terms of the general perceptions of FPS workers of the fairness of HR selection policies.

Specifically, the factor analysis would provide the theoretical framework for the key themes through data reduction process and should relate the extent of justice and fairness of current

HR selection policies within the FPS.

P a g e | 267

For this exercise, questionnaire 17 was again employed to reinforce the consistency within the analysis and outcome considering that the overall objective of this questionnaire was for applicants to rate the fairness of current HR selection policies in their organisation. Before conducting factor analysis, all questionnaire variables needed to be tested for their suitability and also to test whether factor analysis was suitable for the exercise.

Table 7.7 Ministry of Agriculture Fiji – Justice and Fairness of HR Selection Policies

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .862 Approx. Chi-Square 266.410 Bartlett's Test of Sphericity df 28 Sig. .000

As is evident, the KMO statistic values for all variables measuring the fairness of HR selection policies in FPS institutions indicated an excellent reading of 0.862. Further,

Bartlett's test of sphericity indicated that all variables in this measure were suitable for factor analysis, with a significance value of less than 0.05. For the presented data, Bartlett’s test is highly significant (p < 0.001), and therefore factor analysis was appropriate for this measure.

It should be also noted that contamination was contained for this measure, and that has also contributed to its high significance. Further, it has also avail the suitability of the specified data for further analysis. Specifically, the KMO and Bartlett’s test play an important role in accepting the sample adequacy (Preacher & MacCallum, 2003). Further, the Bartlett’s test of sphericity relates to the significance of the FPS and thereby showed the validity and suitability of the responses to the problem being addressed.

P a g e | 268

Table 7.8 Total Variance Explained

Factor Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadingsa Total % of Variance Cumulative % Total % of Variance Cumulative % Total 1 5.561 55.609 55.609 5.247 52.467 52.467 5.227 2 1.487 14.871 70.480 1.082 10.817 63.284 1.614 3 .933 9.327 79.807 4 .570 5.696 85.503 5 .465 4.652 90.155 6 .414 4.138 94.293 7 .192 1.917 96.210 8 .185 1.847 98.057 9 .126 1.261 99.318 10 .068 .682 100.000 Extraction Method: Principal Axis Factoring. a. When factors are correlated, sums of squared loadings cannot be added to obtain a total variance.

Table 7.8, shows the amount of variance that can be explained by the factor analysis.

Figures indicated that there were two strong underlying factors identified during the analysis.

The first highlights the initial eigenvalues which show the amount of variance accounted for by each variable, which are ordered from highest to lowest. In our example, the first factor explains the highest portion of variance at 55.6 per cent, while the second factor explains at least 14.8 per cent. As evident from this analysis, two factors will be retained. After rotation, factor 1 now explains 52.5 per cent of the variance and factor 2 explains 10.8 per cent of the variance. The total variance explained a significant variance reading of 63.3 per cent, which depicts the total variance that can be explained by these two factors. The two factors have recorded eigenvalues greater than 1. These two factors will provide the key themes in the data reduction process which should relate to the extent to which selection procedures are honestly and adequately explained by HR managers in the FPS.

P a g e | 269

Figure 7.2 Justice and Fairness of HR Selection Policies

The scree plot in Figure 7.2 confirms the variance analysis shown in Table 7.8, confirming that within the analysis, two factors will be extracted and retained. The scree plot graphically demonstrates the distribution of variance among the components to confirm in our analysis that a two-component solution is adequate. Specifically, these two factors will provide the theoretical framework for the key themes in the data reduction process which should relate to the extent to which selection procedures are honestly and adequately explained by HR managers in the FPS.

P a g e | 270

Table 7.9 Justice and Fairness of HR Selection Policies

Pattern Matrixa Factor 1 2 q17a. I am treated fairly by my boss .577 .176 q17b. I am fairly rewarded considering the work that I do in .869 .009 my organisation q17c. I am fairly rewarded taking into account my education .940 -.171 and training background q17d. I am fairly rewarded taking into account my work .940 -.052 experience background q17e. I am fairly rewarded for the effort that I put into my .879 -.018 work q17f. I receive enough reward when I do a good job .804 -.026 q17g. I am compensated fairly for the stresses and strains .759 .021 of my job q17h. My pay level equals the worth of my knowledge, skills .543 .100 and abilities (KSA) q17i. I feel that the current HR selection policies in my -.061 .817 organisation are unbiased and non-discriminatory q17j. I feel that the current HR selection policies in my .088 .680 organisation have been improving over time Extraction Method: Principal Axis Factoring. Rotation Method: Oblimin with Kaiser Normalization. a. Rotation converged in 3 iterations.

The 10 items of significance in justice and fairness of HR selection policies were subjected to factor analysis (principal axis factoring) using SPSS Version 21. Factor analysis revealed the presence of two factors with eigenvalues exceeding 1 explaining 52.5 per cent and 10.8 per cent of the variance respectively. An inspection of the scree plot revealed a clear break after the second component, and using Catell’s (1966) scree test, it was decided to retain two factors. To aid in the interpretation of these two factors, Oblimin with Kaiser Normalisation was performed. The rotated solution, as depicted above, revealed the presence of a simple structure (Thurstone, 1947), with both factors showing a number of strong loadings and all variables loading substantially on only one component. The two-factor solution explained a total of around 63.3 per cent of the variance, with factor 1 contributing 52.5 per cent and factor 2 contributing 10.8 per cent. P a g e | 271

A factor loading for factor 1 was much stronger than factor 2. In view of cross loadings experienced within the analysis, it would be prudent to highlight the prominent element with the two loadings as they depict the key driver which shapes up the model.

For factor 1, two elements recorded a value of 0.940 and the elements are: ‘I am fairly rewarded taking into account my education and training background’ and ‘I am fairly rewarded taking into account my work experience background’. In other words, this element can be termed ‘Fair reward in view of KSA and work experience’.

For factor 2, two elements also recorded significant readings at around 0.817 and

0.680. These were, ‘I feel that the current HR selection policies in my organisation are unbiased and non-discriminatory’ and ‘I feel that the current HR selection policies in my organisation have been improving over time’. In other words, this element can be termed

‘Fair and improving HR policies’.

From the analysis, it is obvious that most employees of the FPS were concerned about factor 1 (Fair reward in view of KSA and work experience) and factor 2 (Fair and improving

HR policies), even though more emphasis was placed on factor 1.

Table 7.10 Summary of Analysis Justice and Fairness of HR Selection Policies

Factor Theme

1 Fair reward in view of KSA and work

experience

2 Fair and improving HR policies

P a g e | 272

As presented in Table 7.10, it is obvious that most employees in the FPS consider factor 2

(Just and improving HR policies) as critical in their work life whilst factor 1 (Fair reward in view of KSA and work experience) was also important. T.

7.6 Conclusion

The descriptive and factor analysis jointly deduced that at times employees perceive injustices in the selection decisions of managers in the FPS. In the descriptive analysis, workers are hinting that there is slight degree of bias and discrimination in the application of

HR policies. This is taking into account that State Services Decree 2009 has provided the most unbiased HRM policy to date for Fiji. However, the regulations on Terms and

Conditions of Employment for GWE in Fiji need to be revised to reflect the strategic remunerations of the FPS aligned with an environment of political stability. The final analysis in this chapter confirmed that evidence for biased and discriminatory HR policies were present within the FPS. Further, outcome from this analysis have also positively recorded that

FPS HR selection panel and board members are underqualified.

P a g e | 273

CHAPTER 8

DISCUSSION AND POLICY IMPLICATIONS

P a g e | 274

Table of Contents

8.1 Introduction

8.2 Sub-research question one: What are the HRM challenges facing the FPS and what are the implications on HR selection decisions?

8.3 Sub-research question two: What are the critical HR selection strategy challenges facing the Fiji Public Service in view of HR selection decisions?

8.4 Sub-research question three: Do the HR selection policies of the Fiji Public Service contribute to the concepts of fit and flexibility to develop a highly qualified HR pool for the FPS?

8.5 Sub-research question four (a): Could the key relations of the dominant social groups in Fiji that build and maintain social capital influence HR selection decisions within the FPS?

8.6 Sub-research question four (b): What are the key characteristics of instrumental and affective ties within the FPS that have the potential to influence HR selection decisions?

8.7 Sub-research question four (c): Do the demographics of employees in the FPS ministries examined correlate with a level of sustained social capital and social networking?

8.8 Sub-research question five (a): Do employees perceive justice and/or injustice with the HR selection decisions of the FPS?

8.9 Conclusion

P a g e | 275

8.1 Introduction

The prime objective of this chapter is to highlight the connections between the central research question of the thesis: How does social capital through network ties influence HR selection and decisions in the Fiji Public Service?, and the findings in the qualitative and quantitative analysis. Specifically, nine sub-research questions drive the thesis statement that,

‘Social capital through network ties influence HR selection and decisions within the Fiji

Public Service’.

Following are the sub-research questions developed from the literature review:

Sub-research question one:

What are the HRM challenges facing the FPS and what are the implications on HR selection decisions?

Sub-research question two:

What are the critical HR selection strategies challenges facing the Fiji Public Service in view of HR selection decisions?

Sub-research question three:

Do the HR selection policies of the Fiji Public Service contribute to the concepts of fit and flexibility to develop a highly qualified HR pool for the FPS?

Sub-research question four (a):

Could the key relations of the dominant social groups in Fiji that build and maintain social capital influence HR selection decisions within the FPS?

Sub-research question four (b):

What are the key characteristics of instrumental and affective ties within the FPS that have the potential to influence HR selection decisions? P a g e | 276

Sub-research question four (c):

Do the demographics of employees in the FPS ministries examined correlate with a level of sustained social capital and social networking?

Sub-research question five (a):

Do employees perceive justice and/or injustice with the HR selection decisions of the FPS?

In answering the central questions that drive the thesis, the chapter developments have been organised to correspond to the literature review in view of the above mentioned sub research questions, in chapters five, six and seven.

As outlined in Figure 8.1 and transposed from the main research question of the thesis, one of the hidden objectives was to unravel the sensitivity around the impact of social capital on the HR selection policies and practice of the FPS. The outcome may open up avenues for future research for the different streams and themes associated with social capital and network ties and their impact on HRM, particularly HR selection.

P a g e | 277

Figure 8.1 Social Capital and HR Selection policies and practice of the FPS

Social Capital Management of the 1. Instrumental Ties Fiji Public Service 2. Affective Ties

HR Selection Policies and Future Research Practices (HR Selection)

One of the key themes of this chapter is to ascertain whether the processes shown in

Figure 8.1 are aligned with the main research question of the thesis. Further, the ensuing policy implications will address avenues to tighten regulations for the management of critical

HRM work functions, particularly HR selection in the FPS. A crucial outcome for the FPS would be the adoption of an autonomous institutional framework to ensure insulation from political infiltration (Nankervis et al., 2006). Autonomy would then control the impact of social capital in its two main forms as stated in the thesis. However, critical HRM work functions, particularly HR selection, would be contextualised within the FPS to ensure that the fibre and fabric of key policies are developed and strengthened.

The findings from both the qualitative and quantitative analysis confirmed that social capital through network ties impacted the HR selection policy and practice of the FPS. P a g e | 278

Particularly, perceptions of FPS employees confirmed that instrumental and affective ties were found not to directly support HR selection policies, but have the tendency to affect the application of those policies. The final outcome of the thesis purports to reflect that social capital through network ties do affect the FPS HR selection policy and practice. However, unravelling the ‘silent connections’ between social capital through network ties and HR selection policy and practice provided a sound theoretical base from which future themes and streams of study may eventuate.

The policy recommendation of this thesis also begins to unravel a contemporary approach to managing HR selection within the FPS. Di Milia (2004) conducted a study aimed at reviewing contemporary approaches to management of HR selection in Australia and to test and prove the proposition that HR selection practice has changed. He stated that

“management selection practice has the potential to make a strategic contribution to organisational performance of government departments and institutions in the public sector”

(2004: p 214). This is a crucial finding that will unearth the potential in the current survey for policy implications in the FPS.

The selection for the thesis topic recognised the need to improve and enhance the calibre of the competencies of the FPS, formulating a personnel replacement model so that those ministries surveyed would incorporate it into their strategic management policy approach. In this regard, HR selection is a key role for HR managers (Morehead et al., 1997).

When 322 HR managers were asked to identify the most important policies and programs for the previous five years, HR selection processes were ranked third highest (Fisher and

Dowling, 1999).

This chapter summarises the thesis proposition to confirm or disconfirm that, ‘Social capital through network ties influences HR selection and decisions within the Fiji Public

Service’. The findings to the sub-research questions not only confirm the thesis proposition P a g e | 279

but within the HR selection discipline, they can be applied in a developing country within the

South Pacific region prone to political and social instability. It is understandable that wherever stability is evident, the confidence level of stakeholders increases inducing social equity which promotes affirmative action for the effective and efficient management of key institutions in the private and public sectors. Consequently, the desired productivity and worker performance levels in the public service are achieved. Managers of the FPSC should take heed of the conventional trend that in times of political normality and stability, the evolution of government personnel systems has shown a transition from political patronage and personalism to a system based on merit, political neutrality, continuity, integrity and professionalism. This chapter outlines a road map that will upgrade to and maintain consistency of optimal public service production and delivery.

8.2 Sub-research question one:

What are the HRM challenges facing the FPS and what are the implications on HR selection decisions?

The key HRM challenges facing the FPS included mimicking Western HRM practices, pressure for public service reform from donor agencies and countries and changing the mindset of Fiji public servants. Confirming the literature findings and providing evidence for the research question, the FPS doesn’t have a choice but to mimic Western ideologies and ready-made public service policy directions in view of three key indicators. These are the management of the new republic as Fiji, the cultural constraints exerted by the two dominant races on HR selection in the FPS and, lastly, the fragility of Fiji’s social, political and economic environment aggravated by a period of political instability which left FPS policymakers no alternative but to ensure that the state institutions are operational. P a g e | 280

Further, evidence provided in the literature confirmed that the political and economic instability in Fiji also exacerbated the situation constraining public service reform initiatives.

Because of the need to survive and operate the state institutions, this pressure-cooker situation catapulted Fiji civil service managers into implementing reforms dictated by

‘change agents’, and their reforms mostly include the application of HR systems which have no theoretical basis, hence are not conducive to embracing the merits of public service reform.

A significant policy driver and implication for the FPS would be to decrease reliance on leading economies and technocrats by building and reinforcing an FPS national competency profile data bank managed by the FPSIM. An example of this would be to formulate policies to increase the minimum qualification requirement (MQR) for all technical and administrative designations and the FPSIM would be the ideal institution to not only recommend but run courses and programs for each designation. This would be, of course, after consultations with existing institutions like the University of the South Pacific, Fiji

National University and other external institutions like the University of New South Wales

(UNSW) and the Australian National University (ANU). These universities are prominent institutions for HR selection research in the Asia Pacific Rim and have also been prominent in public service reforms for Oceania.

Further, the current HRM challenges could be managed efficiently through the process of standardisation. Storey & Sisson (1990) confirmed that by convention, standardisation establishes a technical standard, which could be a standard specification, standard test method, standard definition or standard procedure. Thus, standardising the HR selection process of the FPS would invariably mean that there is a standard specification, a unit of instruction that is understood across government departments and ministries managed by the FPSC. Further, by using standardisation, government departments and ministries can P a g e | 281

easily communicate through set guidelines, in order to maintain focus. Hence, one of the prime objectives of standardisation is to eradicate overlapping of policies and conflicts, which is currently evident within FPS departments and ministries. However, the existence of a published standard does not necessarily imply that it is technically correct. On this note, policymakers and practitioners who use the item or service have the responsibility to consider the available standards, specify the correct one, enforce compliance, and use the item correctly; hence validation of suitability is necessary (Storey & Sisson, 1990).

In New Zealand organisations, the greatest change over the past seven years was reported to be the standardisation of HR selection processes (Taylor et al., 1987).

Correlational evidence linking HR selection practice with business outcomes is important here. Terpstra and Rozell (1993) reported high correlations (0.70-0.80) between structured interviews, mental ability tests, biodata and higher annual profits, profit growth and sales growth. Huselid et al. (1996) reported small (0.10–0.16) but significant correlations between global indicators of HR capability (including some specific selection items) and return on assets and profitability.

A critical policy change is required for the standardisation and enforcement of the

MQR for entrance into a designated FPS gazetted position. This would entwine current and future civil servants in Fiji. In this regard, inculcating the notion that change is inevitable and that HRM practitioners are becoming more professional worldwide, the FPS needs to address the shortcoming in contrast to other developing countries of the world. Dowling and Fisher

(1999) surveyed 2795 members of the Australian Human Resources Institute (AHRI) in 1995 and compared the data to a similar survey in 1985. The 1995 data indicated an increased use of the term ‘HR’, a similar level of undergraduate qualifications (34%) and a three-fold increase in graduate qualifications (23%). In the FPS context, the enforcement of the P a g e | 282

requirement of a national MQR has to be implemented across all bands of entry point into civil servant designations.

Further, evidence in the literature also confirmed that key HRM challenges derive from the belief that social capital reinforces kinship and ethnicity status to control HR selection within the FPS. When entering the FPS, public servants from the two dominant ethnic groups transfer to FPS beliefs through their ethnic group’s SCI ideologies, hence mindsets are preconditioned. A key policy implication for this finding is that the current regime has already begun the processes of dissolving and disintegrating the cornerstone for the two dominant ethnic groups. It has begun with the Fijians, and in terms of consistency should move towards the Indian social structures and systems in Fiji. However, key policy advice from the thesis is that the pace of the disintegration processes be relaxed so that it is aligned to the contextualisation and processes of change within the core Fijian and Indian social systems. Hence, the consequences of its aftermath may be smooth to ensure that it does not unnecessarily harm stakeholders within the Fiji social structure. An important anticipated outcome from this recommended process would be gradually changing the mindset of FPS current and future employees.

As per the results and outcome of the research on the FPS and assessment of the current mechanisms in view of this notion of strategic policy change, current HRM challenges could be managed from within. Hence, the following FPS key reform policy drivers are recommended in view of Porter’s (1985) and Storey and Sisson’s (1990) findings

(see Table 8.1).

P a g e | 283

Table 8.1 Fiji Public Service Reform Strategy Focus

FPS Strategy Focus Recommended HRM Strategies

a) Jobs that require close interaction and coordination among groups of individuals b) Performance appraisals that are more likely to reflect long-term and group-based achievements c) Jobs that allow employees to develop skills that can be used in other positions in the firm d) Compensation systems that emphasise internal Innovation equity rather than external or market-based equity e) Broad career paths to reinforce the development of a broad range of skills a) Relatively fixed and explicit job descriptions b) High levels of employee participation in decisions relevant to immediate work conditions and the job itself c) A mix of individual and group criteria for interdependent behavior Quality Improvement d) Performance appraisals that are mostly short-term and results oriented e) A relatively egalitarian treatment of employees and some guarantees of employment security f) Extensive and continuous training and development of employees a) Relatively fixed and explicit job descriptions that allow little room for ambiguity b) Narrowly designed jobs and narrowly defined career paths that encourage specialisation, expertise and efficiency Cost Reduction c) Short-term, results-oriented performance appraisals d) Close monitoring of market pay levels for use in making compensation decisions e) Minimal levels of employee training and development

Adapted from Porter (1985) and Storey and Sisson, (1990)

FPS managers need to fully grasp that HRM has continually evolved over the years to become more strategic. The classifications have also evolved, from personnel management to

HRM to people management, and policies need to be modified to suit the evolution and respective needs of the public of Fiji.

All in all, current FPS employees should be encouraged by the FPS, and preferably the FPSIM, to undertake extensive training and development programmes in undergraduate and graduate studies. The policy drive is that there should be a shift in the initiation of personnel development from the employee to the employer. This would enable the FPS to P a g e | 284

undertake continuous employee competency assessment through competency needs assessment and development. As employees are up-skilled, they would certainly be filling the void in the institutions. The FPS institutions would also be using a highly centralised personnel replacement chart (PRC) (see Appendix 9.6, Figure 4.1) to keep track of competency development and management. The chart shows the present performance and the likelihood of promotion for each potential candidate.

8.3 Sub-research question two:

What are the critical HR selection strategies challenges facing the Fiji Public Service in view of HR selection decisions?

The critical events in Fiji that impacted HRM in the FPS substantiated the HRM challenges, as illustrated in Figure 5.6 (P177), and continued to shape the theoretical framework for HRM in the FPS. The analysis through an explanatory effects matrix shows that the HR selection strategies challenges facing the FPS over the years were exacerbated through the promulgation of mainstream legislation which always supported ethnic and raced-based HR selection into the FPS. The determinants of these mainstream legislation changes emanated from an unstable social environment where the two dominant ethnic groupings and their networks always wanted control of and full status in the administration of the government and bureaucracy.

Ethnic tensions have also been responsible for distorting the merits, disrupting the

FPS’s autonomy and shattering the rationality of HR selection decisions. The above were evident in the HR selection policies for the FPS when Prime Minister Rabuka came into power after the first military coup in 1987 and authorised the promulgation of the 1990

Constitution of Fiji. This key regulatory framework exemplifies an ethnic and raced-based P a g e | 285

policy preference for Fijians and Rotumans in HR selection into the FPS. Specifically, this constitution regulates the selection of not less than 50 per cent Fijians and Rotumans and not less than 40 per cent of the members of other communities. As evident in the analysis of this thesis, Fijians’ supremacy within the population of the FPS has always been alarming. For example, in 2011, the Fijian population in the FPS was recorded at almost 67 per cent.

The analysis also found that when the ethnic and raced-based policy was removed due to public outcry, its replacement, the 1997 Constitution of Fiji also had elements of ambiguity, for example: ‘(d) the composition of the state service at all levels should reflect as closely as possible the ethnic composition of the population, taking account, when appropriate, of occupational preferences.’ In other words, it encourages the dominant races and others to cling to their social capital as they enter the FPS.

The explanatory effects matrix as exemplified in Figure 5.6 found that the ensuing policy changes after Commodore Frank Bainimarama took over the democratically elected government in 2006, the State Services Decree (2009) – Part 6, Section 9, was considered the best by stakeholders. When analysed, there is no evidence of discrimination in the legislation as it provides a leeway for the enhancement of meritocracy in HR recruitment and selection.

However, its institutional legitimacy has been challenged (Kotobalavu, 2012), in view of the different definitions by key stakeholders.

The critical findings for this thesis also confirmed that the fundamental principles or established precedents that regulate HR selection policies, the ‘push’ factor for regulatory changes, was emanating from the social capital power base for the two dominant races, and formally transferred across to influence the bureaucracy through the democratic processes and parliamentary representation. Further, it could also be argued that the ‘push’ factor for regulatory changes may have also been transferred across informally as FPS employees exerted their social capital networking within their respective institutions. P a g e | 286

The key HRM challenges facing the FPS included: obtaining and building theoretical frameworks to guide policy formulation in the FPS; choices about particular HR policies that best achieve organisational fit; and building HR and KSA that are difficult to copy.

Considering that HR selection is a stand-alone function in HRM, it can’t actually be critically examined in isolation if practitioners need to measure its effectiveness; hence FPS managers need to understand the internal fit syndrome which encompasses aligning a firms HR practice with each other (Baird and Meshoulam, 1989). This would enable the particular HR function to affect strategy to enable a fit between the FPS business strategy and its HR strategy.

However, this exercise would also consider an external fit requiring aligning a firms HR practice with its strategy.The HR strategic policy choices for the FPS need to consider its internal and external environmental factors. As evident within the political climate of Fiji in the recent decade, the RFMF’s control on HR selection within the FPS has drastically affected HR policy choices as the military appoints its own people to assume strategic designations with the FPS.

In trying to build HR and KSA for the FPS that are difficult to copy and obtain, this study is initially recommending the differentiation between administrative and technical officers employed by the FPS. In this regard, policies that administer distinctive HR forecasting and planning; HR selection; compensation; work performance and appraisals; and training and development should be formulated. This initiative should be tasked to a decentralised Management Improvement Division (MID) of the FPSC, and the tasks would expect to be conducted by the individual ministries’ MIDs. Further, the MQR classifications could be reclassified to competency modelling, which is a job analysis method that identifies the necessary worker competencies for high performance.

The onus is for FPS executives to incorporate relevant SHRM theories into workable policies. Another important lesson from the work of Lengnick – Hall et al. (2009) for the P a g e | 287

FPS, is that for SHRM to thrive, it must remain sensitive to shifts in the business environment where commercialisation has been seen as bridging the gap between SHRM in the private and public sectors. In its endeavour to improve public service delivery, the current regime in Fiji formulated a strategic framework for change using key milestones to direct and drive the public service machinery. However, as a policy initiative, this thesis is recommending an assessment of the effectiveness and efficiency of its application with a view to ascertaining its relevancy to guide Fiji out of continuous instability and uncertainty.

A key strategic policy decision is downsizing and FPS managers should always ensure that the motive for downsizing should be related to shifts in the business environment of the FPS.

In this regard, concrete data analysis depicting accurate relativity of HR strategising to economic growth phases must dictate and defend the rationale for downsizing or otherwise.

On that note, FPS managers should be actively involved in the planning process as ‘strategic partners’ at the corporate level of decision-making with FPS customers, integrating FPS business strategies with the strategic directions of the Fiji Government. They should also have the required competencies to perform as strategic partners in the delivery of exceptional service, and must prove that they are capable of performing diverse task and roles within their

HR executive designations.

8.4 Sub-research question three:

Do the HR selection policies of the Fiji Public Service contribute to the concepts of fit and flexibility to develop a highly qualified HR pool for the FPS?

Key findings for the above sub-research question confirmed the constraints faced by the FPS which hinder its strategic approach to attaining the concepts of fit and flexibility.

These included choices about particular HR policies that best achieve organisational fit, P a g e | 288

building theoretical frameworks to guide policy formulation for the FPS and the importance of clarity in HR policies on selection and worker performance; (See figures 5.7 and 5.8).

It is advisable that in view of the political instability in which the FPS is operating,

HR selection practice strategies should be cautiously and carefully aligned to the changes in external and internal environmental factors. In this regard, HR selection practitioners in the

FPS may vary strategies along a number of dimensions, therefore considering a number of strategic typologies to apply. The rationale for this ideology is for the FPS to implement HR reform policies relative to the economic and political phases of development of Fiji, and not to be too aggressive given the merits and fibre of organisational characteristics. This is a key lesson for the FPS.

Further, findings concluded that it would be conducive for the FPSC and its managers to frame and build a theoretical framework to guide its HR selection policy formulation. This daunting tasks needs to be bestowed on a FPSIM. The following institutions could be formally requested to assist the set-up of the FPSIM: University of the South Pacific, School of Business, UNSW@ADFA, and the University of Waikato, Management School. These institutions have to operate in parallel with the FPSC, and be tasked to undertake theoretical assessments as bases for policy formulation and public service reform initiatives, guiding the strategic directions of the FPS in determining the right size and structure of the civil service through appropriate consultation and reviews. The University of the South Pacific, School of

Business, UNSW@ADFA, and the University of Waikato, Management School were chosen because the researcher for this thesis began contemplating the ideology for the research intent in these institutions. It would also be appropriate for the FPSIM concept to expand the current institution, to the centre for training and development (CTD), formerly the government training centre (GTC). P a g e | 289

Lastly, findings also confirmed that for the FPS operational or functional HRM do exist to ensure that the current regime continues to manage state institutions and the bureaucracy; however the strategic aspects of HRM may not be evident. This is due to economic, political and societal constraints as exemplified in Figure 5.7 (p 184). The central idea behind the strategic aspects of HRM is that all initiatives involving how people are managed need to be aligned with and in support of the “organisations” overall strategy

(Mello, 2011) see Figure 5.8. Given the current political instability, the FPS may not be able to achieve this substantive outcome. Hence, the context of current HR policies ascertained that the mainstream regulations are crystal clear; however ministerial and departmental regulations are heavily infiltrated, hence causing ambiguities within the context of the policies that may create confusion, duplication and complications in their application.

8.5 Sub-research question four (a):

Could the key relations of the dominant social groups in Fiji that build and maintain social capital influence HR selection decisions within the FPS?

The key relations derived from the analysis initially included the relationship of the dominant social grouping as shown in the latest census of Fiji in 2007. Figures gathered from the census indicated that Fijians recorded 56.8 per cent of the total population of Fiji while

Indians recorded 37.5 per cent. Meanwhile, people of other ethnicity recorded around 5.7 per cent. It is also important to note that while the census confirmed an increase in the total population of Fiji by approximately 8 per cent, considering 1996 as the base year, the Fijian population increased by around 20 per cent, while the Indian population decreased by 7.4 per cent. P a g e | 290

It is common knowledge in Fiji that the Indians have been dominating the business sector of Fiji. One has to only wander through the streets of the major cities in Fiji to notice their ownership of businesses, including shops, modes of public transportation and other forms of business. This research couldn’t gather empirical evidence on this; however, a minor example showed the dominance of Indians through their social capital grouping within the business community.

In regard to representation in government and Cabinet, the domination of Fijians was clear. For example, Appendix 9.7 presents the first Civil List of Fiji in 1874, where the total number of public servants was 47. The analysis found that around 28 per cent of these were

Fijians, while 72 per cent were Europeans. The domination of Europeans are understandable because Fiji had just been ceded to the British Crown, and all forms of government administration were controlled and managed by Britain.

Further, in the election of the Council of Ministers in 1967, the ethnic representation was 30 per cent Fijians, 50 per cent Europeans and 20 per cent Indians. The analysis showed that the domination of Europeans was slowly decreasing, while the representation of Fijians increased to around 30 per cent. In this Cabinet line-up, the Indians emerged at around 20 per cent.

In 1972, the dominance of Fijians continued and the pinnacle was reached when a high chief of Fiji was made Prime Minister. In this Cabinet line-up, 58 per cent were Fijians,

21 per cent were from the Indian community and 21 per cent were from the European community. It was also noted that the Fijian representation increased to 58 per cent, while the

Indians recorded 21 per cent, an increase of about 1 per cent.

In 2011, and as exemplified in Table 6.9, the ethnic composition for the Cabinet line- up in Fiji recorded an overwhelming majority of Fijians at around 70 per cent while Indians P a g e | 291

recorded 30 per cent. This was a surprising given that the government of Frank Bainimarama was advocating a multinational government to eradicate any form of racial and ethnic preference. However, he implemented contradictory tendencies in his indigenous development strategies, where he demonised affirmative action openly, but on the other hand he reinvented it under the rubric of rural development (Ratuva, 2013).

As is apparent, even though the Fijians dominated the national population of Fiji and were well represented in government and the bureaucracy, the Indian population also had a solid foundation backed by their sound economic position in the society. Hence, it is obvious that the strong representation of the dominant social groupings and their networking have the potential to influence HR selection decisions within the FPS.

8.6 Sub-research question four (b):

What are the key characteristics of instrumental and affective ties within the FPS that have the potential to influence HR selection decisions?

This finding confirmed that the key characteristics of instrumental and affective ties within the FPS have the potential to influence HR selection decisions within the FPS. A summary of key arguments indicated that ethnicity was a key characteristic for the instrumental ties variables: school, religious and political affiliations within the FPS. Hence, it was the binding factor and the most critical determinant that aggregated instrumental tie variables in confirming the dominance of Fijians within the FPS. Figures gathered from the analysis confirmed this overwhelming majority of Fijians within the FPS’s total population at around 67 per cent in 2011.

As exemplified in Figure 6.4 (see p222), and at the individual analysis level, the high schools that dominated the readings in the analysis were ethnic Fijian schools, totalling P a g e | 292

around 30 per cent. This verified the dominance of the school network within the Fijian population of the FPS, and its magnitude reflects its potential to influence HR selection and decisions within the FPS. The Indian ethnic group was also distinctively represented throughout the survey.

Further, a key instrumental tie variable is politics and, as exemplified in Appendix

9.6, a 95 per cent non-response rate to the initial pilot study question, ‘What political party are you affiliated with?’, prompted the researcher to use an alternate-form reliability approach to strengthen the reliability of the research instruments for the major survey within the FPS. In using alternate-form reliability, the exercise took care of provocative and sensitive elements around the questions given the current political climate in Fiji; hence the revised question for the survey reads: ‘What political party did you vote for in the last election?’ The outcome for the major survey indicated a handsome response rate for this question. The survey analysis also recorded that for this key instrumental tie variable, members of the SDL Party recorded the highest response at around 55 per cent, followed by

Labour Party at 27.5 per cent. The other parties recorded the remaining 17 per cent. The SDL and CAMV political parties have been aligned with the Fijian ethnic grouping, while the NFP and Labour Party have been aligned with the Indian ethnic grouping and network ties. This reading again confirmed the dominance of Fijians boosted by their political affiliations.

Lastly, a critical instrumental tie variable is religious organisations. The survey requested respondents to state their religion. The ensuing analysis, as illustrated in Figure 6.5, confirmed the domination of the Methodist Church at around 45 per cent, followed by members of the Hindu religion at a modest 30 per cent. Members of the Catholic Church and

SDA recorded 5 per cent each. As stated earlier, the domination of the Methodist Church has its roots in the formation of the FPS and cessation of Fiji to the UK when Fiji high chiefs adopted the Christian religion. P a g e | 293

In cross-examining school, religious, and political affiliation with ethnicity, key findings confirmed that ethnicity is highly significant within the school, religious and political affiliation for the FPS population. It is the main determinant that binds instrumental tie variables (school, political and religious) within the FPS population under survey. Hence, ethnicity which stems from the social system and networking for Fijians and Indians has been the key ingredient for instrumental ties within the FPS.

On the other hand, figures for affective ties indicated that a majority of employees within the FPS were keen on working with people from different communities and family backgrounds. Further, these FPS employees highlighted that working with friends is rather weakly rated and almost insignificant. In this regard, community relationships, family relationships and working with friends are key characteristics of affective ties that have the potential to positively influence HR selection and decisions. In other words, their impact is almost insignificant. This may be due to the current political climate in Fiji, where issues in regard to family, friendships and community relationships are highly sensitive. A key recommendation would be to run the survey again when Fiji is more stable.

Further, and according to the survey, respondents were more flexible in answering instrumental rather than affective ties questions. The rationale behind this could be the same as mentioned earlier, since current FPS employees are discreet in discussing issues that involve community, family and friendship. However, in cross-examining community, family and friendship ties with ethnicity, the findings deduced that ethnicity influences have been insignificant, even though ethnicity is a key determinant of family. On that note, FPS employees were rather non-ethnic affiliated when it comes to choosing friends and also when it comes to assessing the importance of belonging to one’s community.

A critical determinant of the insignificance of ethnicity towards affective ties variables would be the dominance of heterogeneity variables, including age and KSA within P a g e | 294

the FPS. As recorded the sample ministries analysed, indicated a booming generation X population, those in the 26 – 34 age group. The trend in this age group is that most are willing to cross the ethnic divide to portray a new crop of public servants. Further, as illustrated in

Figures 6.1 and 6.2, the total schooling years for current FPS employees was around 15 to 18 years, indicating that most FPS employees would have completed high school and either enrolled or completed their tertiary education. This again displayed a high level of KSA would be limiting the dominance of ethnicity for affective ties and networking.

8.7 Sub-research question four (c):

Do the demographics of employees in the FPS ministries examined correlate with a level of sustained social capital and social networking?

The heterogeneity variables examined in this study reflect the demographics of employees in the FPS population. These included the representation of dominant ethnic groups within the FPS, age analysis and KSA/competency for current employees.

The domination of the Fijian ethnic group in the FPS recorded 62 per cent in 2003, 63 per cent in 2004, 64 per cent in 2005, and 67 per cent in 2011. On the other hand, Indians recorded 34 per cent in 2003, 33 per cent in 2004, 33 per cent in 2005 and 32 per cent in

2011. It is also important to note that the total number of public servants in Fiji decreased by

3.4% per cent from 2004 to 2005, increased by 1.5 per cent from 2003 to 2004 and decreased by 9.5 per cent from 2005 to 2011. The finding concluded that ethnicity correlates highly with the level of sustained social capital and social networking of the two main ethnic grouping: Fijians and Indians. This is reflected in the ethnicity growth rates for both dominant groups. P a g e | 295

The current age distribution analysis for the FPS is illustrated in Table 6.14. In trying to ascertain whether the age variable correlates with a level of sustained social capital and social networking within the FPS, the analysis revealed a booming Generation X population for both ministries. In other words, the 26 – 34 age group dominated at around 40 per cent for the organisations surveyed. At the same time, the prominence for the Fijian ethnicity was recorded around 73 per cent for the ministries surveyed. Hence, the assumption here is that age may be a crucial factor that has been sustaining the dominance of the Fijian ethnicity within the FPS. However, this finding differs with earlier discussions when age is cross- examined with affective ties.

This part of the finding eventuated after KSA/competency was cross-examined against ethnicity. The general assumption here is that the classification for competency data records for the FPS (Table 6.25, p242) could be strengthened if the FPS strives for fair ethnic representation, and as evident in Table 6.24, Fijians have been the dominating ethnic grouping in the FPS. Hence, if regulations are enacted to promote fair ethnic representation, as exemplified in the 1997 Constitution of Fiji, the KSA data records for the FPS may be reinforced and strengthened.

8.8 Sub-research question five (a):

Do employees perceive justice and/or injustice with the HR selection decisions of the FPS?

In examining the above sub-research question, the analysis focussed on the outcome for key questionnaire instruments. These included assessing the fairness of current HR selection policies within the FPS, assessing the fairness of HR selection processes and how well they are implemented in the FPS and fair treatment during selection procedures. The latter includes the competency level for HR paneland board members within the FPS. P a g e | 296

When respondents’ perceptions were assessed in regard to the fairness of current HR selection policies in the FPS, the findings confirmed that employees were divided. Some felt they were treated fairly by their work managers, while others felt that they had not been fairly treated at all. Hence, a decisive majority outcome was not recorded, hinting that the current

HR selection policies within the FPS are not fair to all employees. . All in all, for employees of the sample Fiji ministries there were slight degree of bias and discrimination in the current

HR policies.

The findings also confirmed that occasionally considerable number of employees felt they were treated fairly during the HR selection processes and decisions. On the hand, others felt they have not been fairly treated at all. Hence, the FPS still needs to address the fairness of how current HR selection processes are implemented in this critical HR function.

An additional issue highlighted was the fair treatment during selection procedures. In this key assessment instrument, the majority of employees felt they were treated unfairly during HR selection processes. For example, the majority of responses indicated that they were seldomly asked their opinion during HR selection interviews. Further, a general consensus confirmed that the FPS needs to urgently upgrade the qualification base line for their HR selection panel members. These HR selection panel members would also need to be trained specifically and comprehensively in HR selection.

All in all, the general consensus of FPS workers indicated a perception of injustice with the HR selection decisions of the FPS. The key lesson for the FPS would be to immediately streamline the HR function of selection and its major legislation. Further, they need to analyse the effectiveness of current HR selection policies with other line ministries and departments to ascertain whether there is a need to change policies and decentralise this critical HR work practice. P a g e | 297

8.9 Conclusion

In view of the current political instability facing the FPS and its stakeholders, the findings confirmed that revival of the justice perceptions for FPS workers is critical to ensure that inefficiencies are minimised and controlled. This will reinforce current predictors of exceptional performances, which will improve the robustness of the HR selection system.

However, given the current social and political environment in Fiji, sound decision-making is critical for the implementation of HR selection policy choices that best achieve organisational fit.

The crucial HRM challenges facing the FPS have been exacerbated by the era of social and political instability which has constrained public service reform initiatives, and distorted elements of rational decision-making and management. However, in view of the need to survive and operate state institutions, FPS managers had to implement reforms dictated by ‘change agents’. These public service reform initiatives included the application of Western HRM practices and pressure for public service reform from donor agencies and developed economies. The challenge of changing the mindset of FPS workers is an integral component for efficiency and effectiveness in the management of state institutions. However, the current period of political and social instability may hinder this crucial HRM challenge.

The current status of these challenges within the FPS may have implications for HR selection decisions. These would include bypassing the meritocracy requirements of the fine legislation, including the State Service Decree 2009. As is evident, the challenges exacerbated by the current period of political and social instability have also made the application of good legislation an arduous task. This is evident with the application of the mandatory retirement age policy as stipulated within State Service Decree 2009. P a g e | 298

Findings of this thesis also confirmed through the explanatory effects matrix (Figure

5.6) that HR selection strategy challenges facing the FPS over the years have being predetermined through the promulgation of mainstream legislation which has supported ethnic and raced-based HR selection into the FPS. The key determinants emanated from an unstable social environment in view of the clashes between the two dominant ethnic groupings and their network in Fiji.

It was also confirmed in this thesis that in view of the current period of political instability for Fiji, the FPS may not be able to achieve the alignment of HR selection policies and practices with the stated business outcomes of the organisation. Hence, developing a highly qualified HR pool for the FPS would be a challenging task indeed. The thesis also ascertained that the mainstream regulations are crystal clear; however ministerial and departmental regulations are heavily distorted and fatigued by unnecessary interventions, causing ambiguities that may create confusion and complications in their application.

Further, the current degree of social capital for the dominant races should be continually assessed, since political instability may be a prime determinant for the magnitude of social capital for each dominant ethnic grouping. The findings in this thesis confirmed that ethnicity is a critical determinant for instrumental tie variables (school, politics and religion), hence the magnet and common denominator in uniting those variables to assert the dominancy of the Fijian ethnic group and its social capital through networking. It is also the main determinant of Indian social capital, even though the influence of the Fijian social capital through instrumental ties and networking has been overwhelming. At the outset, affective ties’ influence and its relationship with ethnicity has been marginalised, since it has no direct control with friendship and members of the same community. However, it has a direct influence on family relationships, but in view of the current political situation in Fiji, it tends to shy away at the moment. P a g e | 299

The magnitude for the correlation outcome has to dictate the texture of HR selection policies applied. The analysis suggested that as stability settles in, a social capital conducive to the management of the FPS will be the antecedent, and the individual social capital for the dominant races will begin to diminish. In this regard, the newly moulded social capital will embrace a multi-ethnic and diverse Fiji and complement fair and unbiased HR selection regulations. An outcome which is invaluable to society will be the production of social capital that reinforces economic and political development. In other words, distinctive social capital variables including high school name, political affiliation, religious grouping, ethnicity, township/city/rural community and the provinces respondents belong to would be viewed positively as complementary variables to social capital building. These variables have the potential to positively influence not only who gets selected into the FPS ministries, but also work performance and productivity issues. Within the FPS, mangers would then be aggressive in the choice and implementation of HR selection policies as the environmental preconditions would be favourable.

A study of reforms in the Fiji public sector by Reddy et al. (2004) identified several aspects of policy initiatives that contributed to a large public sector, particularly the public service. They found that the Fiji Government has been adopting reform policies mostly based on protection of domestic industries and a large public enterprise sector. The current military regime in Fiji also tried to tackle a large public service operating expenditure by lowering the retirement age to 55 years in 2007. However, they were not able to accurately assess the net gains from that exercise because turnover rate figures were significant, resulting in loss of invaluable working experience and competencies in the FPS. Even though it has been the responsibility of governments to ensure that unemployment figures are managed and contained, the current regime in Fiji did not measure real figures to depict loss. However, even since the implementation of this strategy in 2007, the Fiji Government has not been able P a g e | 300

to deliver a growth rate to match the increasing domestic demands and formulate a robust civil service with the potential to reinforce excellent public service delivery. Hence, this is a critical challenge for Fiji public service managers.

Further, as the current regime in Fiji introduced a policy initiative to reduce the retirement age of public servants to 55 years, they did not accurately forecast the relative rate at which school leavers and trained personnel enter the Fiji labour market. In other words, irrespective of the implementation of a compulsory retirement age, the ratio of retirement to entry for the FPS was dictated by the new entrants to the public service. In view of this, the government would always have a large public sector and consequently a larger public debt.

However, the policy advice of this thesis is that managers of the FPS need not be too extravagant or aggressive in the implementation of the SHRM theories. They need to vary personnel practices as a function of organisational characteristics, particularly the aggressiveness of the application of personnel practices needs to be aligned with texture and rigour of the stability of the external environment. Hence, critical HR policy strategic choices for the FPS have to be carefully scrutinised and then applied.

P a g e | 301

CHAPTER 9

CONCLUSIONS

P a g e | 302

Table of Contents

9.1 Introduction 9.2 Conclusions about the research questions

9.3 Conclusions about the research problem

9.4 Implications for theory

9.5 Implications for policy and practice

9.6 Limitations

9.7 Implications for further research

9.8 Conclusion

P a g e | 303

9.1 Introduction

The central research question of this thesis reads: ‘How does social capital through network ties impact HR selection policy and practice in the Fiji Public Service?’, Thus the thesis statement for this survey reads: ‘Social capital through network ties influence HR selection policies and decisions within the Fiji Public Service’. Driving the thesis statement, the following sub-research questions were designed considering the literature review of the study, and the historical narratives that exemplifies the Fiji context of the thesis.

The sub-research questions are:

 what are the HRM challenges facing the Fiji Public Services and what are the

implications on HR selection decisions?

 what are the critical HR selection strategy challenges facing the Fiji Public Service in

view of HR selection decisions?

 do the HR selection policies of the Fiji Public Service contribute to the concepts of fit

and flexibility to develop a highly qualified HR pool for the FPS?

 could the key relations of the dominant social groups in Fiji that build and maintain

social capital influence HR selection decisions within the FPS?

 what are the key characteristics of instrumental and affective ties within the FPS that

have the potential to influence HR selection decisions?

 do the demographics of employees in the FPS ministries examined correlate with a

level of sustained social capital and social networking?

 do employees perceive justice and/or injustice with the HR selection decisions of the

FPS?

P a g e | 304

The answers to the above sub-research questions summarise the general perceptions of FPS employees in view of the influence of social capital through networking on the FPS

HR selection systems, policies and, particularly, selection decisions. If the findings suggest that the HR selection system has a significant level of social capital and social network influence, then the FPS needs to accord priority and urgently reform HR selection and recruitment policies. On the other hand, it may also need to ascertain the benefits of social capital in public service operations, and acknowledge its formal use in the FPS.

However, if the findings of this survey suggest that the level of social capital and social network influence on HR selection decisions is highly insignificant, then it may be ascertained that the FPS’s HR selection processes has established autonomy and insulated its operational elements from political and social infiltration. In other words, the specific outcome of HR selection is strategically aligned with the FPS’s business environment. With this rationale as a guide, a pilot study was initially conducted, followed by the major survey of three Fiji government departments.

Hence, the prime objective of this concluding chapter is to either confirm or disconfirm the thesis statement. This would be justified when the connections between the central research question of the thesis: How social capital through network ties impact HRM policy and practice in the Fiji Public Service?, and the findings from the qualitative and quantitative analysis are confirmed.

9.2 Conclusions about the research questions

As explained within the findings, sub-research questions one, two, three, four (a), four (b), and four (c) were formulated from the literature review Chapter 2 of the thesis, and examined the following key pillars of the thesis: human resource management and the FPS; Human P a g e | 305

resource selection and the FPS; the strategic approach of HRM and the FPS; social capital and the FPS; networking ties and justice and employee discretions in the FPS.

Examining HRM in the FPS, it was envisaged that the best point to begin the investigation would be to ascertain whether the current challenges as depicted in Figure 5.2 are substantiated. Findings confirmed that the critical events in Fiji that impact HRM in the

FPS (Figure 5.1) substantiated the critical HRM challenges illustrated in Figure 5.2. In other words, the HRM challenges depicted in Figure 5.2 are survival strategies for the operation of the FPS and the bureaucracies. The justification for this survival strategy package has been exacerbated by periods of continuing political and social instability.

In examining HR selection within the FPS, the key outcomes linked themselves to

‘survival strategies for the operations of the FPS and the bureaucracies’, as found in examining sub-research question one. However, the critical finding is that, for the FPS, meritocracy has always been challenged since the FPS has not been properly mandated for full autonomy and insulation from political interference in its application of rules and regulations. Hence, the application of critical HR function policies, including job evaluation and HR selection, would be distorted because the uninsulated FPS ministries are vulnerable to the influence of social capital through networking.

Examining the strategic approach of HRM in the FPS, the thesis confirmed that the

FPS needs to turn away from depending on ‘survival packages’ as mentioned earlier and enhance the contextualisation of contemporary and relevant HR work practices which confirm the fit or otherwise between the FPS strategic directions and its stated business outcomes. The findings chapter confirmed certain strategic policy directions for making rational decisions about HR policies that best achieve organisational objectives. In the process, it is assumed that unnecessary infiltration and ambiguities may diminish. P a g e | 306

In assessing the sub-research question that examined social capital within the FPS, the major findings summarised the following substantive outcomes. Initially, Fijian ethnicity was quite prominent for instrumental ties due to a dominant social networking of their religion

(Methodist Church), their preferred political party (SVT), and their school networks (Fijian- dominated schools). Hence, the correlation between the stated instrumental ties variables bonded by ethnicity reinforced Fijian dominance over the years. In this process, ethnicity, a heterogeneity variable has been highly influencing instrumental ties, hence affecting employee reactions to perceptions. On the other hand, affective ties within the FPS have been marginally influencing HR selection decisions and policies due to the sensitivity surrounding its reactions with ethnicity. However, it has a tremendous potential to correlate highly with age and KSA, which may influence employee reactions to perceptions.

Lastly, when employee perceptions of justice/injustice of HR selection decisions were analysed, workers hinted that there was some degree of bias and discrimination in the application of HR policies within the FPS.

9.3 Conclusions about the research problem

The major research problem for this study reads: ‘Social capital through network ties influence HR selection policies and decisions within the Fiji Public Service’. In view of the conclusions stated in section 9.2, the thesis now confirms that social capital through network ties influence HR selection policies and decisions in the FPS. However, the degree of influence varies as a result of the control of heterogeneity variables including ethnicity; age and KSA (see Figure 1.0). The most influencing variable has been ethnicity. This conclusion confirms the theoretical framework of the study, and ultimately defends the initial declaration of the thesis. P a g e | 307

9.4 Implications for theory

This thesis has also so far produced a published article titled: ‘HR Selection Distortions: A

Theoretical Framework for the Fiji Public Service,’ (2012), Journal of US – China Public

Administration, ISSN 1548 – 6591, 9 (7) pp. 775-790.

In addition to the literature review (Chapter 2), the analysis in terms of the stated sub- research questions also took into consideration the critical events in Fiji’s history that provided the narrative and shaped the HRM theoretical background in Fiji, particularly the

FPS. It is anticipated that these narratives would be the basis for formulating HRM in Fiji, which has not been done before; hence it is a substantive contribution to theory on HRM in

Fiji. In this regard, it is anticipated that this thesis will act as the major vehicle to initiate the development of a theoretical framework for HRM in Fiji and other Pacific Island countries.

In Fiji’s very peculiar case, periods of continuing political and social instability have indirectly formed a unique HRM theoretical framework through a strategic survival package that has gone against the rationality of contemporary HRM work practices.

An additional theoretical implication from sub-research question two is that the Fiji

HRM theoretical framework has been also shaped by uncontrolled infiltration from social capital and networking since independence in 1970.

A theoretical implication from the findings in sub-research question three for the FPS would be to reform and contextualise current HR work policies considering the interests of its key stakeholders and this has to be aligned with the realities regarding its social, economic and political environment. A new strategic direction would then be formulated in line with the realistic internal and external elements to confirm the fit or otherwise.

In sub-research questions four (a), (b) and (c), a major substantive contribution to theory confirmed the influence of ethnicity on instrumental ties variables (school, politics and P a g e | 308

religion). The decisive influence of Fijian ethnicity is a substantive finding in its own right.

However, records from the study support the same theory for Indian ethnicity in the FPS.

The potential contribution for sub-research questions five (a) may support the initiative to fully utilise those excellent HRM regulations in place in the FPS. Within a developing country like Fiji, the application of fine legislation is still a crucial restraint in the management of public services institutions. Hence, key implications for theoretical development would be to contextualise the fibre and constructs of policies with the realities of the environment so that it may be well received, understood and managed. This exercise should be subjected to all FPS HR selection policies and also recommended to be implemented across other work functions. This policy recommendation may address and contain degrees of bias and discrimination in the application of HR policies within the FPS and in other SPI countries. From the discussion so far, it is clear that equity and consistency are the key pillars for HR selection decisions, as the dynamic nature of the FPS internal and external environments continue to distort the employer/employee interface. In addressing the issues of fairness and equity, organisations should be seen to be acting in accordance with the principles of distributive and procedural justice.

9.5 Implications for policy and practice

The political and economic stability of Fiji is conducive to relevant reform initiatives of the

FPS following trends in social and economic development. Hence, in view of this finding, it is crucial that the FPS managers target and address the situation that generates perceived inequity (for example, through dysfunctional behaviour, such as leaving the organisation, resignation, unofficial absence, refusing to work or cooperate with those employees who are P a g e | 309

perceived as over-rewarded), by formulating the desired policy levels to address inequity and reinforce motivation levels.

The thesis recommends the following policy changes within the FPS in view of the need to manage the impacts of differing characteristics of Fiji’s social structure in the HR selection system of the FPS. For the current regime in Fiji, it is recommended that in due course, preferably the post 2014 general elections era, all operational aspects of government and state, should focus on efficiency, effectiveness and higher work performance.

A significant policy driver and implication for the FPS would be to decrease reliance on leading economies and ready-made policies by consultants, but be guided by FPSIM to formulate a relevant HR theoretical framework, against which policies should be developed.

Of particular importance is HR selection, recruitment and worker performance, from which

HR selection should be based. For example, in New Zealand organisations, the greatest change over the past seven years was reported to be the application of consistent policies of

HR selection processes (Taylor, Keelty and McDonnell 2002). Of importance is some correlational evidence linking HR selection practice with business outcomes. Terpstra and

Rozell (1993) reported high correlations (0.70-0.80) between structured interviews, mental ability tests, biodata and higher annual profits, profit growth and sales growth. Huselid,

Jackson and Schuler (1997) reported small (0.10–0.16) but significant correlations between global indicators of HR capability (including some specific selection items) and return on assets and profitability.

Supporting the literature, the data provided evidence that the current policy objectives, strategies and key performance indicators for the FPS needs to be guided and assessed through the FPSIM and other supportive institutions which will improve accuracy of decision-making and strategy formulation. Further, since practitioners and academics can’t ascertain the precedent ideology or the preconditions to base the development of theories to P a g e | 310

guide the relevant work practices, a policy initiative from this perspective is that policy objectives, strategies and key performance indicators for the FPS be guided through some theoretical HRM base.

Secondly, HR policy strategic choices for the FPS need to consider its internal and external environmental factors. As evident within the political climate of Fiji in the recent decade, the RFMF’s control of HR selection has influenced HR policy strategic choices for the FPS. The key policy change initiative would be that if the democratically elected government (post 2014) ascertains that it is conducive to appoint military officers in senior

FPS designations, then this internal factor should be considered as critical for the FPS’s

HR policy strategic choices. What is ideally recommended for this critical factor could be a cadetship program similar to the one being designed at UNSW@ADFA where midshipmen and officer cadets enroll in undergraduate programs in Arts, Business, Engineering, Science and Technology (www.defence.gov.au/ADFA/Training/). Further, the PSC also needs to consider the labour market in Fiji including the tertiary institutions and turnover rates as key determinants of HR policy strategic choices.

On the other hand, the FPS must identify those external influences that will impact on the organisation and the management of its human resources. Some of the major influences include; laws and regulations regarding employment, cultural, demographic, social values and factors relating to industrial relations and the quality of work life.

In trying to build HR and KSA for the FPS that are difficult to copy and obtain, this study is initially recommending the differentiation between administrative and technical officers employed by the FPS. In this regard, distinctive policies that administer HR selection be formulated for core and supportive functions within the PSC. It is evident that current policies on most HRM sub-functions employed by the PSC do not differentiate administrative and technical competencies and their formal application (see appendix 9.3, 9.4 and 9.5). P a g e | 311

Hence, the recommendations are that current policies on HR selection and most probably recruitment be revised to distinctively reflect and highlight the differentiation. Further, a policy initiative for the FPS would be the formulation of comprehensive ‘human resource packages’ which should detail; the place of the job within the organizational hierarchy, a HR plan, the appropriate job analysis framework, the recommended HR selection process and the procedure for making the final decision, a performance management and appraisal system for the job, a recommended compensation and benefits package, and the recommended HR audit process, taking into account the HR climate within the organization. These HR packages should be designed to reflect the cyclical nature of the FPS’s business cycle considering its status within the Pacific region and with other comparable economies.

As a policy initiative, this thesis is also recommending an assessment of the effectiveness and efficiency of current regulations so as to ascertain those that may maintain the operations of state institutions until such time when political instability fades itself out of Fiji’s social and political environment and guide Fiji out of the doldrums of continuous instability and uncertainties. A key strategic policy decision would be to ensure that employee retention level is maintained at desirable level. On that note, FPS managers should be actively involved in engaging strategic partners at all levels of decision making with current FPS employees. The key policy initiative here is for the FPS to incorporate its business strategies with the strategic directions of the Fiji Government, whilst considering the welfare of its employees. In the long run, it is anticipated that these employees may become strategic partners in the delivery of exceptional service within the FPS.

The political and economic stability of Fiji is conducive to relevant reform initiatives of the FPS following trends in social and economic development. Findings from this key research question confirmed that the strategic approach to HRM for the FPS is highly dependent on stable correlations between the economic, social, political, organisational and P a g e | 312

cultural context in Fiji. The outcomes of these correlations are invaluable for FPS managers because understanding the variability of environmental conditions would enable the alignment of specific HRM systems with the desirable organisational context.

It may also be advisable to formulate an optimal combination of core competencies

(KSA) for different designations within the different ministries of the FPS. As a preliminary initiative, all designations within Fiji ministries would be differentiated into two key divisions: technical and administrative. In this regard, the specifications of the key performance indicator benchmark would be matched with the worker performance outcomes for the specific designation holder.

Further, in view of the intangible aspects of services, the input and output performance indicators for FPS institutions must be quantifiable if performance and productivity are to be measured accurately. Instead of using a five-point Likert scale to summarise the annual performance review, it would be advisable to apply the five-point

Likert scale to the elements of the job description, managed by a HRM database through a relevant statistical analytical package. Hence, an FPS human resource management information system (HRMIS) needs to be set up as an independent unit within the FPS to control and monitor HR job evaluation, job analysis and particularly HR selection.

For the FPS managers, the critical issue that often arises in HR selection in view of selecting new employees or selecting a current employee for promotion is that: Are they measuring correct competency or work performance readings against the key performance indicators (KPI) the competencies of the incumbent? It may also be advisable to incorporate both phases depending on the character of the job content and specifications needed.

The diversity of the services rendered from the FPS ministries also means that each function needs to develop its own worker performance assessment measures for the P a g e | 313

distinctive tasks of their ministries or departments. Further, the distinctive ministerial or departmental worker assessment measures should be aligned against a national framework on worker performance for the FPS. The issue for such cases is that without due flexibility, it is obvious that the worker could apply for sick leave (with or without sick leave form). In this regard, the importance of an absence management policy is suggested by this thesis to be an important deterrent to high absenteeism among Fiji public servants.

Further, it is recommended that for a forty hours working week, an additional key deterrent policy would be to shift the daily starting time for FPS workers from 8am to 9am as practiced in many leading economies of the world today. This would stabilise the “home – work” responsibilities and stresses, particularly organising other chores like getting the kids safely to school etc.

To ensure that mistakes are minimised, a FPS contemporary strategic direction policy on HR selection and performance could be formulated. The corporatisation ideologies for this policy initiative would be controlled from within the ambit of FPS Corporate Division. In implementing key HRM processes, the FPS would actively encourage dialogue from all stakeholders, setting policy objectives to streamline routine operations. All in all, the transformation and reform processes should be subjected to legal requirements and governance regulations enshrined in the revised Fiji Public Services Act. An independent HR work practices monitoring team coordinated through a tripartite membership framework should be set up with the involvement of the FPSIM, the University of the South Pacific,

UNSW@ADFA, University of Waikato Management School, FPS customers’ representative, a non-government organisation (NGO) affiliate and a private sector representative. This requires evaluating the impact of different interventions at each stage along the results chain.

Sound measurement devices are essential to tracing those aspects of service delivery that are not meeting the expectations of the public. This finding is crucial for the FPS, whereby P a g e | 314

regular customer feedback on the differing aspects of public service can be assessed and evaluated.

This policy directive is crucial for the FPS, to not only select the most qualified employees but also to accurately measure public servants’ performance, and continually review current worker performance benchmarks through the process of job analysis. This policy initiative aligns itself to Baird and Meshoulam (1989)’s finding that even though HR selection is a stand-alone function in HRM, it can’t actually be critically examined in isolation if practitioners need to measure its effectiveness. Hence, without standard benchmarks, the competitive environment available to the private sector will not be available to service-oriented institutions. Without the formulation of the appropriate benchmarks for each job classification, functional classification or respective FPS departments, the performance appraisal methodology employed will be error prone, invalid, unreliable and will measure irrelevant performance indicator attributes. In this context, the FPS may be rewarding an underperformer who is also unproductive in their annual excellence reward programme. All in all and to improve efficiency and effectiveness in the context of changing environment, it is important for the FPS to take note that the nature of work and the jobs are changing rapidly with increasing educational standards (MQR) and increasing stakeholder expectations.

9.6 Limitations

A major limitation for this study has been the current environment of the study which has been very challenging. Fiji has been under a cloud of political and social instability since

1987, and the unstable environment has influenced major stakeholders, particularly within the

FPS. In this regard the management of HR selection has also been under duress. For a P a g e | 315

researcher during this period, the exercise has also been challenging since survey respondents in a politically unstable economy tend to display nervousness when being engaged in the survey.

However, managers of the FPS and employees engaged in the survey have all been very cooperative, even though the researcher perceived instability during the engagement.

One of the major limitations was time management, and this is understandable considering the ‘Fiji time’ perceptions of FPS employees. However, a little patience was the amicable solution.

9.7 Implications for further research

Bell et al. (2004) proposed certain theoretical limitations which would have positive implications for future research; firstly, that future research is needed to better understand exactly how expectations fit into the larger organizational justice framework and the implications of justices expectations for understanding and enhancing applicant perceptions.

They also proposed that further studies need to be done on the role of expectation strength in determining the effects of justice expectations because as the strength of an expectation increased, so did its impact on an individual’s cognition, attitudes and behaviours .

The relevancy of the research by Bell et al. (2004) on a future theoretical framework is substantial for the following reasons: firstly, an applicant’s past experience in a hiring environment may influence his or her current justice perception level. Further, a person’s extant beliefs about an organisation, taking into account his/her personal characteristics and the environmental influence, is a strong determinant of justice perception. Lastly, an applicant’s indirect experience in the same institution, taking into account his/her personal characteristics and the environmental influence – for example, he or she may not be formally P a g e | 316

advised as to the reason why he/she is not selected for the job – has significant implications on him/her

It is also prudent for future researchers to dwell further on the theoretical framework presented in Vosikata (2012), and conceptualise variables that integrate the specific agents and elaborate on their theoretical stance. Rocco and Plakhotnik stated that “a conceptual framework related concepts, empirical research and relevant theories to advance and systematise knowledge about related concepts or issues” (2009: p122) are crucial as basis for a sound theoretical research. They further stated that “such a framework would help researchers define the concept, map the research terrain or conceptual scope, systematise relations among concepts and identify gaps in the literature” (2009: p 126). Since the current study concentrates on the public service, this would be prudent for future

Lastly, a public policy initiative for future governments in Fiji, and the region, is to propel the classification of HRM systems from undeveloped to dominant adhocractic, where attaining international classification is necessary especially to not only upgrade regional classification of HRM competencies, but to also highlight the worth of competencies developed. This would augur well with the policy initiative recommended in this thesis for the establishment of the FPSIM supported by its strategic partners in Fiji and the region.

9.8 Conclusion

In view of the current external environmental factors facing the FPS, the analysis confirmed that it is indeed a critical challenge for HRM doctrines to be efficiently implemented by FPS managers. However, the policy advice of this thesis is that managers of the FPS need not be too extravagant or aggressive in the implementation of the SHRM theories. In this regard, they need to vary personnel practices as a function of organisational characteristics, P a g e | 317

particularly the aggressiveness of the application of personnel practices needs to be aligned with the texture and rigour of the stability of the external environment. Hence, critical HR policy strategic choices for the FPS have to be carefully scrutinised and then applied.

In this aspect of the analysis, it was found that, given the current social and political environment in Fiji, rational decision-making is critical for the implementation of HR selection policy choices that best achieve organisational fit. Further, the social, political and economic environment, particularly the current degree of social capital for the dominant races, should be continually assessed. It also envisaged that there is a positive correlation between political instability and the magnitude of social capital for each dominant ethnic grouping. The degree for the correlation outcome has to dictate the texture of HR selection policies applied. The analysis proposed that as stability settles in, a social capital that is conducive to the management of the FPS will be the antecedent, and the individual social capital for the dominant races will begin to diminish. In this regard, the newly moulded social capital will embrace a multi-ethnic and diverse Fiji and complement fair and unbiased HR selection regulations. A critical outcome which is invaluable to society will be the production of social capital that reinforces economic and political development. In other words, distinctive social capital variables including high school name, political affiliation, religious grouping, ethnicity, township/city/rural community and the provinces respondents belong to would be viewed positively as complementary variables to social capital building. These variables have the potential to positively influence not only who gets selected for the FPS ministries, but also work performance and productivity issues. Within the FPS, managers would then be aggressive in the choice and implementation of HR selection policies as the environmental preconditions would be favourable.

The military regime in Fiji also tried to tackle a large public service operating expenditure by lowering the retirement age to 55 years in 2007. However, they were not P a g e | 318

comprehensive enough to be able to accurately assess the net gains from that exercise because turnover rates were significant, resulting in loss of invaluable working experience and competencies in the FPS. Even though it has been the responsibility of governments to ensure that unemployment figures are managed and contained, the current military regime in

Fiji did not measure real figures to depict loss. However, even since the implementation of this strategy in 2007, the Fiji Government has not been able to deliver a growth rate to match the increasing domestic demands and to formulate a robust civil service with the potential to reinforce excellent public service delivery. Hence, this is a critical challenge for Fiji public service managers.

P a g e | 319

Appendix

Appendix 9.1 Daily Mean Wages of Wage Earners by Industry Group in Fijian Dollars

P a g e | 320

Appendix 9.2 – Key Designations held and influenced by the Fiji Military Forces, 2011

Designation Military Career Civil Servants President Yes Prime Minister Yes PS (PM’s office) Yes PS - PSC Yes, but Military appointed Office of the AG (AG – Minister) No, but Military appointed Deputy Solicitor General ( Ministry of No, but Military appointed Justice) Prisons and Correction Services – Yes Commissioner of Prisons Minister of Finance Yes Ministry of Finance (PS) Yes, but Military appointed Government Printer Yes Ministry of Itaukei Affairs - Minister Yes Ministry of Itaukei Affairs - PS No, but Military appointed Ministry of Defence,National No, but Military appointed Security Immigration, Minister Ministry of Defence,National No, but Military appointed Security Immigration, PS Department of immigration - Director Yes Commissioner of Police Yes Deputy Commissioner of Police No, but Military appointed Ministry of Labour,Industrial No, but Military appointed Relations Employment (Minister) Ministry of Labour,Industrial Yes Relations Employment PS Ministry of Foreign Affairs - No, but Military appointed Minister Ministry of Foreign Affairs - PS Yes, but Military appointed Ambassadors to Foreign Missions Military appointed, a few military personnel Office of the Auditor General No, but Military appointed Supervisor of Election No, but Military appointed Judicial Department – Chief Justice No, but Military appointed Judges and Magistrate No, but Military appointed Office of the Director Public No, but Military appointed Prosecutions - DPP ADPP No, but Military appointed Ministry of Information,National Yes Archives Library Services - Minister Ministry of Information,National No, but Military appointed Archives Library Services - PS Ministry of Strategic Planning -PS Yes, but Military appointed Bureau of Statistics Yes Ministry of Education - Minister No, but Military appointed Ministry of Education - PS No, but Military appointed Ministry of Health - Minister No, but Military appointed Ministry of Health - PS Yes, but Military appointed Ministry of Local Government - Yes Minister Ministry of Local Government -PS Yes, but Military appointed P a g e | 321

Ministry of Women Social Welfare - No, but Military appointed Minister Ministry of Women Social Welfare - Yes, but Military appointed PS Department of Social Welfare - No, but Military appointed Director Department of Youth Sports Yes Ministry of Primary Industries - No, but Military appointed Minister Ministry of Primary Industries - PS Yes Ministry of Fisheries Forestry - Yes Minister Ministry of Fisheries Forestry - PS Yes Ministry of Lands Survey - Minister Yes Ministry of Lands Survey - PS Yes Ministry of Industry Trade - Yes Minister Ministry of Industry Trade - PS Yes Ministry of Transport Works - Yes Minister Ministry of Transport Works - PS Yes Ministry of Provincial Development Yes National Disaster - Minister Ministry of Provincial Development Yes National Disaster - PS Divisional Commissioner – Central Yes Divisional Commissioner – Eastern Yes Divisional Commissioner – Western Yes Divisional Commissioner – North Yes Ministry of Sugar -Minister Yes Ministry of Sugar - PS yes Ministry of Public Yes Enterprise,Tourism,Communication Civil Aviation – Minister Ministry of Public No, but Military appointed Enterprise,Tourism,Communication Civil Aviation – PS (Fiji Civil List, 2011)

P a g e | 322

Appendix 9.3 Fiji Public Service Commission – Annual Confidential Report

STAFF – IN GP CONFIDENCE 36

REPUBLIC OF FIJI STAFF REPORT NOTE: This is a restricted document which will be seen only by a few senior officers. Please state the facts clearly and correctly and do not hesitate to express your views frankly.

REPORT For the period from ______to: ______

SECTION I (to be completed by the officer reported on in his own handwriting).

FULL NAME: In Block Letters. Surname/Family Name First E.D.P. No.

Date of Date of joining Date of Appointment to Confirmed? YES/NO Birth Government Service: Permanent Establishment:

Department: Date of Appointment To Substantive Post: Grade

Substantive Post: Present Incremental Salary: Month:

1. Any periods spent acting during period under review – Record the post held and period, and percentage allowance.

2. State briefly the duties performed during the period under review.

3. Do you consider that you might be more usefully employed in some other part of the Public Service? If so, where (Give reasons).

4. Are you taking any course of instruction in your spare time? YES/NO. If YES give details of the course.

5. Give details of qualifications held and date degrees, diplomas, professional, technical qualifications and service examinations were obtained.

6. Give brief details of any contributions to the literature of your science or profession.

Date:______Signature: ______P a g e | 323

SECTION II REPORT ON QUALITIES AND PERFORMANCE OF DUTIES

Instructions to the Reporting Officer

Listed below are fifteen factors important to performance on the job. For each factor five statements are given. Carefully read each of the statements listed for the first factor, then choose the statement that you believe best describes the officer being assessed. Tick the box to the right of the chosen statement. Repeat this procedure for each of the factors. These reports are not to be shown to the officers on whom they are written but the subject of an adverse report should be conveyed to him/her in writing by the reporting officer or any other officer instructed to do so.

REMARKS 1. QUALITY OF WORK His/her work frequently contains an unacceptable percentage of errors or shows evidence of poor judgment. Doesn’t seem to have enough concern about the quality of his/her work. Quality of work is about the same as that of most employees of this grade and type of work Nearly always turns out a good job. Rarely makes mistakes. Anything he/she does, he/she does almost to perfection

2. QUANTITY Excessively slow; output consistently below standard. Works slowly, rather low production. Completes average amount of work; works at a steady pace. Better than average producer; volume more satisfactory. Works rapidly; consistently completes exceptional amount of work.

3. KNOWLEDGE REQUIRED BY THE JOB Handicapped quite often in his work because of lack of knowledge, understanding or information. Isn’t quite up to par with respect to the knowledge and information required by the job. Has adequate knowledge to do a satisfactory job. Has more job - related information at his/her finger tips than most in this line of work. Complete knowledge of this and related jobs and the inter – relationships of the jobs.

4. INITIATIVE AND RESOURCEFULNESS Definitely not imaginative; waits for others to furnish ideas. Something of a routine worker, who only infrequently contributes something new. Comes up with a new idea now and then. Has more imagination than most in this job classification and type of work; frequently thinks of new ways of doing things.

5. JUDGMENT Very erratic in his/her ability to reach logical conclusions. Finds it quite difficult to analyze a variety of facts to reach sound conclusions. Of the makes poor decisions. Makes an average number of errors or judgment. Can discriminate between relevant and irrelevant details to reach generally sound conclusions. Almost invariably reaches correct conclusions even in the most difficult problems.

7. DEGREE OF SUPERVISION REQUIRED Requires constant supervision or direction. Goes to his/her supervisor or superior for help more often than seems necessary. Acts on his/her own in activities that are usual to the work. Requires less direction or supervision than most employees of this classification. Able to proceed on his/her own with little or no supervision.

8. ABILITY TO COMMUNICATE (ORAL) Vague and ambiguous in expression of ideas. Ability to express himself/herself is somewhat limited. Expresses himself/herself without difficulty. Shows organization and consistency in expression of ideas. Unusual clarity and facility of expression.

P a g e | 324

Appendix 9.4 – Fiji Public Service Commission – Annual Performance Assessment

P a g e | 325

ANNEXUREI PUBLIC llRVICI COMMISSION (111) SKILLS (Relevant to the l>OftJ MO• Sltllb PtewWe-el 1tv the te>o-.. .n.. ~:u-. -r ·:]..L• , [2) -r (1) ~., llt.btr• er Dept. 1Nll] I C•l Communication SRllls -r -, -r -r Knowledoe and Und~tandJng 1------,~---1 I Personal Values ~..-----i-----·------.----- Relatlonshli> S~llls Occupatlonal SRllls Manooement and Leadenhlp SRilk Technkal Ex;perti~ Attocfi the MO~ wnlkaticn PUT a, OVERALL AUEUMENT RATINGS (I) IWP Torgets and Achievements Agreed Ratino

{II) PSC Code of Conduct Aorffd Rating

(Ill) S~llls Aorffd Ratioo

Offictt'• Pln•I Ovtr•ll P1rhrm•nc1 l•tl119 iR Ptrc.tnll ~ To r:altvlal• tlN F1naf Owraff l>Hfolmaoa In P.fC~ f"•J odd tlN J Avr-d RatirPfp f(il (If) & (111}) abo.,. and dN/ttM by lS and multiply by 100. ~ tM o_._ "' th• 8oJI 01 tn. Offbr'I Fltaf 0.--al ~nc• Rating. l 9$•• - Outdonc/lflfl l'Mformonc. [Hr9hly fltt«l fw promotion!-.dwWon of 0>ntroct) ~ llO'• - Good PM'fonttonut {Fltt«l lor promotion'ut«t1/on of «111tra

Signature of Offker: .. ~ .. ·~··· · · ·· ...... ~~...... ~ ...... ~ ...... Date: ..... ~~ ...... PUT C. RECOMMENDATIONS I. COllllINTS AND •ICOlllllNDATIONS OF THI llllllDIATI SUPDVUO•

[/~ drwrtvt/I crf tit• Offlc:.r. a'-! for 1rt~n11nt rscon1,,,_ndot....,4-.)

Signature-Immediate Supervisor Oeiignatlon Date: U. UCOIHllNDAllON OP THI SUP ..VHOIJHUD OP DIVISION

(I) I concur/do not concur with the Immediate Supervisor's onessment and I recommend that the Officer's contract: to be extended for another term of .... --..... m...... yearh]/not to be extended [delete the inapplicable].

ERD 2011 Page2

P a g e | 326

P a g e | 327

ANNEXUIEZ GDIDE TO AllllT llOIEDIA11 SDEIVUOll IN THE FILUMG OF APA

CAUllON TO IMMEDIATE SUPERVISOR This ~ a legal doc1m1ent and requies the OOser.iaOOn of all milor detaib. Any attempt to prt-empt the decisiln of the Permanent Secretmy hough him reporting ~ a breach of the PSC Code of Conduct mid u.iJ reflect the quaiy of the bnmediate 5upeJVisof and can lead to disqilil~ted well before the tine 3 4· ·All worh completed Mlhin aid IC!Tle weU before the tine line " 3- ·All wcrh completed IUittm the line line 2· ·Time line ¬ conmteni>y ochred to . 1- ·SWm ism ore a~O/l lote with very few on tine I TOTAL UTING FOi IWP lClllEVU(ENl NOTE TO UWDIA TE llPEIVUOI tt the agreed Qualfy h ~uolity of cxhievement to the highest ~1ille stondcrf the Of!Ker's rating shall be a S, this figure shall be pbced in the appropriate skit in the Total Ratng for IWP Achievement above and Annual Perform(l)(e Assessment [APA) Section 3 - Assessmenl of the Of!Ker by the lmmecfiate Supervisof(i) - IWP Achievemenl The same procedure shall also be observed fa the other two criteria of Qu

ANNEXURE2

l'>.dom d-- ..,..._ ..-.,-.. ~. -••.ml<>bhty. ,-.,ff!.-_ lnMlwcH-1, ooc...... ,.U,u.tv, ....,,...,, ~. hdl repute en! \lll• ttvough '*"""" .....dl.tbc or ,,.,1orm dutlel ~ e>d ln'OWtlol\I' ""-' nol P«f«m a. on •-It Of•-•-,,,_

If the Oflket lo !lll"tn o S °' h9 rat.19. IN S 1t.ould be tr°"""'fd to the opp1cpnot. 1lot In APA Sedlon 5 · A•-l'M'flt qt tbt O!~ by IM i.n~t~ SupcMlof{il) PSC Code of Cancb.t and oho In the booc In APA Part B 0..-aU Auesvnent Rattnq,(11) PSC Cade ol Cord.id AQrttd Aailn11c

- An ...... i...... t-...tivond ..th ...._,..,,In lit...... o1 -~In,,,...... An ..,,p1ioy- '"'"'CCI: wtlti itan urd ctl'if9"'l•.., th. CM.In• of~ ~tt. publft wrvite ...... ,.,, ...... _ ...... ari"'9 I• !ho ·--of ~nt In""',.. ...,.. - ""'"treat--.-_.,.....---...,. ...- ...... --_, """".... ·-.,, ~of'"""1""cl OM

•--M ..,,.,i..,.. ,.,...... ,,~ w•h .. 1-11>1 - ·-dr.. i.ono •'- by - ., oUlhoolty In !ho ·~· ""-try ....,._ ot .....-- ..... bocl,i c..till_.I... Al>-'*"'- ,,,utll _ Ol>Clt•IWIOU ~llUolay ~ ~ dMK the _._..... 'Will, 01111 .,._.... °' - _...... , of tllo ltalt el o ...... --....,.... An-pio.-. ....,,. -..andtcloo .-.. '"""'"'""°ld,oryoonlbc:lol-•I ~er.,.._, ln-,...,ltt..,...... ,_inthe publk •-n- A".....,1_,....,, .,,.~_,..,..,,.,.,.,.blna"-- T,.--.., A - mwt noc. n t"9 _,,,. ol orn - with-•-., tlM P4A>tc l«Vke...... ,_or mw-1"'9 W«motlon In - to• f-1l0f ...... ,,,.,i..dt0t lo modo ""~-...... 1wpti.l1 An...,,,._ - - MO... lmpt_...,. of olftdutl& - - croothorltil In orcillr to - obenoftt Of """--+.t lhe ...... _ °' ......

P a g e | 329

P a g e | 330

P a g e | 331

P a g e | 332

ANNRXURl2

S-M~ Moro... 11a!I..,....,.. _ ..._...... obl4yto..,.,..d..., tholeb. --Strategtc \/hlon S-the ~ ~ pirtlw9. I I I ...... I ...... Computor lltwa

Upon o thorough WldentaidlnQ of the Rotlncr lndlce. ..uler the vorlou• •ub-lwodlngi.. ~ Immediate ~ wlll tt- ~ to rote the Offbr lmp>O(• 19* di• 01- I> !If.,.,!( ond wld ....ct IO be ~d to •ult th

Thl1 -•that tt the Obow J.t cloo• ~ lo.,tur• 10m1t-' U"' ,,...... ,,... '~ '°"''''od toi the Po'4..,,. lmmodlal• ~"-""°' lho.rlJ U1.,t'loo• I lho "''"Inv """1\ in th• bi~ proWlod undor oacl• 11A> - heodinQ .,,,..,...,!• Su.,.,_. lhould al..uoyl b.. In mind that the and the Coucm!1Mf1t rwlle. on thetn to be honeft so that rner11orlou• Offkerl and 1tnder - puformlng -. cire ldtntlf\cd and the cle>fr.,fng to be rttorntd ond U.. 'hconvt" - Ot'I' tn al producttu. OfflUI* and 1tte «~t ol ()Ja Serw. Reform<.

'

ANNEXURE2

The<• or• - 171 molo< 1ub - heading• In Ihm above •-"' and the\I ore Con'""''*"tlewl Sltllh. llg relevant ilols bv the lmrnedlata Supervjscx_

Whftl the Im~ Supeorwar hos flnoH~d the ADtlng lrdce. for each lllb - ~ itUI• - IOO'lt e - Tht Otflc• ho\ lllo ol lh. 1fifl41t •t.1111- 7S~ J - The Officer i-haN or ,,_.. of thr rei..-.t ilolll> - 500. 2 The Officer lxn a quortff or more of tJw r~ sllill> - 25'1. I • The Olfker l1llt i.,, lt.. a quotlw of the ~i..-.t iltllh le ..~.taaOHket ' Ut • Ofllttt'> 1'CJme ls AIPtKi Conimunkotlon ~ I<~ and Undeotandlng P9f\onol ~ :} T..... llvt.r•• an A.ipio'• --rat.np• lo< oaoh ...... _ ... th.,, tbould bo -...... , u.. ao•- ratlno> ...,...,., -.cmd RelotlOnlhipo Sllill• 1 Occ.upotlonol Slollk .a -tnd>o""""""'"'l>o~•lttAPA~•C110 Slolh ~ment and leodmfllp Sllflh 2 T.clvric;cd E>-1>• s TOTAL 2• To find~'> Slollk Aatlno l wll be 24 • lS (all 7 ~ haue o mmdmum tear• of S eadJ i-a 35 II the mo~l mum totaQ • !> frotlnq t«nl for the wnot. S1>llt A11e>Jmen.t) The<"°'• AWlo'• tcete uAll be 24 7 • JA:!. The ;l.43 d>ould be tr

If ..u c et••""•" 1" r.tlnw with.. 1ultert0nde OHi.er, die c••• ••uhl lte Nfen.-1 .. peur lmm.-ild• Supervber •• cenclll•te

P a g e | 333

Appendix 9.5

HR Selection: Appointment and Promotion - The Fiji Public Service - Process & Merit

Criteria

5.1 Advertising of Vacancies

All vacancies have to be advertised in the Public Service Official Circular and the local print media. When a substantive vacancy arises, it is to be submitted to the Commission on plain paper together with a completed and signed “Staff Requisition Form” (GP Form 140), which shall contain the following information:

1. Ministry heading

2. Designation

3. Functions & principal accountabilities of the post

4. Minimum qualifications requirement

5. Salary range.

6. Any special requirement

Deadline for Application

Duration of 21 days from the date of publication is the standard time frame. A further period of 5 working days may be allowed for applications sent through the po

Receipt of Applications

All application forms must indicate the date it was received, preferably an official stamp to be used for this purpose. A vacancy folder to be opened, in which all applications to be kept for each vacancy. Such vacancy folder should contain the following information:

1. staff requisition

2. vacancy notice

3. functions and principal accountabilities

4. application forms

5. letters of acknowledgement P a g e | 334

Acknowledgement of Application

After applications are closed, acknowledgment letters are to be prepared and issued to all applicants concerned. Alternatively, these could be acknowledged as and when received and a list of all applicants received together with a bio-data of the candidates. A copy of the acknowledgement letters and bio-data is placed in the vacancy folder.

P a g e | 335

5.2 Preparation of Bio-Data

The preparation of bio-data is an important and the core component of post processing. It consolidates all the relevant background information on the applicants for vacancies from both within the public service and from outside. The information include:

1. Current position

2. Date of Birth/Age

3. EDP/FNPF number

4. Ministry/Department/Company

5. Qualifications

6. Work Experience

7. Annual Confidential Report Grading/References & Testimonials

The time frame for data preparation and analysis will depend largely on the availability of information sources (Application forms, Personnel Files, Civil List, MIS, Government Training Centre and Ministries/ Departments) for verification of authenticity. The data should be correct and portrays the full background information of the applicants.

a Scheduling and Short Listing

Applications are sorted, separating serving from non-serving applicants and then classified into two main categories: qualified and unqualified.

b Notification of Interview

A short-list of qualified applicants is prepared according to the criteria set out in Part 2, Regulation 5(3) of the Public Service (General) Regulations, 1999. The short-listed candidates are notified and invited/called for an interview, specifying the vacancy concerned and date, time and venue of such interview.

c Interview Panel

An interview panel should comprise not less than three members who should hold positions above that of the grade of the vacancy in question. Preferably ethnicity and gender should be considered when appointing a panel.

d Interview P a g e | 336

Part 2, Section 4 of the Public Service Act, 1999 requires, consistent with Section 140 of the Constitution that all appointments or promotions must be made on the basis of merit after an open, competitive selection process. This is further amplified under Regulation 5 of the Public Service (General) Regulations 1999. An interview is conducted to determine and ascertain the level of knowledge and understanding of the job/post; it also enables a face to face assessment of the personal attributes and presentational ability of the applicants.

(www.psc.gov.fj)

P a g e | 337

5.3 Staff Board Submission a. Preparation of Commission Paper

When preparing a Commission Paper, the assessment should be based on the relative suitability and capacity of the applicants for the post and taking account of the following work related qualities:

1. skills and abilities;

2. qualifications, training and competencies;

3. standard of work performance (ACR and other assessment);

4. capacity to perform at the level required;

5. demonstrated potential for further development;

6. ability to contribute to team performance;

7. any other relevant matters

Where the merits of the best two remaining applicants are equal in all respects, the following matters are to be considered to determine the most suitable candidate for the post:

1. length of service

2. gender & ethnicity

The submission of the Ministry to the Public Service Commission should be properly documented and accompanied with relevant documents that will substantiate the recommendation, this should include the following:

1. submission proper

2. ACR for all serving applicants and any other assessment, the latter especially for applicants from outside the Service;

3. Interview assessments and results;

4. Biodata (CV) of all the applicants; and

5. Ethnic and gender representation within the Ministry/Department.

6. (www.psc.gov.fj) P a g e | 338

5.4 Procedures of Recruitment of Expatriates

Recruitment of Expatriates is made only in exceptional circumstances where local expertise is not available to fill the post. The recruitment process is expensive and also time consuming. As required under Section 141 of the Constitution, in the case of the appointment of a person who is not a citizen of Fiji, the concurrence of the Prime Minister must be sought and obtained. To assist Ministries/Departments ease the process in the recruitment of expatriates, the procedure set out below is to be followed: -

a Advertisement

In accordance with Regulation 5(2)(a) of the Public Service (General) Regulations 1999, an appointment or promotion may only be made if the vacancy was notified (advertised) in a Public Service Official Circular within the last year as open to any citizen of the state

If at the closing of the advertisement period, no local application is received or local applicants are found to be unsuitable to fill the post, then the post may be advertised abroad.

b. Advertisement Abroad

Permission and approval to advertise the post abroad must be sought from the PSC. It is advisable that the Ministry of Foreign Affairs is also informed as prospective candidates may seek assistance /further information from Fiji’s diplomatic Missions abroad. It is a normal procedure to obtain at-least three (3) quotations from the potential advertisers (Newspaper Publishers) abroad. Assistance may be sought from the foreign diplomatic missions in Fiji as to which Publishers to contact, including addresses, fax numbers, etc.

c. Costs

The cost of the advertisement is to be met from within the Ministry/Department budget/funds.

d Recruitment Process

The normal recruitment process is to follow. Assistance from Fiji’s diplomatic Missions abroad may be sought for the purpose of obtaining additional information on the applicants or to interview them so as to minimise cost.

P a g e | 339

e Police and Medical Clearance

The candidate identified to fill the post should also be requested to arrange (at their expense) his/her Police and Medical reports which are to be sent to the Ministry.

f Commission Submission

The Ministry/Department should forward the following documents to the Commission for its consideration and decision: -

1. Commission Submission;

2. Staff Board Paper with Chief Executive Officer’s endorsement;

3. Vacancy folder;

4. Short listed applicants CV, certified true copies of academic transcripts, references, testimonials;

5. police and medical report;

g Prime Minister’s Concurrence

Once the Commission has approved the recruitment of the expatriate, the Commission Paper with a covering minute will be forwarded by the Commission to the Prime Minister seeking his concurrence.

h Contract Documents

Once the Prime Minister has concurred with the recruitment of the expatriate, the Commission will prepare the contract documents together with an offer of appointment which will be sent to the respective Ministry for forwarding to the appointee.

i Acceptance of Offer

If the appointee accepts the offer to take up the position on the terms and conditions specified, he/she should sign the Contract documents (and to be witnessed) and return these to the respective Ministry.

j Arrangement of Passage

The Ministry, alternately the officer may arrange for his/her passage and transfer of personal effects to Fiji. Three competitive quotes must be obtained and forwarded to the PSC for approval. The PSC will forward the quotes to the Ministry of Finance and National Planning P a g e | 340

for issue of a Passage Order. Alternatively the officer may pay for his/her own passage and on assumption of duty will submit his/her claim for reimbursement.

k Accommodation

The Ministry should in advance liaise with the PSC and PWD for suitable Government Quarters to be ready for appointee’s accommodation.

l Immigration Procedures

Normally the Immigration Department grants Visitors Visa to all expatriates on first arrival to Fiji. The Department is required to take the appointee’s Passport and a copy of the Contract document (only the first and last pages) to the Immigration Department for endorsement and grant of Work Permit.

(www.psc.gov.fj)

5.5 Procedures for Appointment Temporary Officers and Temporary Relieving Officers a. Temporary Officers/Employers/Wage Earners

Recruitment of temporary officers must be in accordance with the requirements of section 140 of the Constitution section 4 subsections 4(2) and 4(3) of the Public Service Act 1999 and Regulations 5 and 8 of the Public Service (General) Regulations 1999. A temporary appointment is usually made when post becomes vacant, either temporarily or substantively and recruitment is likely to take longer than usual due mainly to scarce skills and when the appointment of an incumbent is critical due to the essential nature of the job. The terms and conditions governing temporary appointees are the same as those for permanent officers/employees/wage earners.

(www.psc.gov.fj)

5.6 The procedure to follow when making a temporary appointment is as explained below: a. Temporary Appointment

There are two options (l) To either select prospective and suitably qualified and experienced candidates either from the pool of applications maintained by the Ministry which are received from time to time OR (2) To advertise the post in the Public Service Official Circular (PSOC) P a g e | 341

and the local media inviting applications from suitably qualified and experienced persons, (section 5.1 above is followed, except the deadline for applications at (a) could be shortened from the usual 21 day, say 14 or 10 days.

Option (i):

Once the selection of prospective and suitably qualified and experienced candidates from the Ministry’s pool of applications id done a short-listing of the applications/applicants is undertaken to determine the most suitably qualified and experienced applicants/candidates. After this determination is done and completed the applicants/candidates are invited to an interview. The procedure/process at section 5.2 above is followed to select the most meritorious candidate for temporary appointment.

Option (2):

Once the procedure/process at section 5.1 above is completed up to (c – Acknowledgement of Applications) a short-listing of the applications/applicants is undertaken to determine the most suitably qualified and experienced applicants. The short-listed applications are then invited for an interview following the procedure/process at sections 5.2 and 5.3 above.

Before a temporary appointment is made, the necessary medical and police clearance is obtained.

Temporary Relieving Appointment

A Temporary Relieving appointment is usually necessitated by the absence of short duration arising out of a post usually at the base grade, becoming either temporarily vacant due to the substantive holder proceeding on annual or sick leave, and also when a post becomes substantively vacant when the incumbent leaves office, and in both cases an immediate relief is warranted and justified for uninterrupted delivery of service, and an acting appointment cannot be made because of the grade of the post which does not make it eligible for such appointment. c. The procedure to follow in making a temporary relieving appointment is explained below:

There are two options (l) To either select prospective and suitably qualified and experienced candidates either from the pool of applications maintained by the Ministry which are received from time to time OR (2) To advertise the post in the Public Service Official Circular (PSOC) and the local media inviting applications from suitably qualified and experienced persons, P a g e | 342

(section 5.1 above is followed, except the deadline for applications at (a) could be shortened from the usual 21 day, say 14 or 10 days.

(ii) Option (i): Once the selection of prospective and suitably qualified and experienced candidates from the Ministry’s pool of applications id done a short-listing of the applications/applicants is undertaken to determine the most suitably qualified and experienced applicants/candidates. After this determination is done and completed the applicants/candidates are invited to an interview. The procedure/process at section 5.2 above is followed to select the most meritorious candidate for temporary appointment.

(iii) Option (2): Once the procedure/process at section 5.1 above is completed up to (c – Acknowledgement of Applications) a short-listing of the applications/applicants is undertaken to determine the most suitably qualified and experienced applicants. The short- listed applications are then invited for an interview following the procedure/process at sections 5.2 and 5.3 above.

(iv) Before a temporary appointment is made, the necessary medical and police clearance is obtained.

(www.psc.gov.fj)

P a g e | 343

Appendix 9.6 Pilot Study – Fiji Public Service Survey

Introduction

The pilot study was deliberately prepared for this chapter in view of Zhang and Shaw

(2012) finding that researchers need to enhance perceptions of their readers in view of credibility in the methods they employ for their research. In that regards, the writer of this thesis has chosen to address credibility effects in the research instruments, particularly why did the Fiji Government was chosen, and particularly the Fiji Public Service Commission.

Further, the analysis of the same signifies reliability and validity ratings for the research instruments as evident. In their study, Maiyaki & Mokhtar (2011) explored a small sample data on the determinants of customer behavioural responses in the Nigerian retail banking.

Hence, the context of the survey guided them to examine instrument validity, reliability and subsequently the data normality. Results show that the instrument is reliable and the data for preliminary study exhibit reasonable normality. Their research explored and validated the instrument of the various antecedents of consumer behavioural responses most of which were hitherto ignored. Their findings have been instrumental in preparing the blue prints for the

FPS pilot study.

Further, Gay, Mills and Airasian (2006) articulated that a pilot test is considered to be like a “dress rehearsal” in which a small scale trial of the study is conducted prior to the full- scale study. Maiyaki & Mokhtar (2011) complimented Gay, Mills and Airasian (2006)’s finding and asserted that their study which is a pilot test was carried out in order to achieve some objectives. Firstly, it was done to test the validity and reliability of the instrument of the study and secondly, it aimed at obtaining an insight into the real conditions of the actual study. Thus, this would enable the researcher to anticipate and adjust to potential problems during the full-scale research. Among the major concern of pilot test is the instrument validity and reliability. P a g e | 344

The objectives of the pilot study for the Fiji public service survey was derived from

Maiyaki & Mokhtar (2011) model, where a pilot study was conducted on a Fiji public service department; Department of Immigration. Acknowledging Maiyaki & Mokhtar (2011) model the pilot study for the Fiji public service survey was also conducted to test the validity and reliability of the instrument of the major survey, which is mostly deductive where the model is dominant (please refer to appendix C), and the theoretical framework for this study has been designed as an enquiry into the social issues distorting the hiring processes in Fiji’s

Public Service. Subsequently, the research instrument used for gathering the raw data is survey where structured questionnaires will be used for the case study analysis and also for the generalization to the larger population of the Fijian Public Service. The survey instruments (questionnaires) have been designed to cover all key variables for examination considering the anticipated arrangement and organization of data.

Further, the Fiji Pilot study was also aimed at obtaining an insight into the real conditions of the actual study, given the current era of political instability in the Fiji Islands.

Fiji’s current Prime Minister and Commander of the Fiji military; Commodore Frank

Bainimarama overthrew a democratically elected government in Fiji accusing them of being inefficient, ineffective, unproductive and racist, promising stakeholders that he and his regime will clean up the “faults” in the government systems.

Hence, one of the prime objectives of the pilot study on the Fiji Public Service is to assess the initial reactions of the public servants and ascertain what elements of the research instruments are mostly provocative and sensitive given the current instability of the environment that the state institutions are operating under. In this regard, this would enable the researcher to anticipate and adjust to potential problems during the full-scale research and ascertain the validity of the survey instruments by modifying the current instruments so as to P a g e | 345

ensure that the instrument is measuring what it is supposed to measure and not something else (Maiyaki & Mokhtar (2011).

Validity of Research Instruments

In his study, Smith (1991) ascertained that validity is defined as the degree to which the researcher has measured what he has set out to measure. Kerlinger (1973) further explored the finding in propping the question: Are we measuring what we think we are measuring? Gatewood et al. (2011) also highlighted that when we are concerned with the accuracy of judgements or inferences made from scores on HR selection measures, such as predictors, we are interested in the validity, and in this sense validity refers to the degree to which available evidence supports inferences made from scores on selection procedures.

They went on further to say that from a HR Selection procedure perspective, HR selection researchers want to evaluate their inferences, that is, they want to know how accurate are the predictions they have made, based on data collected from a selection procedure such as employment interview (Litwin, 1995). In view of the Fiji case study design framework, where the mixed method approach composes of a structured survey and archival and publicly available documents, it is probable that the validity of such instruments may not be measured.

However, the instruments need to be subjected to reliability measures to ensure that the instruments are free from error to yield a consistent outcome.

Reliability of Research Instruments

Zikmund et al. (2003) found that reliability is the degree to which measures are free from error and therefore yield consistent results. They found that imperfections in the measuring process affect the assignment of scores or numbers in different ways each time a measure is taken and these would contribute to low reliability scores. Sekaran and Boulgie

(2010) compounded Zikmund et al. (2003)’s study by highlighting that reliability of a P a g e | 346

measure indicates the extent to which an instrument is error free and thus, consistent and stable across time and also across the various items in the scale. Hence, to comprehensively subject the research instruments as against reliability measures, alternate-form reliability and internal consistency reliability tests were activated in validating the research instruments.

Alternate-form reliability

Litwin (1995) ascertained that alternate-form reliability provides one way to escape the problem of the practice effect, and it involves using differently worded items to measure the same attribute. Specifically, he advised changing the actual wording of the items themselves. On this note, when the Pilot Study on the Fiji Public Service institution was conducted, and when respondents were asked of their political preferences, using the undermentioned question, almost 95% of respondents did not fill in the space provided.

Item 1

Please write your preference.

What political party are you affiliated with? (Please write)

The 95% non-response rate to the above question gathered from the Fiji Public Service institution under survey hinted that this element of the research instrument is provocative and sensitive given the current political climate in Fiji. Hence, to reinforce and strengthen the reliability of the instruments of research for the major survey in Fiji’ Public Service, an alternate-form reliability approach was adopted whereby the question in item 1 above was changed, and this is specified in item 2.

P a g e | 347

Item 2

Please write your preference.

What political party did you vote for in the last election? (Please write)

An additional question was included in the questionnaires as per the advice of my supervisors. The question is: Total years of Schooling (Primary + Secondary + Tertiary).

The question was included to enable the research to fully grasp the depth of the competency bank within the selected Fiji Public Services ministries. Further, the competency banks may enable the HR division to create skill matrices for each job or job family, and subsequently profiling the competency base for each worker to the related tasks and roles of the designation (Dessler, 2012). Additionally, the impact of this question is vital since it may enable those Fiji Public Service institutions survey or any organization for that matter to formulate a management personnel replacement chart (please see figure 4.1), and the chart specifically outlines current worker performance ratings. In particular, the charts show the present performance and likelihood for promotion for each current employee. The readings as evident and exemplified in figure 4.1, will also determine HR selection of those from within the organisation for promotion. Lastly, the model will also help HR personnel to determine whether, there is a need to select from the external environment, because the profiles that the designation needs is not available from within.

P a g e | 348

Figure 4.1

Management Personnel Replacement Chart

Dessler (2010)

Internal Consistency Reliability of Research Instruments

Litwin (1995) highlighted that Internal Consistency Reliability (IRS) is another commonly used psychometric measure in assessing survey instruments and scales, which is applied to groups of items that are thought to measure different aspects of the same concept.

It is an indicator of how well the different items measure the same issue and Litwin (1995) affirmed that this reliability measure is important because a group of items that purports to P a g e | 349

measure one variable should indeed be clearly focussed on that variable. Conventional theories and particularly Litwin (1995) affirmed that internal consistency reliability is measured by calculating a statistic known as Cronbach’s coefficient alpha, named after the

20th century psychometrician who first reported in 1951. He further stated that the coefficient alpha measures internal consistency reliability among a group of items combined to form a single scale, hence a statistic that reflects the homogeneity of a scale.

The pilot study on the Fiji Department of Immigration emulated a pilot test of an ongoing project where a few samples of Fiji public servants were randomly selected. This is in line with the recommendation by Malhotra (2008) and (Maiyaki & Mokhtar (2011) that the sample size for pre-test is normally small, ranging from 15-30 respondents but it be increased substantially if the test involves several stages. Hence, a total of 50 copies of questionnaires were personally distributed and 26 were returned out of which 2 were not properly completed and hence, not considered for analysis. In that regards, a response rate of fifty two percent

(52%), this is highly commendable.

Calculating Internal Consistency Reliability of the Pilot Study

In their study, Sekaran and Bougie (2010) affirmed that the most popular test of inter- item consistency reliability is Cronbach's alpha coefficient; hence this test was employed in the pilot study to measure internal consistency of the instruments employed. Cronbach's alpha is a measure of internal consistency, that is, how closely related a set of items are as a group. A "high" value of alpha is often used (along with substantive arguments and possibly other statistical measures) as evidence that the items measure an underlying (or latent) construct. However, a high alpha does not imply that the measure is uni-dimensional. If, in addition to measuring internal consistency, you wish to provide evidence that the scale in question is unidimensional, additional analyses can be performed. Exploratory factor analysis is one method of checking dimensionality. Technically speaking, Cronbach's alpha is not a P a g e | 350

statistical test - it is a coefficient of reliability (or consistency). In addition, Sekaran et al.

(2010) found that reliability of a measure indicates the extent to which an instrument is error free and thus, consistent and stable across time and also across the various items in the scale.

The research explored and validated the instrument of the various antecedents of selection distortions, including networking ties, and analysis showed that the instrument is reliable and the data for preliminary study exhibit reasonable normality.

In calculating the cronbach's alpha measure, the completed questionnaires were analysed using SPSS version 20 for windows. Before the distribution of the questionnaires, the draft was initially submitted to experts and also to the typical respondent for face and content validity. The whole process was completed within the period of four weeks in the months May/June 2010.

P a g e | 351

Table 5.2

Reliability Analysis for Pilot Study - Fiji Public Service Survey

Reliability Statistics

Questions Cronbach's Alpha Cronbach's Alpha N of Items Based on Standardized Items

11 0.811 0.794 10

12 0.884 0.901 10

13 0.885 0.885 10

14 0.762 0.752 10

15 0.513 0.545 10

16 0.659 0.715 10

0.905 0.906 10 17 0.812 0.813 10 18 19 0.786 0.796 10

20 0.692 0.724 10

Mean 0.771 0.783

Source: Fiji Public Service survey data – Pilot Study - Vosikata, S (2011)

Cronbach's alpha is a measure of internal consistency, that is, how closely related a set of items are as a group. As depicted in table 7.2, Cronbach's alpha value for questions 11,

12, 13 and 14 were relatively high. However, for question 15, the value was 0.513, a low and insignificant score. As depicted by Zikmund et al. (2003), the low reliability score was due to certain imperfections in the measuring instruments of question 15, and this was rectified prior P a g e | 352

to the major research in the Fiji Public Service. The Cronbach's alpha value for questions 16,

17, 18 and 19 were again relatively high with an average of 0.771.

A "high" value of alpha is often used (along with substantive arguments and possibly other statistical measures) as evidence that the items measure an underlying (or latent) construct. However, a high alpha does not imply that the measure is uni-dimensional. If, in addition to measuring internal consistency, you wish to provide evidence that the scale in question is uni-dimensional, additional analyses can be performed. Exploratory factor analysis is one method of checking dimensionality. Technically speaking, Cronbach's alpha is not a statistical test - it is a coefficient of reliability (or consistency). In their study, they affirmed that the most popular test of inter-item consistency reliability is Cronbach's alpha coefficient; hence this test was employed in the pilot study of the Fiji Public Service survey to measure internal consistency of the instruments prior to the major survey. The data was analysed using SPSS version 20 for windows and the results out outlined as per figure 1.1.

Conclusion

In view of the political environment in Fiji, it has been anticipated that most public servants would be reluctant to open up and be flexible in answering the research question regarding political affiliation, hence the need for an alternate-form reliability mechanism. In other words, this would provide an avenue to escape the problem of the practice effect using differently worded items to measure the same attribute. The outcome of the pilot study accurately addressed this problem and wordings were changed to address the political scenario of Fiji and ensure that FPS workers were flexible to answer questions regarding political affiliations of respondents. It was also found that FPS workers were at ease in regards to questions on religious and ethnicity issues.

Further, for the ten questions prepared for the questionnaires, results for the internal consistency reliability test found that almost all questions met the credible reliability readings P a g e | 353

of 70% level. However, only one question recorded a 51% cronbach alpha reading which was rectified prior to the major research in the FPS.

However, before the distribution of the questionnaires, the draft was initially submitted to my supervisors for their comments and also to present evidence that the typical respondent for face and content validity was met. In the process, the pilot study explored and subjected the research instruments to a typical FPS environment. The analysis on reliability has showed that the instrument is reliable and the data for preliminary study exhibit reasonable normality.

P a g e | 354

Appendix 9.7 The First Public Service in Fiji – Civil List 1 - 1874

1. Cyril H.H. Irvine (Clerk of the Executive Council and Despatch Clerk in the Colonial Secretary’s Office) 2. 3. Job enrichment (skill flexibility, job variety, responsibility) 4. Self-managed or autonomous work groups 5. Employee participation in decision making 6. Systems to encourage feedback from employees

Judicial Department 1. Joseph Hector Garrick, Esq., Barrister at Law (European Judge of the Central Court of Fiji and Chief Police Magistrate of the Colony) 2. Ratu Marika (Native Judge of the Central Court of Fiji) 3. Sydney Chas. Burt, ESQ., (Crown Prosecutor) 4. Mr. William Isaac Thomas (Registrar of the Central Court of Fiji and Sheriff of the Colony). 5. Mr. John Harman (Interpreter and Clerk to the European Judge and Chief Police Magistrate) 6. Mr John St. Julian (Clerk of Petty Sessions for the Police Court, Levuka).

Colonial Secretary’s Office

1. Mr. John Langford (Clerk) 2. Mr. David John Chisholm (Clerk) 3. Doctor John Cruickshank (Medical Officer to the Government) 4. Mr. John David William Vaughan (Housekeeper and Overseer of Stores and Works).

Postal Department 1. Captain Frederick Platt Blackmore (Chief Post Master of the Colony).

Police Department

1. Mr. James Doxey Senior (Sub Inspector of Police) 2. Mr William W Thomas (Clerk).

Gaols

1. Mr. John Cox (Gaoler, Levuka) 2. Mr. John Williams (Warden of the Gaol, Levuka) 3. Mr. James Molloy (Gaoler and Sub Inspector of Police, Lovoni).

Treasury department

P a g e | 355

1. Mr. Charles E Gundry (Accountant of the Colony)

Customs Department

1. Mr. David Henry Smart (Collector of Customs) 2. Mr. Nicholas Samuel Hedstrom (Harbour Master) 3. Mr. Thomas Thomas (Clerk and Bond Keeper) 4. Mr. Sydney Wilson (Boarding Officer) 5. Mr. William Good (Boarding Officer) 6. Mr. D. J. Christy (Light Keeper)

Department of Native Affairs

17. Henry Charles Thurston, ESQ; (Stipendiary Magistrate for the South Coast of Viti Levu and Kadavu); 18. Alexander Eastgate, ESQ; (Stipendiary Magistrate for the North Coast of Viti Levu and Yasawa) 19. John Packer Gore Martin, ESQ; (Stipendiary Magistrate for Vanua Levu and Taveuni) 20. Horace Caton Emberson, ESQ; (Stipendiary Magistrate for the Windward Islands, including Moala, Totoya, and Matuku, and to be Deputy Collector of Customs at Lomaloma 21. Maafu (Roko Lau) 22. Tui Cakau (Roko Cakaudrove) 23. Ratu Epeli (Roko Tailevu and Naitasiri) 24. Ratu Savenaca (Roko Central Islands or Lomaiviti) 25. Tui Macuata (Roko Macuata) 26. Tui Bua (Roko Bua) 27. Tui Dreketi (Roko Rewa) 28. Cagi Levu (Roko Kadavu) 29. Ratu Kini (Roko Nadroga) 30. Ratu Isikeli (Roko Ra) 31. Roko Tui Bau (Roko Ba and Yasawa) 32. Matanitobua (Roko Namosi

Immigration Agent and Protector of Imported Labourers

1. Mr. Henry Bentley Clerk to the Secretary for Native Affairs 1. Mr. Robert Robertson

(Fiji Government Gazette No. 5, 1874; Vosikata, 2013).

P a g e | 356

Reference List 1. Acquaah, M., & Eshun, E. (2010). A longitudinal analysis of the moderated effects of networking relationships on organizational performance in a sub-Saharan African economy. Human Relations, 63(5), 667-700.

2. Acquaye, E., 1984, Principles and Issues. Land Tenure and Rural Productivity in the Pacific Islands. FAO, Rome; Institute of Pacific Studies, University of the South Pacific, Suva; and South Pacific Regional Environment Programme, Noumea.

3. Adam, F., Roncevic, B., (2003), Social Capital: Recent Debates and Research Trends, Social Science Information, 42 (155 – 183).

4. Adams, J. S., (1963), Toward an understanding of inequity, Journal of Abnormal and Social Psychology, 67 (422 – 436).

5. Adams, J. S., (1965), Inequity in social exchange, In L. Berkowtz (Ed.), psychology, 2 (267 – 299. 6. Adler, P., Kwon, S., (1999), Social Capital: The Good, the Bad and the Ugly. University of Southern California, LA.

7. Agneessens, F., & Wittek, R. (2008). Social capital and employee well-being: Disentangling intrapersonal and interpersonal

8. Aguinis, H., & Smith, M. A. (2007). Understanding the impact of test validity and bias on selection errors and adverse impact in human resources selection. Personnel Psychology, 60(1), 165-199.

9. Albright, C.S., Winston, W, Zappe, C.J., 2004, Data Analysis for Managers, Thomson Learning, Belmont.

10. Anderson, C.A. (1983). Imagination and expectation: The effect of imagining behavioural scripts on personal intentions. Journal of Personality and Social Psychology, 45, 293–305.

11. Andrews, R., & Brewer, G. A. (2013). Social capital, management capacity and public service performance evidence from the US states, Public Management Review, 15(1), 19-42.

12. Anbari, F., T., (2002), Quantitative methods for project management (2nd ed.), New York: Injustice Institute for Learning.

13. Antoci, A., Sacco, P., L., Vanin, P., (2007), Social Capital accumulation and the evolution of social participation, The Journal of Socio – Economics, 36 ( 2007), 128 – 143.

14. Appana, S. (1998). New Public Management and Public Enterprise Restructuring in Fiji, Fiji Institute of Applied Studies (1) 1. P a g e | 357

15. Argyle, Michael; Adrian Furnham & Jean Ann Graham (1981), Social Situations. Cambridge: Cambridge University Press.

16. Armstrong, J. S., (1967), Derivation of Theory y Means of Factor Analysis or Tom Swift and His Electric Factor Analysis Machine, The American Statistician, 21 (5), 17 – 21.

17. Arthur, J. B., (1994), Effects of human resource systems on manufacturing performance and turnover, Academy of Management Journal, 37 (3) 670 – 686.

18. Bacharach, S. B. 1989, Organizational theories: Some criteria for evaluation. Academy of Management Review, 14: 496 –515.

19. Bachrach, D., Bendoly, E., & Podsakoff, P. M. (2001). Attributions of the “causes” of group performance as an alternative explanation of the relationship between organizational citizenship behavior and organizational performance. Journal of Applied Psychology, 86(6), 1285-1293.

20. Bae, J., Rowley, C., (20100), Changes and continuities in South Korean HRM, Asia Pacific Business Review, 9 (4), 76 – 105.

21. Bair, L., Meshoulam, I., (1988), Managing two fits of strategic human resource management, Academy of management, 13 (1) 116 – 128.

22. Baldacchino, G., (1999), Small Business in small island nations: A case study from Fiji, Journal of Small Business Management 37 (4), 80-84.

23. Baleiwaqa, T., (2005), What went wrong?, A historical overview of Fijian modern political development: 1960 – 2000, Department of History, Davuilevu Theological College, Fiji.

24. Barber, B., Odean, T., (1998), Boys will be boys: Gender over confidence and common stock investment, University of California, CA,.

25. Barnett, M. L., & King, A. A., (2008), Good Fences Make Good Neighbours: A Longitudinal Analysis Of An Industry Self-Regulatory Institution, Academy of Management Journal, 51 (6) 1150 – 1170.

26. Barney, J. B., Wright, P. M., (1998), Human Resources Management, 37 (1 ) 31 – 46.

27. Barney, J., (1991), Firm resources and sustained competitive advantage, Journal of Management, 17 (1) 99 – 120.

28. Barrett, M. A., & Hort, L. K. (2000). The use of tacit knowledge in selection decisions in universities. Athens, Greece: Decision Sciences Institute.

29. Bauer, T. N., Maertz, C., Dolen, M., & Campion, M. A. (1998). A longitudinal assessment of applicant reactions to an employment test, Journal of Applied Psychology, 83, 892-903. P a g e | 358

30. Bauer, T. N., Truxillo, D. M., Sanchez, R. J., Craig, J. M., Ferrara, P., & Campion, M. A. (2001). Applicant reactions to selections: Development of the selection procedural justice scale (SPJS). Personnel Psychology, 54, 387-419.

31. Beatty, R. W., Ewing, J. R., & Tharp, C. G. (2003). HR’s role in corporate governance: Present and prospective. Human Resource Management, 42(3), 257-269.

32. Becker, B. E., Huselid, M. A., (1998), High performance work systems and firm performance: a synthesis of research and managerial performance, Personnel and Human Resources Management, 16 (1), 53 – 101.

33. Bell, B. S., Ryan, A. M., & Wiechmann, D. (2004). Justice expectations and applicant perceptions. International Journal of Selection and Assessment, 12, 24-38.

34. Bell, B. S., Ryan, A. M., & Wiechmann, D. (2006). Consequences of justice expectations in a selection system. Journal of Applied Psychology, 91(2), 455-466.

35. Benbasat, I., Goldstein, D. K., Mead, M., (1987), The Case Research Strategy in Studies of Information Systems, MIS Quarterly, 11 (3), 369-386.

36. Bettenhausen, K., Murnighan, J. K., (1986), The emergence of norms in competitive decision making groups, Administrative Science Quartley, 30 (350 – 372).

37. Beverly Little, "The Performance of Personnel Duties in Small Louisiana Firms: A Research Note,' Journal of Small Business Management (October 1986), pp. 66-69. Bibliography.

38. Beverly Little, "The Performance of Personnel Duties in Small Louisiana Firms: A Research Note,' Journal of Small Business Management (October 1986), pp. 66-69.

39. Bommer, W. H., Dierdorff, E. C., (2007), Does prevalence mitigate relevance? The moderating effect of group-level OCB on employee performance, Academy of Management Journal, 50 (6), 1481 – 1494.

40. Bourdieu, P., (1980), Social capital, Acts of the Social Science Research 31 (2-3).

41. Bourdieu, P., (1986), The forms of capital. In J. Richardson (Ed.) Handbook of Theory and Research for the Sociology of Education, New York: Greenwood.

42. Bourdieu, P., (1991), The peculiar history of scientific reason, Sociological Forum, 6 (1).

43. Boxall, P. (1998). Achieving competitive advantage through human resource strategy: Towards a theory of industry dynamics. Human Resource Management Review, 8(3), 265.

44. Boxall, P., Purcell, J., (2000), Strategic human resource management: where have we come from and where should we be going?, International journal of management review, 2 (1), 183 – 203. P a g e | 359

45. Breaugh, J. A., Employee recruitment: Current knowledge and important areas for future research, (2008), Human Resource Management Review, 18 (103 – 118).

46. Brennan, R., Al – Husan, F. B., (2009), Transferring western HRM practices to developing countries, Personnel Review, 38 (2), 104 – 123.

47. Brockner, J., & Wiesenfeld, B. M. (1996). An integrative framework for explaining reactions to a decision: The interactive effects of outcomes and procedures. Psychological Bulletin, 120(2), 189-208.

48. Brookfield, H.C., (1988). Fijian Farmers Each On His Own Land: The Triumph Of Experience Over Hope. The Journal of Pacific History, 23(1), 15-35.

49. Browne, M. W., (1968), A comparison of factor analytical techniques, Psychometrika, 33 (3) 267 - 333.

50. Bryman, A., Social Research Methods, Oxford University Press, Great Clarendon Street, Oxford.

51. Burns, A., (1963), Fiji, H.M. Stationery Office, Volume 8 of the Corona Library, University of Michigan.

52. Cabrera, E., & Bonache, J. (1999), An expert HR system for aligning organisational culture and strategy, Human Resources planning, 22 (1), 51 – 60.

53. Caillier, J. G., I want to quit: A closer look at factors that contribute to the turnover intentions of state government employees, State and local government review, 43 (2), 110- 122.

54. Callan, V. J., Gallois, C., & Ramsay, S. (1997). Social rules and attributions in the personnel selection interview. Journal of Occupational and Organizational Psychology, 70, 189-203.

55. Campbell, D.T., (1969), Reforms as experiments, American Psychologist, 24 (409 – 429).

56. Campbell, D.T., (1975), Degrees of freedom and the case study, Comparative Political Studies, 8 (178-193.

57. Carbery, R., & Garavan, T. N. (2007). Conceptualizing the participation of managers in career-focused learning and development: A framework. Human Resource Development Review, 6(4), 394-418.

58. Carlson, K. D., Connerley, M. L., Mecham III, R. L., (2002), Recruitment evaluation: The case for assessing the quality of applicants attracted, Personnel Psychology, 55 (2), 461 – 490. P a g e | 360

59. Cassar, V., & Bezzina, C. (2005). People must change before institutions can: A model addressing the challenge from administering to managing the Maltese Public Service. Public Administration and Development, 25(3), 205-215.

60. Cattell, R., & Vogelmann, S., (1977), A comprehensive trial of the scree and KG criteria for determining the number of factors. Multivariate Behavioral Research, 12(3), 289-325.

61. Cavaye (1996), Case study research: a multi – faceted research approach for IS, Information system journal, 6 (227 – 242).

62. Chan, D., & Schmitt, N. (2004). An agenda for future research on application reactions to selection procedures: A construct-oriented approach. International Journal of Selection and Assessment, 12, 9-23.

63. Chand, G. (1998). The labour market: A decade of restructuring. In G. Chand and V. Naidu (Eds.), Fiji: Coups, Crisis and Reconciliation, 1987-1997 (pp. 33-68). Fiji Institute of Applied Studies, Suva

64. Chand, G., Prasad, B. C., Reddy, M., & Chand, A. (1993). The woes of structural adjustment policies: The 1994 Fiji government budget. Fiji Institute of Applied Studies, Suva

65. Clifford, J, (1994) Job Analysis: Why Do it, and How Should It be Done?, Public Personnel Management Vol 23: 2.

66. Cohen-Charash, Y., & Spector, P. (2001). The role of justice in organizations: A meta- analysis. Organizational Behavior and Human Decision Processes, 86, 278-321.

67. Coleman, J. S., (1988), Social capital and schools, The Education Digest, 53 (8) 6.

68. Coleman, J. S., (1988), Social capital in the creation of human capital, The American Journal of Sociology, 94, 95 – 120.

69. Colquitt, J. A. (2001). On the dimensionality of organizational justice: A construct validation of a measure. Journal of Applied Psychology, 86, 386-400.

70. Colquitt, J. A., & Shaw, J. C., 2005, How should organizational justice be measured? In J. Greenberg and J. A. Colquitt (Eds.) The handbook of organizational justice: 113-152. Mahwah, NJ: Erlbaum.

71. Colquitt, J. A., Conlon, D. E., Wesson, M. J., Porter, C. O. L. H., & Ng, K. Y. (2001). Justice at the millennium: A meta-analytic review of 25 years of organizational justice research. Journal of Applied Psychology, 83, 425-445.

72. Colwill, N. L., Perry, R. P., & Struthers, C. W. (1992). An attributional analysis of decision making in a personnel selection interview. Journal of Applied Social Psychology, 22(10), 801-818. P a g e | 361

73. Commonwealth Secretariat Report, 2006.

74. Courty, P., & Marschke, G. (2004). Benchmarking performance. Public Finance and Management, 4(3), 288-316.

75. Creswell, J.W., 1994, Research Designs, Sage publications, London.

76. Crocombe, R. G., 1968. Improving Land Tenure. Technical Paper No. 159. South Pacific Commission, New Caledonia.

77. Crocombe, R. G., 1984. Registration, Security and Productivity in the Pacific Islands: Experiences and Potential. Land Tenure and Rural Productivity in the Pacific Islands. FAO, Rome; Institute of Pacific Studies, University of the South Pacific, Suva, Fiji; and South Pacific Regional Environment Programme, Noumea.

78. Crocombe, R. G., 1987. Overview: The pattern of change in Pacific Land tenures. Land Tenure in the Pacific. 3rd edition. University of the South Pacific, Suva.

79. Crocombe, R. G., 1989, The South Pacific: An Introduction. 5th edition. Institute of Pacific Studies, University of the South Pacific, Suva.

80. Cropanzano, R., & Greenberg, J. (1997). Progress in organizational justice: Tunnelling through the maize. International Review of Industrial and Organizational Psychology, 12, 317-372.

81. Davidson, M., & Friedman, R. A. (1998). When excuses don’t work: The persistent injustice effect among black managers. Administrative Science Quarterly, 43(1), 154-183.

82. De Cieri, H., Holland, P., Sheehan, C., (2007), Attracting and retaining talent: exploring human resource development trends in Australia, Human Resource Development International, 10 (3), 247 – 262.

83. Delery, J., E. (1998), Issues of fit in strategic human resource management: implications for research, Human resource management review, 8 (3), 289 – 309.

84. Department of Public Enterprises Report, Fiji, 1998).

85. Dessler, G., (2012), Human Resources Management, Prentice Hall, Saddle River, New Jersey.

86. Doing Business in Fiji: 2012 Country Commercial Guide for US Companies Report.

87. Dubin, R. (1976). Organizational effectiveness—Some dilemmas of perspective. Organization and Administrative Sciences, 7(2), 7-14. P a g e | 362

88. Duncan, R., Cutherbertson, S., & Bosworth, M. (1999). Pursuing economic reform in the Pacific. Pacific Studies Series No. 18. Manila: Asian Development Bank.

89. Dutt, B, Volavola, M., 1977. History of Land Surveying in Fiji. Ministry of Lands & Mineral Resources, Suva.

90. Eaton, C., 1988. Vakavanua land tenure and tobacco farming. Rural Fiji. Institute of Pacific Studies, University of the South Pacific.

91. Edwards & Bagozzi (2000). On the nature and direction of relationships between constructs and measures, Psychological Methods, 5(2), 155-174.

92. Edwards & Bagozzi (2000). On the nature and direction of relationships between constructs and measures, Psychological Methods, 5(2), 155-174.

93. Edwards & Bagozzi (2000). On the nature and direction of relationships between constructs and measures, Psychological Methods, 5(2), 155-174.

94. Eisenhardt, K. M., (1989), Building Theories from Case Study Research, Academy of Management, 14 (4), 532 – 550.

95. Eisenhardt, K. M., Bourgeois, L. J., (1988), Politics of strategic decision making in high velocity environment: Towards a mid-range theory, Academy of Management, 3 (1), 737 – 770.

96. Elkins, T. J., & Philips, J. S. (2000). Job context, selection decision outcome, and the perceived fairness of selection tests: Biodata as an illustrative case. Journal of Applied Psychology, 85, 479-484.

97. Ezigbalike, I. C., Benwell, G, L., 1992. "Cadastral `Reform'--What They Never Tell You About The Cultural Costs", Proceedings of the International Conference on Cadastral Reform `92. Dept. of Surveying & Land Information, The University of Melbourne, Melbourne.

98. Ferrier, W. J., (2001), NAVIGATING THE COMPETITIVE LANDSCAPE: THE DRIVERS AND CONSEQUENCES OF COMPETITIVE AGGRESSIVENESS, Academy of Management Journal, 44 (4), 858 – 877.

99. Fiji Government Report (2007)

100. Fiji Civil List, 2010

101. Fiji Civil List, 2011.

102. Fiji Blue Book, (1891 – 1937), Colonial Secretary’s Office, Suva.

103. Fiji Blue Book, (1917), Colonial Secretary’s Office, Suva. P a g e | 363

104. Fiji Blue Book, (1918), Colonial Secretary’s Office, Suva.

105. Fiji Department of Public Enterprise Report, 1998, pp6. Suva.

106. (Fiji Parliamentary Paper no. 110 of 2006)

107. Fiji Royal Gazette, 1874 - 1879, Fiji local government and subsidiary legislations, Government Printer, Suva.

108. Fiji Royal Gazette, 1875, Fiji local government and subsidiary legislations, Government Printer, Suva.

109. Fiji Royal Gazette, 1887, Fiji local government and subsidiary legislations, Government Printer, Suva.

110. Fiji Royal Gazette, No. 1, 1874.

111. Fiji Royal Gazette, No. 1, 1875.

112. Fiji Royal Gazette, No. 3, 1874.

113. Fiji: The State of the Nation and the Economy Report, (2008), Fiji government, Suva.

114. Fiske, S., & Taylor, S. E. (1991). Social cognition (2nd ed.). New York: McGraw- Hill Book Company.

115. Floyd, F. J., & Widaman, K. F., (1995), Factor analysis in the development and refinement of clinical assessment instruments, Psychological Assessment, 7(3), 286-299.

116. Folger, R., Konovsky, M., (1989) Effects of procedural and distributive justice on reactions to pay raise decisions, Academy of Management Journal, 32 (1), 115 – 129.

117. Ford, K. J., MacCallum, R. C., Tait, M., (1986), The application of exploratory factor analysis in applied psychology: a critical review and analysis, Personnel Psychology, 39.

118. Ford, R., McLaughlin, F., (1985), Nepotism, Personnel Journal, 64(9), 56 – 61.

119. Ford, R., McLaughlin, F., (1986), Nepotism: Boon or Bane, The Personnel Administrator, 31(11), 78 – 87.

120. France, P. (1969). Charter of the land: Custom and colonisation in Fiji. Melbourne: Oxford University Press.

121. France, P., 1969. The Charter of the Land: Custom and Colonisation in Fiji, Oxford University Press, Melbourne.

122. Frankema, E., (2010), The Colonial roots of land inequality: geography, factor endowments, or institutions, Economic History Review, 63 (2), 418 – 451. P a g e | 364

123. Freyd, M., (1923), The graphic rating scale, Journal of Educational Psychology, 14 (83–102).

124. Gatewood, R., Field, H. S., 1987, Human Resource Selection, Thomson Learning, Ohio.

125. Gatewood, R., Field, H. S., 2001, Barrick, M., Human Resource Selection, Thomson Learning, Ohio.

126. Gatewood, R., Field, H. S., 2011, Human Resource Selection, South – Western Cengage Learning, Mason, Ohio.

127. Gauntlett, D., (2011), Making is Connecting: The social meaning of creativity, from DIY and knitting to YouTube and Web 2.0, West Sussex, John Wiley and Sons Ltd.

128. Gay, L., Mills, G., & Airasian, P. (2006). Educational research: Competencies for analysis and applications. New Jersey: Pearson Education, Inc.

129. Gersick, C., (1998), Time and transition in work teams: Towards a new model of group development, Academy of Management Journal, 31 (9 – 41).

130. Ghebregiorgis, F., Karsten, L., (2007), Human resource management and performance in a developing country: the case of Eritrea, International Journal of Human Resources Management, 18 (2), 321 – 332.

131. Ghoshai, S., & Tsai, W., (1998), Social Capital And VALUE CREATION: THE ROLE OF INTRAFIRM NETWORKS, Academy of Management Journal, 41 (4), 464 – 476.

132. Gilliland, S. W. (1993). The perceived fairness of selection systems: An organizational justice perspective. Academy of Management Review, 18, 694-734.

133. Gilliland, S. W. (1994). Effects of procedural and distributive justice on reactions to a selection system. Journal of Applied Psychology, 79, 691-701.

134. Gilliland, S. W., & Chan, D. (2001). Justice in organizations: Theory, methods, and applications. In N. Anderson, D. S. Ones, H.K. Sinangil, and C. Viswesvaran (Eds.), Handbook of Industrial, Work, and Organizational Psychology: OrganizationalPsychology (Vol. 2, pp. 143-165). London: Sage.

135. Gilliland, S. W., & Hale, J. M. S. (2005). How can justice be used to improve employee selection practices? In J. Greenberg and J. A. Colquitt (Eds.), Handbook of Organizational Justice (pp. 411-438). Mahwah, N.J.: Lawrence Erlbaum Associates, Inc.

136. Gilliland, S. W., Truxillo, D. M., Steiner, D. D., (2004), The importance of organisational justice in personnel selection: defining when selection fairness really matters, International journal of selection and assessment, 12 (1/2). P a g e | 365

137. Gilmore, C. B., Fannin, W. R., (1982), The Dual Career Couple: A Challenge to Personnel in the Eighties, Business Horizons, 25(3), 36 – 42.

138. Glenn M. McEvoy, "Small Business Personnel Practices,' Journal of Small Business Management (October 1984), pp. 1-8.

139. Glenn M. McEvoy, "Small Business Personnel Practices,' Journal of Small Business Management (October 1984), pp. 1-8.

140. Gomez – Mejia, L. R., Nunez – Nickel, M., Gutierrez, I., 92001), The role of family ties in agency contracts, Academy of Management Journal, 44(1), 81 – 95.

141. Goodman, J., (1980), An examination of Referents used in the Evaluation of Pay, Organisational Behaviour and Human Performance, 12 (1), 170 – 195.

142. Gorsuch, R. L., (1983), Factor Analysis (2nd ed.), Hillsdale, NJ: Erlbaum.

143. Great Council of Chiefs Fiji, Resolutions of Bose Levu Vakaturaga, (1881 – 1890), National Archives of Fiji, Suva.

144. Greiling, D. (2005). Performance measurement in the public sector: the German experience, International Journal of Productivity and Performance Management, 54(7), 551 - 567.

145. Greiling, D. (2006). Performance measurement: A remedy for increasing the efficiency of public services? International Journal of Productivity and Performance Management, 55(6), 448-465.

146. Grimshaw, B., (1907), From Fiji to the Cannibal Islands, George bell & Sons, London.

147. Guion, R., M., (1965), Industrial Psychology as an Academic Discipline, American Psychologist, 20 (10), 815 – 821.

148. Guion, R.M. (1976). Recruiting, selection, and job placement. In M.D. Dunette (Ed.), Handbook of industrial and organizational psychology. Chicago: Rand McNally.

149. Guion, R. M. (1987). Changing views for personnel selection research. Personnel

Psychology, 40, 199-213.

150. Guttman, L. A., (1954), A new approach to factor analysis: The radix. In P. F. Lazarsfeld (Ed.) Mathematical thinking in the social sciences, New York: Columbia University Press. P a g e | 366

151. Harold, C. M., & Ployhart, R. E. (2004). The applicant attribution-reaction theory (AART): An integrative theory of applicant attributional processing. International Journal of Selection and Assessment, 12(1-2), 84-98.

152. Harrington, B. (2001). Organizational performance and corporate social capital: A contingency model. Social Capital of Organizations, 18, 83-86.

153. Harrington, B. (2008), Pop Finance: Investment Clubs and the New Investor Populism, Princeton University Press.

154. Harris, S., Sutton, R., (1986), Functions of parting ceremonies in dying organisations, Academy of Management, 29 ( 5 – 30).

155. Hauck, V. E., (1983), Employing Relatives in Government, Journal of Collective Negotiations in the Public Sector, 12(4), 283 – 294.

156. Heinrich, C. J. (2004). Improving public sector performance management: One step forward, two steps back. Public Finance and Management, 4(3), 317-315.

157. Hewitt, T., Wangwe, S., & Wield, D. (2002). Seeing eye to eye: Organizational behaviour, brokering and building trust in Tanzania. Public Administration and Development, 22, 97-108.

158. Hofstede, G., Cultural Dimensions in Management and Planning, Asia Pacific Journal of Management, 1984.

159. Holden, N., (2002), Cross-cultural Management: A Knowledge Management Perspective, Harlow: Prentice Hall.

160. Hu & Bentler (1999). Cut off criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives, Structural Equation Modeling, 6(1), 1-55.

161. Hu & Bentler (1999). Cut off criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives, Structural Equation Modeling, 6(1), 1-55.

162. Hu & Bentler (1999). Cut off criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives, Structural Equation Modeling, 6(1), 1-55.

163. Huselid, M. A., (1995), The impact of human resource management practices on turnover, productivity, and corporate financial performance, Academy of management journal, 38 (3) 635.

164. Huselid, M. A., Delaney, J. T., (1996), The impact of human resource management practices on perceptions of organisational performance, Academy of Management Journal, 39 (4), 949 – 969. P a g e | 367

165. Huseyin, A., Ali, B., Ekiz, E., (2006), The effects of nepotism on human resource management; The case of three, four and five star hotels in Northern Cyprus, The International Journal of Sociology and Social Policy, Patrington: 26, pg 295.

166. Iles, P., (2007), Employee resourcing and talent management, Annual Review of Psychology, 46 (237 – 364).

167. Jääskeläinen, A., & Lönnqvist, A. (2009). Designing operative productivity measures in public services. VINE, 39(1), 55-67.

168. Jääskeläinen, A., Uusi -Rauva, E. (2011), Bottom-up approach for productivity measurement in large public organisations, International journal of productivity and performance management, 60 (3), 252 – 267.

169. Jackman, Robert W. and Ross A. Miller. 1998, Social Capital and Politics, Annual Review of Political Science, 1 47-73.

170. Jackson, S. E., Schuler, R. S., (1995), Understanding human resources management in the context of organisations and their environments. In M. R. Rosenzweig & L. W.Porter (eds.), Annual review of psychology, 46 (237 – 264).

171. Jackson, S. E., Schuler, R. S., Rivero, J. C., (1989), Organisational characteristics as predictors of personnel practices, Personnel Psychology, 42 (4), 727 – 785.

172. Johnston, J., (2005), Performance measurement uncertainty on the Grand Canal: ethics and productivity conflicts between social and economic agency, International Journal of Productivity and Performance Management, 54 (7), 595 – 612.

173. Kepes, S., & Delery, J., E. (2007), HRM Systems and the Problem of Internal Fit [online]. In: Boxall, Peter (Editor); Purcell, John (Editor); Wright, Patrick (Editor). Oxford Handbook of Human Resource Management, The. Oxford; New York: Oxford University Press, 2007: 385-404.

174. Kerlinger, F.N., (1973), Foundations of behavioural research (2nd ed.), New York: Holt, Rhinehardt, and Winston.

175. Kerlinger, F.N., Lee, H.B., (2000), Foundations of Behavioral Research (4th ed.), Fort Worth, TX: Harcourt College Publishers.

176. Kidder, T., (1982), Soul of a new machine, New York: Avon.

177. Kim, P. S., & Hong, K. O. (2005). Searching for effective HRM reform strategy in the public sector: Critical review of WPSR 2005 and suggestions. Public Personnel Management, 35(3), 199-215. P a g e | 368

178. Kimbugwe, K., Lewis, R., & James, N. (2009). Labour inputs in public sector productivity: Methods, issues and data. Economic & Labour Market Review, 3(4), 52-59.

179. Klehe, U. (2004). Choosing how to choose: Institutional pressures affecting the adoption of personnel selection procedures. International Journal of Selection and Assessment, 12(4), 327-342.

180. Klimoski, R., (1991), Theory presentation in human resource management, Human Resource Management Review, 1 (4), 253 – 271.

181. Klingner, D. E., Campos, V. P., (2001), Strengthening Personnel Management in Developing Countries: Lessons Learned, Lessons Forgotten and an Agenda for Action, Public Organisation Review, 30 (1).

182. Klingner, D. E., Campos, V. P., (2002), Building Public HRM Capacity in Latin America and the Caribbean: What Works and What Doesn’t, Public Organisation Review, 2 (4) 349.

183. Kluger, A. N., & Rothstein, H. R. (1993). The influence of selection test type on applicant reactions to employment testing. Journal of Business and Psychology, 8, 3-25.

184. Knouse, S. B. (2002). The role of attribution theory in personnel employment selection: A review of the recent literature. The Journal of General Psychology, 116(2), 183- 196.

185. Konig, C. J., Klehe, U., Berchtold, M., & Kleinmann, M. (2010). Reasons for being selective when choosing personnel selection procedures. International Journal of Selection and Assessment, 18(1), 17-27.

186. Kotobalavu, J., (2009), Resolving the crisis in Fiji: a personal perspective, In Firth, J., Fraenkel, J., Lal, B., The 2006 military takeover in Fiji: a coup to end all coups?, 375-84.

187. Kramar, R., Bartram, T., De-Cieri, H. D., Noe, R. A., Hollenbeck, J. R., Gerhart, B., & Wright, P. M., (2011), Human Resources Management in Australia, McGraw Hill, Sydney.

188. Lal, B., (1988), Power and Prejudice: The Making of the Fiji Crisis, Wellington: New Zealand Institute of International Affairs.

189. Lal, B., (2008), A time bond lies buried: Fiji’s road to independence, 1960 – 1970, The Australian National University, Canberra, ACT.

190. Landau, S., Everitt, B., (2004), A handbook of statistical analyses using SPSS, CRC Press LLC, Florida. P a g e | 369

191. Lasaqa, I., (1984), The Fijian people: before and after independence, Australian National University Press, Canberra.

192. Laumann et al., 1978, p. 458), Community structure as interorganizational linkages. Annual Review of Sociology, 4(1), 455-484.

193. Lawson, S., (1990) The Myth of Cultural Homogeneity and Its Implications for Chiefly Power and Politics in Fiji, Comparative Studies in Society and History, 32(4), 795– 821.

194. Lawson, S., (1993), Ethnic Politics and the State in Fiji, Working Paper 135, Canberra: Peace Research Centre, Australian National University, Canberra

195. Lazar, A., Zinger, A., Lachterman, B., (2007), The influence of prefeedback selection justice on perceptions of overall procedural justice and organisational attractiveness in a real life selection procedure, European journal of work and organisational psychology, 16 (1), 94 – 109.

196. Lee, E., Kim, S., (2006), Best Practices and Performance-Based HR System in Korea, Seoul Journal of Business, 12 (1).

197. Lee, T. H., Gerhart, B., Weller, I., Trevor, C. O., (2008), Understanding voluntary turnover: Path-specific job satisfaction effects and the importance of unsolicited job offers, Academy of Management Journal, 51 (4), 651 – 671.

198. Leenders, R. A. J., Gabbay, S. M., (1999), Corporate social capital and liability, Boston: Kluwer Academic Press.

199. Lei & Wu (2007). Introduction to Structural Equation Modeling: Issues and Practical Considerations, Educational Measurement: Issues and Practice, 33-43.

200. Lei & Wu (2007). Introduction to Structural Equation Modeling: Issues and Practical Considerations, Educational Measurement: Issues and Practice, 33-43. Karl’s corner standardized coefficients.

201. Lei & Wu (2007). Introduction to Structural Equation Modeling: Issues and Practical Considerations, Educational Measurement: Issues and Practice, 33-43.

202. Leonard, B., (1988), Synergistic design for case studies: Longitudinal single site and replicated multiple site. Paper presented at the National Science Foundation Conference on Longitudinal Research Methods in Organisation, Austin, TX.

203. Leonard-Barton, D. (1988). Implementation as mutual adaptation of technology and organization. Research policy, 17, 251 - 267. P a g e | 370

204. Levine, E., (1977), Legal Aspects of Reference Checking for Personnel Selection, The Personnel Administrator, Berea: 22(9), pg. 14.

205. Linna, P., Pekkola, S., Ukko, J., & Melkas, H. (2010). Defining and measuring productivity in the public sector: Managerial perceptions. The International Journal of Public Sector Management, 23(3), 300-320.

206. Li-Yun, S., Aryee, S., Law, K., (2007), High- Performance Human Resources Practices, Citizenship Behavior, and Organisational Performance: A Relational Perspective, 50 (3), 558 – 577.

207. LLoyd, D.T., 1982, Land Policy in Fiji, Occasional Paper No. 14. Department of Land Economy, University of Cambridge, London, UK.

208. MacCallum, (1992). Model modifications in covariance structure analysis: The problem of capitalization on chance, Psychological Bulletin, 111(3), 490-504.

209. MacCallum, (1994). Alternative strategies for cross-validation in covariance structure models, Multivariate Behavioral Research, 29(1), 1-32.

210. Macnaught, T. J., (1974), Chiefly civil servants? Ambiguity in district administration and the preservation of a Fijian way of life 1896 – 1940, The Journal of Pacific History, 9 (1), 3 – 20.

211. Macnaught, T. J., (1982), The Fijian Colonial Experience: A study of the neo- traditional order under British colonial rule prior to World War II, The Australian National University, Canberra.

212. Malhotra N, Morris T, 2009, Heterogeneity in Professional Service Firms, Journal of Management Studies, Vol: 46, ISSN: 0022-2380, Pages: 895-922.

213. Maiyaki, A. A., Mokhtar, S. S., (2011), Determinants of customer behavioural responses in the Nigerian retail banks: Structural equation modelling approach, African Journal of Business Management, 6 (4), 1652-1659.

214. Mara, Ratu Sir Kamisese CGMG, KBE, CF KStJ, MSD (1997), The Pacific Way: A Memoir, University of Hawaii Press, Honolulu.

215. Markus M L (1989) Case Selection in a Disconfirmatory Case Study. In The Information Systems Research Challenge: Qualitative Research Methods, Volume 1, (Cash J I and Lawrence P R Eds.), pp.20-26, Harvard Business School Press, Boston, MA.

216. Marsh, Hau, & Wen (2004). In search of golden rules, Structural Equation Modelling, 11(3), 320-341. P a g e | 371

217. Martell, K., Carroll, S. J., (1995), How strategic is HRM, Human Resources Management, 34 (2), 253 – 267.

218. Martin, D., & Friedman, R. A. (1998). When excuses don’t work: The persistent injustice effect among black managers. Administrative Science Quarterly, 154(43), 154.

219. Martin, S. S., (2008), Relational and economic antecedents of organisational commitment, Personnel Review, 37 (6), 589 – 608.

220. Masterson, S. S., (2001), A trickle-down model of organisational justice: relating employees and customers perceptions of reactions to fairness, Journal of applied psychology, 86 (4), 594 – 604.

221. Maykut, P, Morehouse, R, 2003, Beginning Qualitative Research – A philosophic and practical guide, Falmer Press, London.

222. McDonald & Ringo Ho (2002). Principles and practice in reporting structural equation analyses, Psychological Methods, 7(1), 64–82.

223. McDonald (2010). Structural models and the art of approximation. Perspectives on Psychological Science, 5(6), 675–686.

224. McDowall, A., Fletcher, C., (2004), Employee development: an organisational justice perspective, Personnel Review, 33 ( 1), 8 – 29.

225. McEvoy, G., (1984), Small business personnel practices, Journal of small business management, 1 - 8).

226. McMahon, R.G.P., Davies, L.G., (1991), Financial Reporting and Analysis in Smaller Growth Enterprises: Evidence on Practice in the North-East of England, Accounting and Finance Research Paper No. 91/4, Discipline of Accounting and Finance, The Flinders University of South Australia, Bedford Park.

227. Mellahi, K., & Frynas, J. (2003). An exploratory study into the applicability of Western HRM practices in developing countries: An Algerian case study. International Journal of Commerce and Management, 13(1), 61-80.

228. Merrett, D., In Nand’s, AA (Ed.), Towards better governance: Fiji’s recent reform experiences, Development Problems and Prospects in Pacific Islands States, pp.245 - 258.

229. Mesch, D. J., Perry, J. L., (1995), Bureaucratic and Strategic HRM: An empirical assessment in the federal government, Journal of Public Administration and Theory, 5 (385 – 402).

230. Miles, M. B., Huberman, M. A., (1994), Qualitative Analysis: An expanded source book (2nd ed.) Thousand Oaks, CA: Sage. P a g e | 372

231. Miles, R., Snow, C. C., (1984), Designing strategic human resource systems, Organisation dynamics, 13 (36 – 52).

232. Millman, J., Von Glinow, M. A., Nathan, A., (1991), Organizational Life Cycles and Strategic International Human Resource Management in Multinational Companies, Academy of Management Review 16 (318 – 39).

233. Milne, R. S., (1975), The Pacific Way – Consociational Politics in Fiji, Pacific Affairs, 48 (3), 413 – 431.

234. Mintzberg, H., McHugh, A., (1985), Strategy formation in an adhocracy, Administrative Science Quartley, 30 (1), 160 – 197.

235. Mirable, R, J, The Power of Job Analysis; Training; Apr 1990; 27, 4; ABI/INFORM Global pg. 70.

236. Moorman, R. H., (1991), Relationships between organisational justice and organisational citizenship behaviours: Do fairness perceptions influence employee citizenship? Journal of Applied Psychology, 76, (845 – 855).

237. Morton, Thomas James (1944), A bibliography, Vuna , Taveuni.

238. Moynagh, M., (1978). Land Tenure in Fiji's sugarcane districts since the 1920's. Journal of Pacific History, 13, 1-2.

239. Namazie, P., Frame, P., (2007), Development in human resource management in Iran, International journal of human resource management, 18 (1), 159 – 171.

240. Nankervis, A., Compton, R., Baird, M, 2002, Strategic Human Resource Management, Thomson, Victoria.

241. Native Land Trust Board (1989), Annual Report, Government Printing Press, Suva.

242. Nayacakalou, R. N., (1975), The Fijian System of kinship and marriage, The Journal of the Polynesian society, 64 (1), 44 – 55.

243. Nayacakalou, R. N., 1971. Fiji: Manipulating the System. Land Tenure in the Pacific. Oxford University Press, Melbourne.

244. Nelton, S., (1998), The bright side of nepotism, Nations Business, Washington, 86(5).

245. Niehoff, B. P., Moorman, R. H., (1993), Does perceived organisational support mediate the relationship between procedural justice and organisational citizen behaviour, Academy of Management Journal, 43 (3), 351 – 357. P a g e | 373

246. Norton, R., (1977), Race and Politics in Fiji, University of Queensland Press, St Lucia, Queensland.

247. Norton, R., (2002), Accommodating Indigenous Privilege: Britain’s Dilemma in Decolonising, The Journal of Pacific History, 37 (2), 133 – 156.

248. O’Connor, B. P., (2000), SPSS and SAS Programs for Determining the Number of Components Using Parallel Analysis and Velicers MAP Test, Behaviour Research Methods, Instruments, & Computers, 32 (396 – 402).

249. Oduro, A., Barr, A.B., 2004, Ethnic and Wage Determination in Ghana, Rochester.

250. Organisation for economic corporation and development, 2001, www.fijitimes.com.fj

251. Ornati, O.A., Eisen, M. J., (1981) Are You Complying with EEOC's New Rules on National Origin Discrimination?, Personnel. New York, 58 (2), pgs 9, 12.

252. Ouchi, W.G. (1980), Markets, bureaucracies, and clans. Administrative Science Quarterly, 25: 129-141.

253. Page, C., Meyer, D., (2000), Applied Research Design for Business and Management, McGraw-Hill, Sydney.

254. Page, C., Wilson, M., Inkson, K., & Meyer, D. (2003). “It’s the situation I’m in”: The importance of managerial context to effectiveness. The Journal of Management Development, 22(10), 841-861.

255. Pari, N., & Frame, P. (2007). Developments in human resource management in Iran. The International Journal of Human Resource Management, 18(1), 159-171.

256. Perrow, C., (1986), Complex organisation: A critical essay, New York: Harper & Row.

257. Perry, C., Coote, L., (1994), Processes of a case study research methodology: tool for management development? Australia and New Zealand Association for Management Annual Conference, Victoria University of Wellington.

258. Pettigrew, A. M., (1985), Culture of politics in strategic decision making and change, In Strageic Decision Making in Complex Organisations, J. M. Pennings (ed.), San Francisco: Jossey – Bass.

259. Pfeffer, J., (1982), New directions in organisational theory: Problems and Prospects, New York: Oxford Univ Press. 264 pp.

260. Pfeffer, J., (1997), Power in organisations, New York: Pitman. P a g e | 374

261. Pidd, M., (2005), Perversity in public service performance measurement, International Journal of Productivity and Performance Management, 54 (5/6), 482 – 493.

262. Pinfield, L., (1986), A field evaluation of perspectives on organisational decision making, Adminstrative Science Quarterly, 31 (365 – 388).

263. Ployhart, R E., & Ryan, A. M (1997), Toward and explanation of applicant reactions: An examination of organization justice and attribution framework. Organizational Behavior and Human Decision Processes, 72, (308 -335).

264. Poole, M. J. F., (1982), Personnel Management in Third World Countries, Personnel Review, 11 (4).

265. Portes, A., (1998), Social Capital: Its origins and applications in modern sociology, Annual Review of Sociology, 24 (1 – 24).

266. Prasad, P.C., 1984. Fiji: Sugar Cane Production & Land Tenure. Land Tenure and Rural Productivity in the Pacific Islands. FAO, Rome; Institute of Pacific Studies, University of the South Pacific, Suva, Fiji; and South Pacific Regional Environment Programme, Noumea.

267. Preacher & MacCallum (2003). Repairing Tom Swift's electric factor analysis machine, Understanding Statistics, 2(1), 13-43.

268. Price, J. L., Mueller, C. W., (1986), Handbook of organisational measurement, Marshfield, MA: Pitman Publications.

269. Putman, R. D., (1993b), What makes democracy work?, National Civic Review, 82(2), 101 – 107.

270. Putman, R. D., (1996a), The decline of civil society: How come? So what?, The John L. Manion Lecture, Canadian Center for Management Development, Ottawa, Ontario, 22 February.

271. Ramage, R. O., (1968), Localisation of the Fiji Civil Service, Civil Service Fiji, Suva.

272. Ramesh, S (2007), Fiji, 1987 – 2007: The story of four coups: Working paper, Sydney.

273. Ratuva, S., (2013), Politics of preferential development Trans-global study of affirmative action and ethnic conflict in Fiji, Malaysia and South Africa, ANU Press, Canberra.

274. Ravuvu, A., 1987. The Fijian Ethos. Institute of Pacific Studies, University of the South Pacific, Suva. P a g e | 375

275. Ravuvu, A., 1988. Development or Dependence: The Pattern of Change in a Fijian Village. Institute of Pacific Studies, University of the South Pacific, Suva.

276. Reed, C., Bruce, W. M., (1993), Dual-career couples in the public sector: A survey of personnel policies and practices, Public Personnel Management, Washington, 22(2), 187 – 200.

277. Renzulli, L. A., Aldrich, H., Moody, J., (2000), Family matters: Gender, Networks, and Entrepreneurial Outcomes, Social Forces, 79 (2), 523 – 546.

278. Richard, O. C., Murthi, B. P. S., Ismail, K., (2007), The impact of racial diversity on intermediate and long term performance: The moderating role of environmental context, strategic management journal, 28 (12), 1213 – 1233.

279. Rocco, T. S., & Plakhotnik, M. S. (2009). Literature reviews, conceptual frameworks and theoretical frameworks: Terms, functions and distinctions. Human Resource Development Review, 8(1), 120-130.

280. Routledge, D., (1974), The negotiations leading to the cession of Fiji, 1874, The journal of imperial and commonwealth history, 2 (3), 278 – 293.

281. Routledge, D., (1985), Matanitu: the struggle for power in early Fiji, Institute of Pacific Studies, USP, Suva.

282. Ryan, A. M., & Ployhart, R. E. (1997). Toward an explanation of applicant reactions: An examination of organizational justice and attribution frameworks. Organizational Behaviour and Human Decision Processes, 72(3), 308-335.

283. Rynes, S. L. 1991. Recruitment, job choice, and post-hire consequences. In M. D. Dunnette (Ed.), Handbook of industrial and organizational psychology (2nd. ed.): 399 – 444. Palo Alto, CA: Consulting Psychologists Press.

284. Salm, S. J., Falola, T., (2002). Culture and Customs of Ghana. Westport, CT: Greenwood Press.

285. Sanders, K., (2008), The impact of individual and shared employee perceptions of HRM on affective commitment: Considering climate strength, (2008), Personnel Review, 37 (4), 412 – 425.

286. Sarker, A. E., Pathak, R. D., (2003), Public Enterprise Reform in Fiji Islands, Public Organisation Review, 3 (55 – 57).

287. Scarr, D., 1984, Fiji: A short history. George Allen & Unwin, Sydney.

288. Schramm, W., (1971), Notes on case studies for instructional media projects: Working paper for Academy of Educational Development, Washington DC. P a g e | 376

289. Schroth, H. A., & Shah, P. P. (2000). Procedures: Do we really want to know them? An examination of the effects of procedural justice on self-esteem. Journal of Applied Psychology, 85(3), 462-471.

290. Schuler, R. (2000). The internationalization of human resource management. Journal of International Management, 6(3), 239-260.

291. Schuler, R., Jackson, S., (1987), Linking competitive strategies with Human Resource Management Practices, The academy of management executive, 1 (3), 207 – 219.

292. Scott, A. M., (2008), When Cupid Strikes in the Office, Employee Benefit Plan Review, New York, 62(8), 26 – 28.

293. Sekaran, U, 2003, Research Methods for Business – a skill building approach, Wiley, Melbourne.

294. Sekaran, U., & Bougie, R. (2010). Research methods for business: A skill building approach (5th ed.). West Sussex: John Wiley & Sons Ltd.

295. Sharpham, J. 2000. Rabuka of Fiji: The authorised biography of Major-General Sitiveni Rabuka. Rockhampton: Central Queensland University Press.

296. Simpson, H. (2009). Productivity in public services. Journal of Economic Surveys, 23(2), 250-276.

297. Singer, M. S., (1993). Fairness in personnel selection. Aldershot, New Zealand: Avebury.

298. Sisson, K. (1990), Introducing human resource management journal, Human Resource Management Journal, 1(1-11).

299. Skinner, D., Managers perceptions of strategic HR change, Personnel Review, 26 (6), 467 – 484.

300. Spate, O.H.K., 1959, The Fijian People: Economic Problems & Prospects. A report for the Legislative Council of Fiji, Council Paper No. 13 of 1959, Government Printing Press, Suva.

301. State Services Decree, 2009, Suva.

302. Stavrou, E. T., Brewster, C., & Charalambous, C. (2010). Human resource management and organizational performance in Europe through the lens of business systems: Best fit, best practice or both. The International Journal of Human Resources Management, 21(7), 933-962. P a g e | 377

303. Stevens (1995). Confirmatory factor analysis of the ITBS, Structural Equation Modeling

304. Stone, R.J, 2002, Human Resources Management, Wiley, Melbourne, Australia.

305. Stone, R.J, 2013, Human Resources Management, Wiley, Melbourne, Australia.

306. Storey, J., Sisson, K., (1993), Managing Human Resources and Industrial Relations, Open University Press, Buckingham.

307. Sutherland, W., (1992), Beyond the politics of race, an alternative history of Fiji to 1992, Australian National University, Canberra.

308. Sutherland, W., (2000), The problematics of reform and the Fijian question, in Akram-Lodhi, A. H. (eds) Confronting Fiji Futures, Asia Pacific Press, Australian National University, Canberra.

309. Sutton, R., Callaham, A., (1987), The stigma of bankruptcy: Spoiled organisational image and its management, Academy of Management Journal, 30 (405 – 436).

310. Swanson, R. A. (2007). Theory framework for applied disciplines: Boundaries, contributing, core, useful novel and irrelevant components. Human Resource Development Review, 6(3), 321-339.

311. Szreter,S. (2000). Social Capital, the Economy, and Education in Historical Perspective, in Social Capital: Critical Perspectives, edited by S. Baron, J. Field and T. Schuller, pp56-78.

312. Tavola, H., (1992), Secondary education in Fiji, Institute of Pacific Studies, USP, Suva.

313. Tayeb, M., (1995), The competitive advantage of nations: the role of HRM and its socio-cultural context, The international journal of human resource management, 6 (3), 588 – 605.

314. Taylor, M. S., Bergmann, T. J., (1987), Organizational recruitment activities and applicants' reactions at different stages of the recruitment process, Personnel Psychology, 40, 261-285.

315. Tessema, M. T., & Soeters, J. L. (2006). Challenges and prospects of HRM in developing countries: Testing the HRM-performance link in the Eritrean civil service. The International Journal of Human Resource Management, 17(1), 86-105. Watson, E. (2007). Who or what creates? A conceptual framework for social creativity. Human Resource Development Review, 6(4), 419-441. P a g e | 378

316. The 1970 Constitution of Fiji, Government Printer, Suva

317. The 1990 Constitution of Fiji, Government Printer, Suva.

318. The 1997Ammenmendt Constitution of Fiji, Government Printer, Suva.

319. Thomas, N., (1985), Planets around the Sun, University of Sydney, Sydney.

320. Thurstone, L., L. (1935), The vectors of mind, University of Chicago Press, Chicago.

321. Thurstone, L., L. (1947), Multiple factor analysis: A development and expansion of the vectors of mind, University of Chicago Press, Chicago.

322. Tolbert, P. S., Zucker, L. G., (1996), The Institutionalisation of Institutional theory, Cornell University ILR School, http://digitalcommons.ilr.cornell.edu/articles.

323. Tullar, W., (1991), Theory development in human resource management, Human Resource Management Review, 1 (4), 317 – 323.

324. Tzanakis, M., (2013), Social capital in Bourdieu’s, Coleman’s and Putman’s theory: empirical evidence and emergent measurement issues, Educate, 13 (2), 2 – 23.

325. Tzeng, O. C. S., (1992), On reliability and number of principal components, Perceptual Motor Skills, 75 (929 – 930).

326. Vallance, S., (1996), The human Resource Crises in Fiji’s Public Sector: trouble in paradise? Public Administration & Development (1986 – 1998); Feb; 16, 1; ABI/INFORM Global pg 91.

327. Vosikata, S., Faifua, D. (2012), HR Selection Distortions: A Theoretical Framework for the Fiji Public Service, Journal of US – China Public Administration, 9 (7) pp 775- 790.

328. Vosikata, S, Reddy, M., Prasad, B., Sharma, P., &Duncan, R., (2004). Understanding reforms in Fiji. Pacific Institute of Advanced Studies in Development and Governance (PIAS-DG), University of the South Pacific (USP), Fiji.

329. Weiner, B. (1986). An attributional theory of motivation and emotion. New York: Springer-Verlag.

330. West, F. J., (1960), Problems of Political Advancement in Fiji, Pacific Affairs, 33 (1), 23 – 37.

331. White, C. M., (2003), Historicizing educational disparity: colony policy and Fijian educational attainment, History of education, 32 (4) 345 – 365. P a g e | 379

332. Williams, T., Calvert, J., (1859), Fiji and the Fijians, D. Appleton and Company, Broadway, NY.

333. Williams, A., (2001), Economic marginalism and the microstate, the impact of donor conditionality requirements in international aid programs on a SHRM training project in Cook Islands, The international journal of human resource management, 12 (1), 99 – 108.

334. Williams, R, 2002, Managing Employee Performance, Thomson Learning, Berkshire House, High Holborn, London.

335. Williams, R.S, 2002, Managing Employee Performance, TJ International, Cornwall.

336. Williamson, I. P., 1982. The Cadastral Survey Requirements of Developing Countries in the Pacific Region - with particular reference to Fiji. The Survey Review, 26(206): 355- 366.

337. Wong, P., & Weiner, B. (1981). When people ask “Why” questions, and the heuristics of attributional search. Journal of Personality and Social Psychology, 40(4), 650.

338. Woolcock, M., Narayan, D., 2000, Social Capital: Implications for Development Theory, Research, and Policy, the World Bank research observer, 15 (2) 225 – 249.

339. Wright, B. E., Pandey, S. K., Public Service motivation and the assumption of person – organisation fit; testing the mediating effect of value congruence, Administration and society, 40 (5), 502 – 521.

340. Wright, P. M., Snell, S. A., (1998), Toward a unifying framework for exploring fit and flexibility in strategic human resources management, Academy of Management Review, 23 (4), 756 – 772.

341. Wright, P. M., & McMahan, G. C. (1992). Theoretical perspectives for strategic human resource management, Journal of Management, 18(2), 295.

342. Yeganesh, H., Zhan, S., (2008), An examination of human resource management practices in Iranian public sector, Personnel Review, 37 (2) 203 – 221.

343. Yin, R. K. (1994), Case study research: Design and methods (2nd ed.), Newbury Park, CA: Sage Publications.

344. Yin, R. K., (1984), Case study research: Design and methods (1st ed.), Beverly Hills, CA: Sage Publications.

345. Yin, R. K., (1989), Case study research: Design and methods, Applied Social Research Series, Vol. 5. London: Sage.

346. Yin, R. K., (1993), Applications of case study research. Newbury Park, CA: Sage P a g e | 380

347. Yin, R. K., (2004), The case study anthology, Newbury Park, CA: Sage Publications.

348. Yin, R. K., (2009), Case study research. Design and Methods, Sage Publications, Thousand Oaks, 4th ed.

349. Zhang, Y., & Shaw, J. R. (2012). Publishing in AMJ—Part 5: Crafting the methods and results. Academy of Management Journal, 55, 8-12.

350. Zikmund, W. G., (2003), Business research methods, South Western Cengage learning, Mason, Ohio, US.