Accounting for Uk Retailers' Success

THE UNIVERSITY OF MANCHESTER - APPROVED ELECTRONICALLY GENERATED THESIS/DISSERTATION COVER-PAGE

Electronic identifier: 16349

Date of electronic submission: 27/09/2015

The University of Manchester makes unrestricted examined electronic theses and dissertations freely available for download and reading online via Manchester eScholar at http://www.manchester.ac.uk/escholar.

This print version of my thesis/dissertation is a TRUE and ACCURATE REPRESENTATION of the electronic version submitted to the University of Manchester's institutional repository, Manchester eScholar.

Approved electronically generated cover-page version 1.0

ACCOUNTING FOR UK RETAILERS’ SUCCESS: KEY METRICS FOR SUCCESS AND FAILURE

A thesis submitted to the University of Manchester for the degree of Doctor of Philosophy in the Faculty of Humanities

2015

TARLOK N. TEJI

MANCHESTER BUSINESS SCHOOL

Contents

LIST OF TABLES ...... 9 LIST OF FIGURES ...... 10 ABSTRACT ...... 11 DECLARATION AND COPYRIGHT ...... 12 ACKNOWLEDGEMENTS ...... 13 PREFACE ...... 14 Chapter 1: INTRODUCTION AND OBJECTIVES ...... 15 1.1. Background ...... 15

1.2. Research problem ...... 16

1.3. Research value ...... 18

1.4. Research framework ...... 18

1.5. Thesis chapters ...... 19

Chapter 2: UNDERSTANDING UK RETAIL ...... 22 2.1. Introduction ...... 22

2.2. Background ...... 23

2.3. Change and retail change models ...... 25

2.3.1. Wheel of retailing ...... 26

2.3.2. Retail Life Cycle...... 26

2.3.3. Accordion ...... 27

2.3.4. Environmental...... 27

2.3.5. Ecological ...... 27

2.3.6. Conflict ...... 27

2.3.7. Model variations and assessment ...... 27

2.3.8. Retail adaptation and resilience...... 29

2.4. Deconstructing the kaleidoscope ...... 30

2.4.1. Introduction ...... 30

2.4.2. Regulation ...... 31

2.4.3. Global influence ...... 33

2.4.4. Socio-demographic ...... 34

2.4.5. Physical retail dimensions ...... 35

2.4.6. Digital retail dimensions ...... 40

2.4.7. Customer Lifetime Value ...... 45

2.4.8. Summary ...... 46

2.5. Conclusion ...... 47

Chapter 3: LITERATURE REVIEW ...... 49 3.1. Introduction ...... 49

3.2. Literature review map ...... 50

3.3. Financial information ...... 52

3.3.1. Introduction ...... 52

3.3.2. Accounting and financial information ...... 52

3.3.3. Usefulness and predictive capabilities debate ...... 54

3.3.4. Ratio Analysis ...... 55

3.3.5. The current ratio and industry differences ...... 57

3.3.6. Going beyond ratio analysis ...... 57

3.3.7. Summary ...... 62

3.4. Governance, risk and company beta ...... 63

3.4.1. Risk and governance ...... 63

3.4.2. Theories of corporate governance ...... 64

3.4.3. Corporate governance rules and reporting ...... 64

3.4.4. Risk ...... 66

3.4.5. Risk and the company beta ...... 67

3.4.6. Summary ...... 69

3.5. Business failure prediction models ...... 70

3.5.1. Introduction ...... 70

3.5.2. The five model types ...... 73

3.5.3. Results of failure model studies ...... 78

3.5.4. Testing UK retail data ...... 80

3.5.5. Summary ...... 83

3.6. Failure theories and retail research ...... 83

3.6.1. Introduction ...... 83

3.6.2. General theories implying failure ...... 83

3.6.3. Business and retail failure research ...... 84

3.6.4. Success and failure continuum ...... 84

3.6.6. Defining retail failure ...... 86

3.7. Business and management research ...... 88

3.7.1. Introduction ...... 88

3.7.2. Marketing measurement frameworks ...... 89

3.7.3. Retail measurement frameworks ...... 90

3.7.4. Performance measurement theory ...... 90

3.7.5. Strategic control...... 90

3.8. Performance measurement and metrics ...... 91

3.8.1. Introduction ...... 91

3.8.2. Financial measurement frameworks ...... 92

3.8.3. Business measurement frameworks ...... 92

3.8.4. Balanced Score Card ...... 93

3.9. Summary of literature review ...... 94

Chapter 4: METHODOLOGY AND METHODS ...... 96 4.1. Introduction ...... 96

4.2. Research objectives related to methods ...... 98

4.3. The pragmatic philosophical world view ...... 100

4.3.1. Ontology and epistemology ...... 100

4.4. Challenges to the approach and methods ...... 103

4.4.1. Introduction ...... 103

4.4.2. Analysing published information ...... 103

4.4.3. Model building ...... 104

4.4.4. Surveys ...... 104

4.4.5. Focus groups ...... 105

4.4.6. Researcher experience and influence ...... 105

4.5. Multiple case study design in three phases ...... 105

4.5.1. Introduction ...... 105

4.5.2. Phase one ...... 108

4.5.3. Coding confirmability testing ...... 111

4.5.4. Phase two ...... 112

4.5.5. Phase three...... 116

4.6. Applying the grounded theory research strategy ...... 119

4.7. First order analysis ...... 119

4.8. Second order analysis ...... 121

4.9. Application of an explanatory matrix ...... 121

4.9.1. Dimensional analysis and the explanatory matrix ...... 121

4.9.2. Explaining the procedural steps ...... 123

4.10. Inductive and interpretive presentation of findings ...... 123

4.11. Summary ...... 124

Chapter 5: FIRST ORDER ANALYSIS ...... 125 5.1. Introduction ...... 125

5.2. The metrics overview ...... 125

5.3. Sifting the metrics ...... 125

5.3.1. Introduction ...... 125

5.3.2. A sifting matrix ...... 127

5.4. Thematic analysis ...... 131

5.5. Customer ...... 133

5.5.1. Customer numbers ...... 133

5.5.2. Customer engagement and satisfaction ...... 134

5.6. Sales ...... 135

5.6.1. Statutory reporting ...... 135

5.6.2. Internal reporting of sales ...... 136

5.6.3. Like-for-Like Sales ...... 137

5.6.4. Summary ...... 137

5.7. People ...... 138

5.7.1. Introduction ...... 138

5.7.2. People engagement ...... 138

5.8. Margin ...... 139

5.8.1. Introduction ...... 139

5.8.2. Failed retailers and margin measurement ...... 139

5.8.3. Successful retailers and margin measurement ...... 140

5.8.4. Summary ...... 141

5.9. Cash ...... 141

5.10. Debt ...... 142

5.11. Property ...... 143

5.12. Stock control ...... 145

5.13. Buying and products ...... 147

5.14. Business model ...... 148

5.15. Risk and regulation ...... 151

5.16. Profit ...... 152

5.17. Return on investment ...... 153

5.18. Brands ...... 154

5.19. Digital ...... 155

5.20. Competition ...... 155

5.21. Supply chain ...... 156

5.22. Market share ...... 156

5.23. Growth ...... 157

5.24. Summary of focus areas and metrics ...... 157

Chapter 6: SECOND ORDER ANALYSIS ...... 160 6.1. Introduction ...... 160

6.2. Twenty Focus Areas ...... 161

6.3. The Twenty 20 ...... 162

6.3.1. Introduction ...... 162

6.3.2. The vital few ...... 163

6.4. Consistently persistent use of metrics ...... 166

6.5. Adapting metrics ...... 169

6.5.1. Introduction ...... 169

6.5.2. Changes in performance metrics exemplars ...... 170

6.5.3. Summary ...... 191

6.6. Being trust intelligent ...... 192

6.6.1. Introduction ...... 192

6.6.2. A framework for understanding metric disclosure practice ...... 192

6.6.3. Reading the conceptual framework ...... 195

6.6.4. The four stages of the journey ...... 196

6.7. An alternative approach to narrative analysis ...... 200

6.8. Mapping the case studies ...... 201

6.9. Summary ...... 204

Chapter 7: CONCLUSION...... 206 7.1. Reflections on the thesis ...... 206

7.1.1. Introduction ...... 206

7.1.2. Aim and objectives met ...... 206

7.2. Theory, knowledge and contribution ...... 209

7.3. Thoughts on further research ...... 212

REFERENCES ...... 213 APPENDICES ...... 226

Appendix 1: List of 194 retailers in distress 2005-2010 ...... 226

Appendix 2: Ratios used - 82 variables in failure prediction models ...... 228

Appendix 3: Exploratory test survey ...... 230

Appendix 4: Original template coding structure ...... 233

Appendix 5: Semi-structured interview meeting agenda ...... 234

LIST OF TABLES Page Table 1.1: Research framework and thesis chapters 19 Table 2.1: Top 50 distressed retailers 2000-2015 23 Table 2.2: Six models of retail change 28 Table 3.1: Shop Direct Group ratio analysis from FAME database 59 Table 3.2: Top 20 UK listed retailers – beta values 68 Table 3.3: Timeline of failure prediction models 71 Table 3.4: Failure prediction models seminal papers’ analysis 72 Table 3.5: Predictive accuracy claims of 11 model types summarised 78 Table 3.6: Eight business failure and risk models applied to Harrods 81 Table 3.7: Eight business failure and risk models applied to Shop Direct Group 81 Table 3.8: Z-score model applied to the 194 UK retail ‘failure’ population 82 Table 3.9: Top ten retailers from the 194 ‘failure’ population classification scores 82 Table 4.1: Objectives, methods and analysis 99 Table 4.2: Top 5 to top 200 UK retail population by market share 108 Table 4.3: Summary of template coding structure 111 Table 4.4: Summary of retail company participant interviews 116 Table 5.1: Attributes of good quality information for retail performance metrics 129 Table 5.2: Properties of resilience applicable to retail performance metrics 130 Table 5.3: Measurement of customer satisfaction and engagement comments 135 Table 5.4: Comments on like-for-like sales measurement 137 Table 5.5: Measuring people engagement via surveys and forums 138 Table 5.6: Margin measurement comments in successful retailers 140 Table 5.7: First ten focus areas and related metrics 158 Table 5.8: Second ten focus areas and related metrics 159 Table 6.1: UK retailer metrics the vital few 163 Table 6.2: The top 20 marketing metrics 164 Table 6.3: The vital few compared with the top 20 marketing metrics 165 Table 6.4: Inter-channel retail performance metric example report 167 Table 6.5: The minimum retail ‘core’ disclosure standard 168 Table 6.6: Retailer C adapting retail performance metrics over time 183

LIST OF FIGURES Page Figure 1.1: Research process framework 18 Figure 2.1: Mega trend and shift to flat spend trend 1997-2014 25 Figure 2.2: Long-term consumer spend trends in the UK 1957-2007 36 Figure 2.3: Medium-term consumer spend trends in the UK 2006-2013 37 Figure 2.4: Share of wallet – household weekly spend categories 2013 37 Figure 2.5: Consumer digital dynamics’ overlay 42 Figure 3.1: Literature review layer map 51 Figure 3.2: Ratio inter-relationships 56 Figure 3.3: Typical cash conversion cycle manufacturing example 58 Figure 3.4: Smart retailer cash conversion cycle example 58 Figure 3.5: New Look Retail Group Consolidated Balance Sheet 2011 61 Figure 3.6: New Look Retail Group Consolidated Income Statement 2011 62 Figure 3.7: Corporate Governance developments timeline 65 Figure 3.8: Tesco wheel 93 Figure 3.9: Sainsbury’s wheel 94 Figure 4.1: Pure and applied research continuum 98 Figure 4.2: Research philosophy’s influence on method choice 101 Figure 4.3: Two dimensions and four paradigms for social theory analysis 102 Figure 4.4: Three phase research design for this thesis 106 Figure 4.5: Detailed research process steps for this thesis 107 Figure 4.6: Transcript excerpt – a first reading 117 Figure 4.7: Process of metrics’ assimilation 118 Figure 4.8: Explanatory matrix for dimensional analysis 122 Figure 5.1: The abundance of metrics’ data extract 126 Figure 5.2: A sifting matrix for retail performance metrics 128 Figure 5.3: Sifting matrix – the focus quadrant 128 Figure 5.4: Sifting illustration of 20 relatively ranked retail performance metrics 131 Figure 6.1: Retailer C key retail performance metrics 1988-89 172 Figure 6.2: Retailer C key retail performance metrics group summary 173 Figure 6.3: Retailer C key group summary sales 2002-03 175 Figure 6.4: Retailer C key group profit & loss account 2002-03 176 Figure 6.5: Retailer C management accounts home shopping 2005 178 Figure 6.6: Retailer C management accounts financial services 2005 179 Figure 6.7: Retailer C performance dashboard 2010-11 181 Figure 6.8: Retailer A retail performance metrics and targets 2013-2017 185 Figure 6.9: Retailer D retail performance metrics in branch reports 2014 187 Figure 6.10: Being trust intelligent through retail metric disclosure 194 Figure 6.11: The conceptual framework journey as a matrix 195 Figure 6.12: Retailer D annual report extract one 198 Figure 6.13: Retailer D annual report extract two 198 Figure 6.14: Retailer D annual report extract three 199 Figure 6.15: Retailer A, journey to becoming trust intelligent 202 Figure 6.16: Retailer B, journey to becoming trust intelligent 202 Figure 6.17: Retailer C, journey to becoming trust intelligent 203 Figure 6.18: Retailer D, journey to becoming trust intelligent 203

ABSTRACT

THE UNIVERSITY OF MANCHESTER ABSTRACT OF THESIS submitted by Tarlok N. Teji for the Degree of PhD and entitled ‘ACCOUNTING FOR UK RETAILERS’ BUSINESS SUCCESS: KEY METRICS FOR SUCCESS AND FAILURE’ Date: 30th September 2015

This thesis provides an understanding of retailers’ performance metrics and measurement. In doing so it lays bare the over reliance on historic published accounting reports as the de facto standard for retail performance reporting. In addition, it exposes the weakness in retail accounting reports as well as retail failure prediction models that are dependent on financial ratios as key variables. This thesis also casts light on the non-financial performance metrics used by retailers.

All retailers use performance metrics but do not always report them in a coherent and defined way to give a transparent picture of their actual performance. The subject of performance, and metrics in particular, can be approached from multiple disciplines, yet there is an absence of detailed guidance or discussion of retail performance metrics, for retail boards, in any literature. To comprehend a UK retailer’s performance, it is argued that there is a prerequisite to understand the full context of the UK retail landscape, and the multitude of metrics, both financial and non-financial, this brings into play when discussing performance measurement.

Accordingly, the objectives of this thesis were to identify: what retail performance metrics are used by retail boards to manage their performance; what these boards claim about their performance in the public domain; and what disconnect there may be between these two areas. A pragmatic worldview in the interpretative tradition frames the research epistemology. This inductive approach is supported by a multiple case study design strategy using informed grounded theory to conduct research into six case companies (four successful and two failed) in order to discover the retail performance metrics they use and report.

The findings show an abundance of metrics in use at retail boardroom level and a ‘sifting matrix’ is devised to cluster the metrics to aid comprehension and ranking into the 20 focus areas which retail boards consider important. These focus areas provide a basis for a suite of metrics, ‘the vital few’ within which six were found to be consistently and persistently used that could form an industry standard. In addition, there was evidence that retailers adapt their metrics as they change, giving substance to the notion of adaptive resilience in performance measurement. Any disconnect between metric use and disclosure was explored through a conceptual framework, ‘a journey matrix’, where retailers are on a journey to becoming trust intelligent with their disclosure of retail performance metrics. The transparent disclosure of retail performance metrics provides the explicit link to gaining trust and demonstrating good governance practice implicit within stewardship theory. The ‘journey matrix’ is also proposed as an alternative developmental viewpoint for analysing retailers’ annual reports and accounts.

The development and disclosure of retail performance metrics lacks guidance on definitions, calculation bases and recommended disclosure. Without guidance, the voluntary proliferation of selective reporting is likely to render performance, as published by retailers themselves, opaque and confusing. This thesis starts the debate about board level retail performance metrics research and provides a framework to assist retail boards to evaluate what they use and what they disclose in their journey to gain the trust of stakeholders.

DECLARATION AND COPYRIGHT

No portion of the work referred to in the thesis has been submitted in support of an application for another degree or qualification of this or any other university or other institute of learning. i. The author of this thesis (including any appendices) owns certain copyright or related rights in it (the “Copyright”) and he has given The University of Manchester certain rights to use such Copyright, including for administrative purposes. ii. Copies of this thesis, either in full or in extracts and whether in hard or electronic copy, may be made only in accordance with the Copyright, Designs and Patents Act 1988 (as amended) and regulations issued under it or, where appropriate, in accordance with licensing agreements which the University has from time to time. This page must form part of any such copies made. iii. The ownership of certain Copyright, patents, designs, trademarks and other intellectual property (the “Intellectual Property”) and any reproductions of copyright works in the thesis, for example graphs and tables (“Reproductions”), which may be described in this thesis, may not be owned by the author and may be owned by third parties. Such Intellectual Property and Reproductions cannot and must not be made available for use without the prior written permission of the owner(s) of the relevant Intellectual Property and/or Reproductions. iv. Further information on the conditions under which disclosure, publication and commercialisation of this thesis, the Copyright and any Intellectual Property and/or Reproductions described in it may take place is available in the University IP Policy (see http://documents.manchester.ac.uk/DocuInfo.aspx?DocID=487), in any relevant Thesis restriction declarations deposited in the University Library, The University Library’s regulations (see http://www.manchester.ac.uk/library/aboutus/regulations) and in The University’s policy on Presentation of Theses.

ACKNOWLEDGEMENTS

I would like to thank all the people who have helped me during the completion of this thesis. In particular my supervisors, Professor Dominic Medway and John Pal who have encouraged, supported and relentlessly provided thorough and detailed feedback on my progress over five years. Also a thank you to Dr Daniel Hampson who provided independent reliability confirmation for the coding methods used.

In addition, I would like to provide a specific individual note of thanks to Professor Peter McGoldrick for getting me started on the research process at Manchester Business School and encouraging me to continue to completion.

Clearly, without the research participants this thesis would not be possible so an acknowledgement to the retail board directors who gave their time and shared their corporate information, as well as the retail experts who provided interviews. Too many to name and most of whom have chosen to remain anonymous.

A final thank you to my family and friends who have supported me throughout the thesis and particularly my critical friend for the tireless reading of chapter drafts.

PREFACE

I retired in 2010 re-entering academia to continue improving my understanding of retailing. This interest began after leaving school in 1975 when I entered a grocery management training programme with International Stores, then a part of BAT Industries, to work in and run supermarkets. At about the same time I also read Economics with industrial experience at The City University, London and continued to work part-time for an area manager across his West London stores. After graduation I joined a big professional practice, now called Deloitte Touche Tomatsu LLP. Training as an auditor and qualifying as a chartered accountant, I also worked with them in management consulting and worked for about two years in the Kenya office.

After the overseas experience, I re-entered the world of retail. Joining Tesco in senior management roles as Head of Corporate Audit, running the retail finance department as Retail Financial Controller, then CFO of the property division. During this time I completed the ‘fast track’ executive management training, part of which consisted of a part-time MBA by distance learning at The Institute of Retail Studies, University of Stirling. What became a ten year commercial experience was enhanced by a move to ASDA Group plc undertaking various senior roles such as Director of Supply Chain and Director of Business Efficiency to assist with the business ‘recovery and breakout’ periods and eventual sale to Walmart.

I then returned to professional practice as lead partner for the UK retail industry sector group at Deloitte LLP whilst also being Enterprise Risk Services Partner for the North of England. During this time I was fortunate to work with most UK retailers and many overseas retail companies. After 15 years as an experienced partner, I retired in 2010 to focus on retail and digital research as well as being an advisor and investor in technology start-ups.

I am a Fellow of the Institute of Chartered Accountants in England & Wales and hold a practising certificate as a registered auditor.

Chapter 1: INTRODUCTION AND OBJECTIVES

1.1. Background

England has been described as ‘a nation of shopkeepers’ by Napoleon (O'Meara, 1822) and today retail has been noted as one of the most competitive business sectors in the UK. “Retailing is a very competitive, low-margin industry and you have to get up every day and earn your bread” Sir Terry Leahy, Sunday Times Business Section p5, 13th June 2010. (Leroux, 2010)

Eight of the top 100 global retailers are from the UK (Deloitte, 2015) with Tesco listed as 2nd after Walmart, yet many retailers operating in the UK continue to struggle as businesses and get into financial difficulty with some failing completely.

The overall global business environment has become increasingly turbulent and volatile (Naisbitt and Aburdene, 1990) over the last 20 years making it harder for even the top performing retailers to avoid distress (Altman and Hotchkiss, 2006, Cameron et al., 1988).

The most recent example of fall from grace is illustrated by the number one UK retailer Tesco, whose share price has collapsed wiping millions off its market capitalisation (Odeluga, 2014, Sparks, 2014). ‘Tesco to face criminal probe after £263 million hole found in profits. Tesco’s senior bosses face the potential threat of jail after the Serious Fraud Office launched a criminal investigation into the ailing supermarket’s £263m accounting scandal’. Russell Lynch, www.independent.co.uk, Wednesday 29 October 2014 (Lynch, 2014).

This clearly demonstrates the current relevance and need to understand retailer performance and attendant reporting (the profit overstatement refers to ‘supplier income’ yet the company has not reported or explained the management judgment or quantum involved in the forecast and estimation of ‘supplier income’ including the potential impact on trading performance e.g. £118m over accrued at the half year to 2015 year-end).

Another example is Woolworths Group plc (‘Woolworths’), where in April 2008 the Board including the Chairman, Chief Executive and Finance Director signed the Annual Report and Accounts (“ARA”) (Woolworths, 2008) and seven months later Woolworths went into administration. The Woolworths illustration is discussed in more detail in Chapters 3 and 4 where the emphasis is specifically about accounting disclosure and imminent failure (i.e. end of life, for definitions of: distress; financial distress; and failure both narrow and broad, see Chapter 3, section 3.6.6). However, it was Woolworths’ music and video distribution business ‘EUK’ that created the tipping point that led to administrators being appointed (Burden, 2014) and this is not apparent from the ARA.

There are sophisticated accounting and reporting requirements of International Financial Reporting Standards (‘IFRS’) (IASB, 2010) and the Companies Act 2006 (HMSO, 2009) yet the question remains unanswered about why failures such as Focus, Comet and Woolworths are not foreseen or predictable and also why some retailers seem unable to deliver sustainable performance. More importantly, retailers appear not to learn from the failures of the past or more recent ‘distress’ situations (McGrath, 1999, Edmondson, 2011, McGrath, 2011).

The accounting profession in the UK and Europe had been debating the usefulness of the current corporate reporting requirements for many years which resulted in the introduction of IFRS. However, as Rake claims, these are still seen as inadequate to deal with the needs of modern day businesses (see the quote below from a press release following a meeting of the ‘big six’ accountancy firms to discuss corporate reporting). “We all believe the current model is broken. We’re not in a very happy situation” Mike Rake, Chairman KPMG International, Financial Times 8th November 2006 (Jopson, 2006)

The Tesco accounting scandal, the reporting model being ‘broken’ and the unexpected failure of retail companies such as Woolworth could be seen as a disconnect between: financial accounting and reporting; governance and the marketing led business strategy; and how retailers are managing and controlling their business performance. In spite of recent efforts by the professional bodies (ICAEW, 1975, ICAS, 2010, ICAEW, 2013, ICAEW, 2014) to address the ‘reporting gap’ and attempt to create business confidence in the audited financial statements (President-ICAEW, 2014), the Tesco example suggests there is still more work to do.

1.2. Research problem

Traditional accounting and reporting being inadequate begs the question whether or not there are other signals (e.g. corporate governance ‘red flags’ such as the chairman and chief executive roles being held by the same person or the auditors commenting on going concern), or performance metrics, that could give an indication that a retailer is showing signs of good or poor performance. Or more generally, are there ‘retail performance metrics’ that successful retailers use to measure performance?

The essence of this thesis is that reviewers of a retailer’s corporate entity performance tend to look in the wrong place at the wrong things i.e. they look to historic published financial information in the form of the ARA and concentrate their primary assessment through the use of ratio analysis. This thesis shows the inadequacy of relying solely on historic published financial information as a means of understanding a retailer’s performance and proposes particular non- financial retail performance metrics for gaining insight.

The UK has a long history of retail excellence e.g. Harrods, with some world class retailers within the Top 100 Global Retailers e.g. J. Sainsbury and Alliance Boots. What is it that these retailers do or how do they manage and monitor performance that sets them apart from the retail failures? Why haven’t ‘distressed’ retailers learnt from the success of excellent retailers? (Sull, 2005, Gino and Pisano, 2011, Hinshaw and Kasanoff, 2012).

The primary sources for information when looking for performance metrics are typically: the published ARA; brokers’ reports; financial media commentary; and company and industry reviews. Notwithstanding the availability of all this information in a digital age what is reported seems inadequate for a proper distinction between success and distress.

Having spent 35 years working with retailers and ten of these years in senior finance roles at two world class retailers (Tesco plc and ASDA Group plc), personal experience suggests that there are internal measures used to steer the business,

One area of enquiry for this thesis is to understand the difference between what companies publish as their trading performance and what retail performance metrics they use internally to manage this yet do not publish. Ancillary to this is why mandatory external reporting rules have not raised ‘red flags’ for retailers in ‘distress’. In conclusion, given the inadequacy of published financial information, the research problem is identifying non-financial1 ‘retail performance metrics’2, actually used by retail boards to manage their companies that can help understand a retailer’s trading performance.

The objectives can be summarised as three key statements:  To identify what retail performance metrics are used by retail boards to manage their performance. In so doing: o To identify any commonalities amongst the performance metrics used by retail board directors. o To determine whether or not retail performance metrics change over time.  To identify what retailers claim about their performance in the public domain; and  To explore any disconnect between the two objectives above i.e. the connectivity between the performance metrics retail boards’ use and those they publicly report.

1 Non-financial refers to any metric that is a non-mandatory requirement for public reporting. In this thesis it could refer to financial information such as weekly takings i.e. weekly sales including VAT. 2 Retail performance metric refers to any information or statistic used to manage and monitor performance that is a non- mandatory reporting requirement such as employee turnover percentage or customer numbers.

1.3. Research value

The economic and social impact of retail failure appears to go further than just shareholder loss. The impact can be seen and felt in: unemployment; loss of pension; supplier’s distress; town, shopping centre and property value deterioration; and loss of revenue for local and national government. For example, when Woolworths collapsed in 2008, 30,000 employees became unemployed (Edgar, 2013) creating a nationwide problem and political focus (Portas, 2011). Consequently, the research may have important implications for the UK economy as 65% of Gross Domestic Product (ONS, 2015a) consumption is in retail and consumer business categories. Similar levels of GDP are noted for other developed western economies with the USA as the largest consumer economy at 70%. Therefore the research results may have some general universal applicability. Helping retailers to avoid failure may prevent the high street, shopping centres and hence communities from becoming blighted as well as preserving employment and pensions for large numbers of the population. Most importantly being able to understand a retailer’s performance may give some comfort to investors, employees, suppliers, government and communities.

1.4. Research framework

The research framework follows the six step flow shown in Figure 1.1 below. The six steps in the research process framework are reconciled to the chapters of this thesis in Table 1.1. Figure 1.1: Research process framework.

Validation check have objectives been met and is research problem answer 1. Problem, impact satisfactory? Yes – stop & publish & value No – go round the loop again & redesign research

6. Findings, 2. Literature analysis & review of theories conclusion & models

3. Critical review 5. Research for relevance to design the problem

4. Identify research gap

Table 1.1: Research framework and thesis chapters

Research step Chapter 1. Problem, impact & value Chapter 1 Introduction and objectives Chapter 2 Understanding UK retail

2. Literature review Chapter 2 Understanding UK retail 3. Critical review Chapter 3 Literature review 4. Research gap 5. Research design Chapter 4 Methodology and methods

6. Results, analysis & conclusion Chapter 5 1st Order Analysis Chapter 6 2nd Order Analysis Chapter 7 Conclusion

A brief summary of the contents of each chapter are given in the next section.

1.5. Thesis chapters

CHAPTER 1: INTRODUCTION AND OBJECTIVES

This chapter introduces the research problem and outlines the research objectives as statements. It explains the potential impact of the research and possible value. It explains the thesis structure in relation to the research framework and chapter headings and provides a synopsis of each chapter as an overview for the reader.

CHAPTER 2: UNDERSTANDING UK RETAIL

The aim of this chapter is to put the UK retail landscape into context. It is considered important to understand the retail landscape to be able to put retailers’ performance into perspective. The chapter begins with examining theories and models of retail change. This is then compared to a description of retail observations through a retail development assessment, tracing back over 25 years using six facets to identify Mega trends and Paradigm shifts. The chapter concludes that there is no one retail change model that appears to explain observations in the UK retail landscape and even hybrid models are falsifiable. One specific recent idea of adaptive resilience currently applied to high streets and town centres is discussed. This may have broader application that the thesis explores in Chapter 6 in relation to retail performance metrics. In addition, there is a discussion of the interdependent multi-relational nature of the six facets for Mega trends and Paradigm shift coupled with the velocity of change in recent times, concluding that it is difficult to forecast the retail outlook with any degree of certainty. Without understanding this retail context, it is unlikely that a retailer’s performance will make sense. Although ‘deconstructing the kaleidoscope’ does not claim to be a theory or a model it provides a way to make sense of today’s UK retail landscape.

CHAPTER 3: LITERATURE REVIEW

The literature review chapter is long and complicated. The reason being that thoughts on performance metrics can be provided by many academic disciplines (e.g. accounting & finance, governance & risk, retail marketing, business and management etc.) as well as other non- academic commentators (e.g. consultants, brokers, trade bodies, regulators). A literature review map is given illustrating the complexity and diversity of academic thought. Each academic discipline is searched for prior study of retail performance metrics with respect to good (success) and bad (failure) performance. There is considerable general business and management research and less so in the retail marketing discipline. The critical review of the literature reveals a lack of, retail boardroom non-financial retail performance metrics, research providing a gap for study.

CHAPTER 4: METHODOLOGY AND METHODS

The methodology and methods chapter details the epistemology and ontology of the research design. The aim is to understand what performance metrics retail board members actually use and consider important to apply their time and attention when making decisions. This is done by using a multiple (six) retailer case study strategy applying informed grounded theory with a particular emphasis on using a dimensional analysis matrix. This is essentially an inductive approach taking a pragmatic world view rooted in the interpretive tradition. The research has been conducted using replicated semi-structured interviews with a cross section of retail board directors within case companies and across a cross section of retail companies. Corporate information over a ten year time horizon has been collected together with secondary data. In addition five retail experts have been interviewed individually, forming a panel of experts, to discuss the findings and get their experience of retail performance measurement and metrics. A mix of qualitative and quantitative tools and methods have been applied to complete the case study research the primary analytical tool being NVivo when searching for commonality in the retail performance metrics used.

CHAPTER 5: FIRST ORDER ANALYSIS

This chapter sets out the ‘first order analysis’ which represents the abundance of retail performance metrics discussed by retail board directors. A sifting framework in the form of a matrix to assess the retail performance metrics for quality and resilience is proposed to assist with the relative ranking of the metrics. The retail performance metrics are then assimilated into a manageable thematic grouping of 20 Focus Areas (FAs). What transpires from this analysis is that retail board members use a mix of financial and non-financial metrics to make decisions about their retail company’s performance with an emphasis on the ‘non-financial’ retail performance metrics. The most striking feature of the analysis is the commonality and relative ranking of the FAs within and across the retail company case studies, although it should be noted that the specific retail performance metrics used varied from case to case. The FAs may present

20 a new way to examine and study retail organisation from a process and performance perspective given that this is what current retail boards actually do.

CHAPTER 6: SECOND ORDER ANALYSIS

This chapter takes the previous analysis to the next level of abstraction and accordingly is titled second order analysis. The data is interrogated using the rigour of the dimensional matrix process until salient dimensions surface. The first of these dimensions is the FAs representing a new way to understand a retailer’s performance given the high degree of commonality within and across the case companies. Then the 20 favoured metrics associated with the 20 FAs (i.e. the ‘vital few’) are noted as the next salient dimension. Another salient dimension noted was the consistently persistent set of ‘six of the best’ retail performance metrics, from within the 20 ‘vital few’, that were used over a long period of time. There is some evidence from the case studies that when it comes to retail performance metrics, retailers are on a journey. The first salient aspect of this journey is that they seem to be adapting their performance metrics as their retail business changes giving some credibility to the notion of adaptive resilience. The primary salient dimension in relation to the journey is the reporting of the performance metrics. Here the retailers are on a journey to becoming ‘trust intelligent’ through improving their capability in the reporting of retail performance metrics. A matrix of quality of reporting and competency is proposed to assess this journey. It is shown that this journey is not necessarily a one directional trajectory and the practice and attitude depends on the four defined stages of a retailer’s development.

CHAPTER 7: CONCLUSION

The concluding chapter summarises the thesis noting that currently it is difficult to understand a retailer’s performance from the traditional published ARA and that retail performance metrics provide better insight. The literature review confirms via accounting and finance research that the ARA and business failure prediction models in particular have significant weaknesses and cannot be solely relied upon for identifying good (successful) or bad (failure) performance. Although there is an abundance of general business, management, retail and performance framework literature, none of it actually provides specific retail performance metrics that retail boards use. The marketing literature is, however replete with suggestions for marketing metrics which have some overlap but are mainly marketing department focused rather than a holistic retail boardroom view. New ways to consider retail companies through the FAs and ‘the vital few’ metrics as well as the consistently persistent ‘six of the best’ set are suggested. A ‘journey matrix’ for assessing a retailer’s journey to becoming trust intelligent through the reporting of retail performance metrics is given as an alternative way to consider narrative reporting in ARA. The contribution to knowledge and theory advancement is the explicit link that is created by the transparent disclosure of retail performance metrics which build trust and demonstrate good governance that is implicit in stewardship theory. The chapter concludes by discussing the research analysis and further ways to enhance the research including additional confirmatory studies.

Chapter 2: UNDERSTANDING UK RETAIL

2.1. Introduction

This chapter sets the scene for the thesis by putting into context the UK retail landscape. To comprehend a retailer’s performance, it is argued that there is a prerequisite to understand the full context of the UK retail landscape, and the multitude of metrics, both financial and non- financial, this brings into play when discussing performance measurement.

Consequently, the UK retail landscape is examined first by a critical review of theories and models of change focusing specifically on retail change. The aim of this is to assess the relevance of these theories and models to observations of change. These observations are made by looking back 25 years to identify changes by reference to the ‘driving forces’ (Porter, 1980, Wrigley and Brookes, 2014). This identification process is to determine and specify which driving forces make up the ‘Mega trends’ (Naisbitt and Aburdene, 1990) and what this researcher calls ‘Paradigm shifts’ (Kuhn, 1962, Kuhn, 1977, Kuhn, 1990) (defined as structural or permanent shifts in direction or behaviour) to get to a holistic view, rather than a single sided analysis e.g. retail supply. In other words there is a need to understand the UK retail landscape to be able to put retailer performance into context.

The way in which the observation is done and described is by using six facets covering: Regulation; Global influence; Socio-demographics; Physical retail dimensions; Digital retail dimensions; and Customer Lifetime Value (CLV). Each of these facets are described in turn with examples of why they are important perspectives individually but also recognising their interdependencies and multi-dimensional relationships. In addition, the turbulence in the retail landscape is considered with reference to the velocity of transition now, compared to 25 years ago.

The chapter concludes by pointing out that single sided analysis, however interesting academically or otherwise, is unlikely to provide a holistic view of the retail landscape. Moreover, any predictions beyond 12 months based thereon will be hard to make with any attendant decision making prone to error. Furthermore the thesis shows the vagaries of over reliance on published financial information3 (examples of Tesco and Woolworths given in Chapter 1: Introduction) and establishes a need to understand how good retailers actually manage their performance rather than conduct superficial assessments based on what they provide in the public domain.

3 Financial information, accounting reports and ARA are used interchangeably in the thesis to represent all ‘historic published financial information’ and derivations thereof e.g. ratio analysis.

2.2. Background

The retail landscape is changing and in the UK there have been a number of examples of retail distress over the last 15 years (see Table 2.1). Consumers have been changing their shopping habits and shifting spending patterns over time as evidenced since records were kept of basket spend (Deloitte, 2009, ONS, 2014b). Some of the names in Table 2.1 are now represented on the retail landscape under different fascia names, e.g. Homestyle and Hamsard, and some have been through a number of ownership changes e.g. Wickes and Bhs; but these issues aside, the table shows that no one sector of retail is immune from distress (Cameron et al., 1988, Altman and Hotchkiss, 2006, Mellahi and Wilkinson, 2010).

Table 2.1: Top 50 distressed retailers 2000-2015

Retailer Retailer Retailer Retailer Retailer Name Sector Name Sector Name Sector Name Sector Name Sector Somerfield g Threshers alc Clinton Cards b Findel hs Wyevale grd Woolworths m Otto:Freemans hs Jessops e TJ Morris m Hamsard clo GUS hs Netto g Empire Direct hs Allied Carpets flr Thorntons bkr Comet e C&J Clark f Blockbuster mv Powerhouse e Laura Ashley clo HMV Waterstones mvb Peacocks clo Blacks Leisure spo Virgin Music mv Allsports spo Wickes diy ScS furn Magnet furn Zavvi mv Kleeneze hsp Bhs clo Toys R Us toy Fenwick dept TJ Hughes dept Aga-Rayburn hmwr Focus diy GAME toy Barratts f Box Clever misc Ottakers b Mfi furn Homestyle furn Stylo f Phones4U e Topps Tiles flr JJB spo Mothercare clo Mosaic Fashion clo Adams clo Land of Leather furn Note: Table 1 shows the largest 50 ‘distressed’ retailers in column rank order (by their revenue size in 2005). Sector classifications used are from Mintel (Mintel, 2005): alc (alcohol); b (books); bkr (bakery); clo (clothing); dept (department stores); diy (do it yourself); e (electrical); f (footwear); flr (flooring); furn (furniture); g (grocery); grd (garden); hmwr (homewears); hs (home shopping); m (mixed); misc (miscellaneous); mv (music & video) mvb (music, video & books); spo (sports); toy (toys).

‘In November 2008, Woolworths went into administration and closed in the UK. They were not the first retailer to come under pressure as the great recession hit, but Woolworths’ physical and symbolic presence was immense. The closure of 800+ high street stores represented a massive hit on high streets and ignited a debate on the state of the high street, the retail sector, and policy towards high streets and retailing. Woolworths were soon followed into administration and closure by other large - and small – retailers, not exclusively trading on the high street. Retail itself was in crisis, and as shop vacancy levels rose sharply and gaps appeared in centres across the country, the state of the high street became a focus of public, media and then political attention’ (Findlay and Sparks, 2014, p13)

The Woolworths story is just one example of a UK retail landscape in the process of rapid transition (Coca-Stefaniak and Carroll, 2014, Wrigley and Brookes, 2014), with many retail failures and the traditional shopping high street in decline, especially post the 2008 financial crisis. This has profound implications for a consumer-based economy, which represents about 65% of gross domestic product by expenditure (ONS, 2015a), as well as creating social and political issues. Accordingly, working towards an understanding of the ‘Mega trends’ and ‘Paradigm shifts’ in the retail landscape and the drivers (Porter, 1980, Williams, 2014, Wrigley and Brookes, 2014) of retailer performance, in particular, becomes an important topic of analysis. As Williams (2014, p2) notes:

‘I have been involved in planning and retail development throughout my professional career. All too frequently I have found that opinion-formers’ attitudes to new development, and particularly its effects, are founded on populism rather than sound analysis of the facts. Similarly, and more worryingly, policy makers’ and decision takers’ approach to new developments is, all too often, based not on careful and objective consideration of sound evidence but on wild frequently unsubstantiated assertions […] It is always pleasing to come across proper objective research that examines how retail trends are affecting existing retail structures’.

Much analysis of retailing, post-2008 focuses on the effect of the ‘credit crunch’ and austerity measures (Hampson and McGoldrick, 2011, Bohlen et al., 2009) or the decline of the high street – see, for example, the Portas (Portas, 2011, BBC, 2011) review, is single sided. There are also many commentators on the outlook for retail globally (Deloitte, 2015) and nationally (BDO, 2015). Although this type of analysis can provide some helpful insights, it can also be singular in context (Findlay and Sparks, 2014) as illustrated by the quote below. ‘There is no shortage of commentary and suggestions over past and future policy for high streets and town centres. Much of this commentary, however, is either politicised or parochial […] The three main reports, Portas, Grimsey, and Fraser represent an interesting development of thought. Portas is very retail supply side focused and highly critical of costs in and management of high streets. Grimsey would not necessarily disagree but is much stronger on the structural changes that have brought high streets to this position and the need for digital solutions and stronger place management to be an integral part of the future. Fraser situates high streets within their broader towns and town centres and focuses on the need for ‘place’ to be overtly embraced in our thinking of high streets and town centres, and a realisation that community focused place-making is fundamental to the successful future of towns.’ ( Findlay and Sparks, 2014, p14).

The Portas (Portas, 2011) remit was narrow to begin with, focusing on high streets only. One ‘mega trend’ of digital impact (Hinshaw and Kasanoff, 2012) was absent from the recommendations and out of town shopping centre development (Fernie et al., 2010b) recommendations, another ‘Mega trend’, were ignored by the government. So any conclusions and recommendations were likely to be sub-optimal. The Grimsey review (Grimsey, 2013), although more rounded in analysis, fell short on sound recommendations. For example, the proposal of a levy on the big grocers to fund high street changes gave rise to the question of why only the grocers? What about the gaming shops, banks, other retailers, coffee shops, payday lenders, discounters etc., which all have a part to play on the high street? Both Portas (2011) and Grimsey (2013) also paid less attention to the points that landlords have a part to play, as do local authorities, on how the retail landscape is shaped from a supply side.

The fortunes of retailing and the high street are actually complex and inter-related. For example there has been a reduction in consumer spending since the onset of this recessionary period (ONS, 2012, ONS, 2014b), however this ‘flat spend trend’ (see Figure 2.1) may not be the only reason retailers and the high street are failing to thrive. Similarly only undertaking a year-on-year comparison of retailers’ like-for-like (LfL)4 trading performance (Deloitte, 2006) can result in

4 Like-for Like Sales often abbreviated to ‘LfL’ and referred to as ‘same store’ sales in the USA is a non-mandatory retail performance metric frequently disclosed by retailers and in this thesis is deemed as a non-financial metric because it is not mandatory to report this.

24 confusion about winners and losers in today’s retail market place because the analysis is likely to show ‘yoyo’ performance from one year to the next i.e. a bad prior year giving a good current year, yet this relative performance is not necessarily an indicator of a consistent retail winner.

Figure 2.1 Mega trend and shift to flat spend trend 1997-2014

1200 Household Consumption (£bn)

shift Flat Spend Trend 1000

800

600

400

200

0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Data source: Office for National Statistics.

To move towards an understanding of retail and high street success and failure it is necessary to look long (Porter, 1980, Porter, 1985, Moran and McCully, 2001, Deloitte, 2009) and deep at ‘Mega trends’ and Paradigm shifts (Naisbitt and Aburdene, 1990, Beinhocker et al., 2009, Kuhn, 1962) as well as current trends (Hampson and McGoldrick, 2013, Wrigley and Brookes, 2014). It is also important to consider what is driving any changes in shopping habits and shifting the spending patterns (Wrigley and Lambiri, 2014) on a longer-term basis, and compare this with the efficacy of current retail business models (Magretta, 2002, Enders and Jelassi, 2000, Harford, 2011, Hinshaw and Kasanoff, 2012, Handy, 2015). Put another way, using ‘snapshot’ analysis only provides a superficial understanding of the issues of retail change and potential retail success and failure.

2.3. Change and retail change models

The idea that change is inevitable and indeed a constant with permanence being illusory, can be traced back in the written word to the teachings of Buddha some 2,500 years ago (Hughes, 2015). Theories and models of change permeate all aspects of academic thought such as the

‘origin of the species’ (Darwin, 1859), as one example in the natural sciences. There are insufficient words available to entertain such a broad based discussion on change and Brown (1991, p140) provides a definitive critical review of cross referencing cycles of change to other disciplines, so this chapter fast forwards and focuses specifically on retail change models (Brown, 1987, Brown, 1991). These retail change models, for ease of reference, are categorised into two broad categories of cyclical and non-cyclical following the approach taken by McGoldrick (2002, p19) with six of the most cited simplified for description below and later assessed for efficacy against today’s retail landscape.

 Cyclical: o Wheel of retailing; o Retail Life Cycle; o Accordion.  Non-Cyclical: o Environmental; o Ecological; o Conflict.

2.3.1. Wheel of retailing

The wheel of retailing (McNair, 1931) is referenced to the work of McNair who suggested that retailing develops through a cycle based on his observations and research into US retail developments comparing the emergence of chain stores and the decline of department stores over ten years. He suggested that retailers commence as low price, low margin, simple retail operations that eventually evolve to become high price, high margin, service differentiated operations vacating the space previously occupied to allow new entrants into the market place. The strengths and weaknesses are noted in a comparative table for the six models aforementioned in Table 2.2, in the summary of this section.

2.3.2. Retail Life Cycle

The retail life cycle is an adoption of the product life cycle model (Rink and Swan, 1979, Guinée, 2002). The retail life cycle has four stages: innovation; accelerated development; maturity; and decline (Davidson et al., 1976). Innovation is the start-up phase leading to accelerated development followed by maturity and then eventual decline. Davidson et al suggested that retail life cycles are speeding up and management may be able to extend the life of the maturity stage to avert decline by taking various actions. Davidson et al (1976, p96) suggest that this is an evolutionary process and hence is deterministic in that: “executives can do very little to counteract it”.

2.3.3. Accordion

The accordion model concentrates on retailer assortment and the perpetual alteration between diversification of ranges and specialisation. First noted in relation to department stores (Hower, 1943) and then elaborated in a view of the history of retail development (Hollander, 1966) in the USA. Some empirical work in the UK on the grocery sector has revived interest in this model (Hart, 1999).

2.3.4. Environmental

Of the non-cyclical models the environmental model (Meloche et al., 1988) has probably had the most research and comment attached to it. In its simplest form, the view is that a retail company’s development is directly influenced by the surrounding environmental circumstances such as: economic; demographic; social; cultural; technological; and legal which make up the structure of the retail market place (Brown, 1987, McGoldrick, 2002). In other words, the environment establishes the conditions and retail management interpret, respond and manage within these conditions.

2.3.5. Ecological

The ecological model (Etgar, 1984) as the name implies draws on ecology and is based on an open systems perspective. Whilst this model recognises the environmental influences in the ecological sense it also incorporates the ability of individual stores in determining survival and focuses on individual decision maker’s ability to effect change (Roth and Klein, 1993).

2.3.6. Conflict

The idea of a conflict model concentrates on the competitive nature of retailing and the response from existing retailers to a new entrant. The dialectical model (Gist, 1968) contends that as the new entrant makes in roads into the market, existing retailers may imitate some of the innovations resulting in a change of form for the existing retailer or new combined formats i.e. a synthesis of retail development as illustrated by Brown (1987, p18).

2.3.7. Model variations and assessment

A summary of the strengths and weaknesses of each of the six models briefly described above are given in Table 2.2 below.

Table 2.2: Six models of retail change Model Strengths Weaknesses Wheel of Retailing  Simple vivid metaphor  Deterministic  Easily remembered  Focus on patterns not process  Aspects observable in western  Price quality continuum economies  Implicit that luxury retail is a failure (high  Trading up concept price, image etc.)  Change is inevitable  Every retailer starts as low price, low  Cycles of change margin, no frills.  Pedagogic value  Management powerless  Historic perspective  Ignores ‘existing’ competitors  Ignores environmental factors  Ignores sudden change – shocks  Cannot predict future change  Falsifying evidence hence lack of universality

Retail Life Cycle  Simple vivid metaphor  Deterministic  Based on product life cycle  Focus on stages with unclear timelines  Easily remembered  Management may influence but cannot  Aspects observable in western change the end economies  Ignores existing competitors  Staged change is inevitable  Ignores environmental factors  Cycles of change  Ignores sudden change – shocks  Cannot predict future change  Falsifying evidence hence lack of universality

Environmental  Recognition of business change as  Exclusively externally focused inevitable  Ignores management  Development in ecological sense but  Ignores competitors within ‘open system’  Lack of pattern  Change can be wide and varied.  Ignores sudden change – shocks  Apply to all geographies  More of an analytical framework than a model or theory  Borrows from contingency theory  Falsifying evidence

Ecology  Change as inevitable driven by rivalry  Akin to physical science analogies  New v Old retail  Ignores management  Mutual adaptation  Ignores competitors  Crises and response  Lack of pattern  Encapsulates shocks or catastrophe  Ignores sudden change – shocks  Apply to all geographies  Falsifying evidence

Conflict  Change as inevitable driven by rivalry  Ignores environmental factors  New v Old retail  Ignores socio-economic context in  Mutual adaptation particular  Crises and response  Focus is on response to competitors only  Encapsulates shocks or catastrophe  Lack of pattern  Apply to all geographies  Falsifying evidence

Adapted and updated from Brown (1991, Table 2, p137)

As shown in Table 2.2 there are significant weaknesses for all the models and they have been criticised as not really being theory but merely descriptive observations of historical events (Hollander, 1960). Where all the models concur is at a high level of abstraction i.e. the ‘micro’ ‘macro’ level of categorisation. Although, at a more detailed level, each can demonstrate some evidence there is plenty of falsifying evidence also.

There have been some hybrid variations and amalgamations of the models (e.g. wheel- environment-conflict combination) that convey retail complexity and these are well documented in

Brown’s (1991) paper. Although they may combine to tackle some of the weaknesses they still suffer from falsification evidence. More recent contributions include the concept of the ‘Big Middle’, providing a very complex model design (Levy et al., 2005), which suggest that the largest retailers compete in the big middle in the long run as this is where most customers can be gained. Retailers can enter at any point (and hence is a departure from the original wheel but they go on to show three interconnected wheels) i.e. low-price or innovator (possibly high-price) but this also suffers the same falsification challenge for instance the model cannot cope with exclusive designer luxury retail e.g. special edition Ferrari cars, yachts, jets, jewellery etc. and presumes that these exclusive designers at this very high premium end, chase growth and the mass market in the same way as WalMart.

2.3.8. Retail adaptation and resilience

Retail change and adaptation has been discussed for some time and change being a constant is accepted by most academics and retailers with most of the research being observational or analytical as mentioned above. This particular emphasis of adaptation and resilience has recently gained some popularity and is a combination of the environmental and ecology notions. Like the six models discussed above that borrow from other disciplines such as biology, economics, ecology, engineering, psychology, sociology and systems for the theoretical underpinning, retail adaptation and resilience does likewise. The application of resilience to the UK retail landscape and the high street in particular (Wrigley and Brookes, 2014) is explored further here and described in the quote below. ‘Are high streets likely to bounce back and resume their pre-crisis trajectories (the engineering interpretation of resilience).Or will the strength of the shock and triggering of other forces change, push the fragile ecologies of many high streets beyond a tipping point where they are unable to bounce back and must move to new configurations (the ecological interpretation). Alternatively, are high street reconfigurations best captured by an adaptive resilience interpretation which focuses on anticipatory and /or reactive reorganisation to cope with changing competitive dynamics following macroeconomic shock.’ Wrigley, (2014, p10).

The theory of ‘adaptive resilience’ as applied to ‘place’ i.e. the high street has some appeal and with sufficient analytical evidence may become grounded as a way to look at high street history and future development.

However, considering the simple principles of supply and demand enshrined in economic theory (Smith, 1776, Keynes, 1937) then there are two components that play out on the high street i.e. consumer demand and retailer supply, with the latter noted in the quote below. “the current debate on the resilience of the high streets […] has predominantly concentrated on supply side composition or outcomes in terms of vacancy rates’. Singleton and Dolega (2014, p41)

It is noted below in section 2.4.4, that consumer demand has been changing since records were first kept (Deloitte, 2009, ONS, 2014b) and academics have this implicit in most of the consumer

29 behaviour research undertaken. So the concept of adaptive resilience can be easily applied without further need for analytical evidence as the Office for National Statistics has been producing an abundance of data and statistics for many years see Figure 2.2. By inference, applying this specific ‘adaptive resilience’ (or bounce forward) concept, to the supply side, retailers (as a unit of study), hasn’t really been done and more importantly its application to the performance metrics used by retailers is absent from research literature. The primary research results discussed in this thesis begin to provide the analytical evidence that the adaptive resilience theory applied to retailer performance metrics has some merit.

Although another attempt is made to integrate multiple models using warehouse clubs (Sampson and Tiger, 1994) in the USA as an example it still falls short of universal application and was devised before the digital impact on retailing arrived, see section 2.4.6. There is no one single model or framework that can explain retail change. What does come out of the review of models including criticisms thereof is the ‘macro’ and ‘micro’ perspective in that both aspects need to be considered when taking a holistic view of retail change. The next section takes this combined ‘macro’ and ‘micro’ view and considers six interdependent multi-relational driving forces over a 25 year UK history to understand the current retail landscape.

2.4. Deconstructing the kaleidoscope

2.4.1. Introduction

In order to understand a UK retailer’s performance, there is a prerequisite to contextualise this performance within the UK retail landscape. The retail change models noted in the previous sections have been shown to be inadequate in explaining retail change so another view is considered below. To get to a historic description of UK retail change development it is necessary to ‘deconstruct the kaleidoscope’, not so much to untangle the complexity but just to enable an assessment of each facet and the driving forces in turn. There is an acknowledgement that this analysis could be done in many ways with different lenses depending not only on who is conducting the research but also its purpose. For instance, one longer-term view coming from the practitioner world (Deloitte, 2014) identifies trends and labels them thematically under four headings: big to small; discounters; technology; and structural when considering historic retail change. However this particular practitioner world analysis is supply side focused. As noted in the introduction, the six facets to analyse the driving forces considered salient are:  Regulation;  Global influence;  Socio-demographic;  Physical retail dimensions (of stores and supply chain – market place);  Digital retail dimensions (market space); and  CLV.

This framework, of external and internal (or macro and micro) factors, is not dissimilar to an integrative framework suggested by Mellahi and Wilkinson (2004, p32) for assessing organisational failure, although the specific areas for focus they noted were different to the above six facets.

2.4.2. Regulation

Some may view regulations and compliance as part and parcel of doing business and from an economic perspective may consider this as a barrier to entry for smaller retailers. But this latter point may have been relevant in the non-digital days when setting up shop meant taking up a presence in a ’place’. There are examples of, now big retailers e.g. Tesco and Marks & Spencer that started on market stalls thereby reducing barriers to entry. Today, putting up a website to trade is even more cost effective. A legal corporate entity and trading website can be established for less than £100 but getting scale, global reach and recognition may require a bit more resource. Nevertheless many academics and commentators acknowledge that regulation such as the abolition of Retail Price Maintenance was a game changer for the grocery sector in the 1960s and further changes in the 1990s on books and health products made a difference to the retail landscape (Hollander, 1960, Bromley and Thomas, 1993, Fernie et al., 2010b, Guy, 1998, Guy, 2013)

This researcher will point out that it was regulation and not mining of customer data that started the relatively recent 25 year trend by the big four grocers towards convenience stores. Prior to this there was an opposite trend, of as many years or more, of a move away from local and high street to superstores (Guy, 1998, Dawson, 2004, Fernie et al., 2010b, Guy, 2013) especially out of town shopping centres (Clarke, 2000). There is a current view that the trend towards smaller stores away from superstores and out of town hypermarkets started with the emergence of ‘convenience culture’ (Wrigley and Lambiri, 2014). The convenience culture argument contends that using customer insight techniques the leading corporate food retailers have detected changes in consumer behaviour resulting in the retailer’s pursuit of convenience stores from big out of town hypermarkets to smaller footprints such as ‘Express’ (Tesco) or ‘Local’ (Sainsbury’s) convenience stores. They argue that the Competition Commission (CC) review of the grocery sector in 2000 only precipitated the drive to make convenience store acquisitions and the consumer behaviour change started earlier. The rise of convenience culture is easy to articulate with hindsight and it is appropriate to assert that the ‘floodgates’ for making acquisitions (e.g. T&S Stores acquired by Tesco and Jacksons by Sainsbury’s) were opened by the CC’s ‘two tier’ view of the grocery market in 2000 (Burt and Sparks, 2003).

However, the real triggers to the strategic shift occurred almost a decade earlier (1990) when loyalty scheme insight was not yet available. Initiatives such as the Tesco Clubcard and other loyalty schemes did not launch until after strategic decisions on store formats had been made as

31 a result of pending legislation. Note that Clubcard was officially launched in 1995 followed by Sainsbury’s Reward Card in 1996, the latter being rebadged and now known as Nectar card.

Yet this strategic shift to open smaller formats was first evidenced by Tesco in 1993, with the opening of their first ‘Metro’ store in Covent Garden, having been in the planning stage for two years. A year later, this new approach to store portfolio development was stated in their 1994/95 ARA in the Chairman’s Statement (Tesco, 1995): ‘Our store development programme over the past year has seen 35 new stores opened, which increased retail floor space by 830,000 sq ft. These comprised 12 superstores, 15 compact stores (supermarkets with sales area less than 26,500 sq ft), 6 Metros and 2 Express stores which traded successfully in their first year. The flexibility provided by these formats has enabled us to adapt to the Government's changes in planning policy which seek to encourage new retail development in town centres. We have been successful in opening stores in areas where there was no previous Tesco representation. This has been largely as a result of careful consultation with local people’ (Tesco, 1995, p3)

Tesco then were quick to recognise the impact of the Government’s proposed changes to retail planning regulations ‘putting Town Centres first’ (Findlay and Sparks, 2014), in Planning Policy Guidance 6 (PPG6) published in 1993, that would restrict large out of town shops. They also lobbied the CC to ensure that the regulatory restrictions on retail would exclude small stores and set about expanding the smaller formats rapidly. The response to legal requirements plus the subsequent insight from ‘Clubcard’ data created a perfect storm for structural and locational changes in the grocery sector which Tesco were the first to exploit. By February 2014, Tesco had grown their smaller footprint stores (<20,000 sq ft.) to number 1,967 in the UK, from zero ten years before. This is an average rate of development of 200 per year. Similar strategies, for the same reasons as Tesco, to go ‘local’ were adopted by Sainsbury’s bringing the grocer retail brands, skills and expertise to the convenience market sector (Wrigley and Lambiri, 2014). There are now 50,747 convenience stores in the UK (ACS, 2014).

To what extent consumers’ behaviour, to shop little and often from the previous big weekly shop, has been shaped by the availability of these convenience stores; or the growth of these convenience stores has been a response to consumer shopping habits changing (Deloitte, 2009, Wrigley and Lambiri, 2014) is difficult to determine without further study and could be a topic for future research.

The regulation facet has relevance not just because it affects strategy shift as illustrated by PPG6 earlier but also impacts the speed of progress as experienced by the proposed merger of Great Universal Stores and Littlewoods due to the recommendations of the CC in 2004. There are also examples of the Office of Fair Trading (now part of the Competition and Markets Authority) levying fines on retailers for colluding on price fixing e.g. Argos, Littlewoods, Woolworths and Hasbro toys (Randall, 2003). In addition, there is now the sector review being undertaken by the Grocery Supplier Adjudicator who has applied to the government to levy multi-million pound fines.

Add to this EU directives and the UK requirements for carbon reduction and the weight of regulation becomes significant where a mishap can have a dramatic effect on the corporate reputation of the retailer be it related to ethical supply chain (e.g. Dhaka factory collapse) or financial reporting (e.g. Tesco) or the payment of ‘fair’ taxes (e.g. Amazon, Starbucks). Consequently, good corporate citizenship (Handy, 2015) becomes a core attribute for being a successful retailer and regulatory compliance a core competency which appears to be frequently underestimated.

The most recent UK government announcement about the ‘living wage’ keeps regulation as a facet for Mega trends and Paradigm shifts in retail dynamics as a salient perspective. Retailers typically employ large numbers of low paid people in their store chains frequently on minimum wage and this ‘living wage’ directive is likely to increase the cost base putting further pressure on retailer business models and margins.

2.4.3. Global influence

Another facet is the global nature of the retail landscape today. This is not just about the global influence on consumer tastes and behaviours or UK retailers going overseas (e.g. Marks and Spencer, Mothercare, Tesco and TM Lewin) or international retailers coming to the UK (e.g. Abercrombie & Fitch, Aldi, Pep and Zara), or the digital world creating one global shop window (e.g. Alibaba, Amazon, Asos, and eBay) or global supply chains or overseas wealth funds acquiring UK retailers (e.g. Harrods and House of Fraser), it is about all of the above creating complexity in a global retail village (Sparks, 1995, Moore et al., 2000, Dawson, 2007, Alexander and Doherty, 2009). The complexity of global influence manifests in many ways for retailers but has been categorised as: strategic; tactical; and operational to enable illustration with a few examples.

At a strategic level, given that the retail spend capacity in the UK is flat, see Figure 2.1, growth can come from overseas. However the decision to go international is complex requiring consideration of which countries, which retail relationship formats e.g. owned, joint ventures, franchised, licensed or by acquisition, as well as which people will go abroad to run the business (Alexander and Myers, 2000, Evans et al., 2008). The experience of UK retailers’ internationalising have been documented in case studies such as Marks and Spencer (Burt et al., 2002b).

At a tactical level, responding to competitive threats from overseas entrants or investors carries complexity in response choices. The arrival of the digital disrupters like Amazon and eBay have added further pressure to competitor response decisions. However, international entrants to the UK market have been physically present before the digital age with companies like Woolworths, Safeway, C&A, Aldi, Lidl, Zara, and since the digital age e.g. Abercrombie & Fitch, J. Crew and in

2015 Pep as examples. Responding to overseas investors can also create market place dynamic changes for instance the acquisition of ASDA by Walmart or Boots by Walgreens taking these out of the public listing limelight and the first rebuttal of a Qatari sovereign wealth fund backed vehicle Delta Two by Sainsbury’s (Peston, 2007).

At an operational level complexity can be manifold. What may appear as simple sourcing decisions may involve multiple inter-related decisions. For instance acquiring commodities such as flour to keep the in-store and regional bakery operations replenished may involve months of planning and even taking financial hedging instruments against commodity price movements. This example does not even consider the complexity of physical distribution and logistics which can be immense (Fernie et al., 2010c, Fernie and Sparks, 2014).

Globalisation is one perspective that should not be ignored when trying to get towards an understanding of the current retail landscape.

2.4.4. Socio-demographic

Much of the analysis in ‘retailing’ has to a large extent been supply-side and focused on the high street (Parker et al., 2014). Although there has been ‘marketing’ research focused on consumer behaviour and the demand side. In practical terms the distinction between the ‘retailing’ discipline and the ‘marketing’ discipline becomes irrelevant as long as the fundamental importance of consumer socio-demographic changes is properly considered. This is such a key ‘demand side’ component in any understanding (see Figures 2.2, 2.3 and 2.4) of retail or business performance, that if politicians, retail experts, consultants and academic researchers ignore this then any subsequent retail analysis is likely to be sub-optimal.

The idea of ‘adaptive resilience’ has been put forward in relation to the high street and this is discussed in more detail later in this thesis (Chapter 6), with reference to retailer performance and retail performance metrics in particular.

However, when demand side analysis is done, shown in Figure 2.2 with Figures 2.3 and 2.4 giving a more recent trend, it can be seen that consumers have been changing their habits and shifting their spending patterns since records were first kept. It can be seen (Figure 2.2.) that the amount of disposable income spent on food and drink has halved (Houthakker, 1957) in 50 years (this is known as Engel’s Law) and since the onset of the recent recession there has been a Paradigm shift whereby weekly spend is relatively flat Figure 2.3. Figure 2.4 shows the three largest spend items per week nowadays are: Housing, fuel and power; Mortgage and interest; and Transport with traditional retail at less than £195 per week.

Figures and statistics aside, academics and other commentators have acknowledged changing consumer behaviour for some time (Helgeson et al., 1984, Solomon et al., 2012), often creating temporary labels for the waxing and waning consumer trends in shopping patterns, such as ‘convenience culture’ (Wrigley and Lambiri, 2014) or ‘value shopper’ (Hampson and McGoldrick, 2013). In any analysis of changing retail context, therefore, the demand side of consumer and shopper trends should clearly be a key component.

2.4.5. Physical retail dimensions

2.4.5.1. Introduction

The physical dynamics of retailing for stores and supply chain have been grouped together as a Mega trend. Store locations, formats and sizes affect the supply chain and vice versa (Clarke et al., 1997, Fernie, 1997, Dawson, 2004, Sparks, 1994, Guy, 2013, Fernie and Sparks, 2014) with research into these areas ebbing and flowing over time. There is a wealth of research including books on both retail store location and supply chain. What is done, as noted in the introduction, is a look back over 25 years to identify the Mega trends observed.

2.4.5.2. Stores

The recent trend, in the last 25 years, towards smaller footprint and convenience stores has already been discussed under regulation and compliance above (see 2.4.2). PPG6 is seen as the real trigger for the Paradigm shift in strategy rather than ‘convenience culture’. This latter understanding arrived a couple of years later through consumer insight from loyalty schemes and reinforced the strategic push towards smaller more local formats. Regulation again through the CC review in 2000 gave the ‘green light’ for the big grocers to aggressively pursue convenience store chain acquisitions

Figure 2.2: Long-term consumer spend trends in the UK 1957-2007

Housing, fuel & power Transport Food & non-alcoholic drink Clothing & footwear Household goods & services Recreation & culture % Misc. goods & services Alcoholic drink & tobacco 40 Percentage of total expenditure on selected categories 1957 – 2006 (based on FES classification) 35 33.5

1957 1958 2005 2006 2007 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Source: Office for National Statistics data tables

Figure 2.3: Medium-term consumer spend trends in the UK 2006-2013

UK Average weekly household spend £600

Flat spend trend

£500 Mortgage & interest

Miscellaneous goods & services

Restaurants & hotels £400 Education

Recreation & culture

£300 Communication

Transport

Health £200 Household goods & services

Housing, fuel & power

£100 Clothing & footwear

Alcoholic drinks, tobacco & narcotics Food & non-alcoholic drinks £0 2006 2007 2008 2009 2010 2011 2012 2013

Source: Office for National Statistics

Figure 2.4: Share of wallet – household weekly spend categories 2013

Housing, fuel & power

Mortgage & interest

Transport

Recreation & culture

Food & non-alcoholic drinks

Restaurants & hotels

Miscellaneous goods & services

Household goods & services

Clothing & footwear

Communication

Alcoholic drinks, tobacco & narcotics

Education

Health Average Weekly Spend in 2013 £517

£0 £10 £20 £30 £40 £50 £60 £70 £80 Source: Office for National Statistics

The resurgence of the discount retail stores is a topic that frequently features in the media (Wood, 2012) and trade publications (Tugby, 2015) and has been subject to periodic academic review (Sparks, 1990, Burt and Sparks, 1994). Discount stores are defined as low cost operators that focus on price as the means for competing although see Burt and Sparks (1994, p205) for a more varied definition of grocery sector discounters. Overseas entrants to the UK market such as Aldi and Lidl created a foothold 20 years ago and with their learning over this time have started to make changes to consumer behaviour and market dynamics. Netto having sold its UK operations to ASDA in 2010 are making a comeback with a joint venture with Sainsbury’s in 2014. This discount sector has been around in regional (e.g. Fultons, Wilkinsons) and niche segments e.g. frozen foods like Bejam and Iceland for many years, including the ambient grocery sector (e.g. KwikSave, Presto, FineFare, Pricerite, Key Markets, Victor Value) and general merchandise (e.g. Woolworths, Original Factory Shop, Dunelm). There would appear to be demand for this type of retail operations from consumer segments from a variety of classes and socio-economic backgrounds. This is especially the case as any social stigma (Wood, 2012, Hobbs, 2013) of shopping in such stores is being replaced by the ‘smart’ shopper driven by recent recessionary times (Hampson and McGoldrick, 2013). Estimates of the size of the market vary but have been estimated at about 10.4% in 2014 (KeyNote, 2015). Thus far, discussion regarding discount and convenience stores is largely about retail formats changing the ‘retail offer’ (Walters and White, 1987) in a physical space and location. What needs to be assessed is the total retail supply and for now the focus is on physical store capacity (digital retail supply is discussed in a later section). We have seen the change in the dynamics of demand in Figures 2.2, 2.3 and 2.4. Yet retail supply capacity, still appears to be on the increase (BCSC, 2007).

The Paradigm shift of consumer shopping patterns is partly explained by the technology impact noted below, where the product can be digitised having the greatest impact (e.g. HMV and Blockbuster), then partly by the rise of the discounters noted above but also by consumers wanting to shop in convenient, safe shopping environments like shopping centres. As Findlay and Sparks (2014, p14) confirm: “both consumers and businesses (including retailers) perceive high streets and town centres as more difficult in many ways than off-centre or decentralised spaces […] for many consumers, town centres can be confusing, unwelcoming, and more costly alternatives’.

With on-line transactions accounting for about 12% (ONS, 2015b) of the total consumer retail spend, the largest impact on the changing high street is the attraction of shopping centres like Westfield (White City and Stratford), The Bullring (Birmingham), Metro Centre (Gateshead), The Trafford Centre (Manchester), Liverpool One (Liverpool) and Trinity (Leeds) etc. One fact that is apparent is the growth of retail capacity. Looking at the trend that retail space capacity is forecast to continue to grow, it can be seen that supply exceeds demand. The bulk of this retail capacity has come in the form of shopping centres (McGoldrick and Thompson, 1991, Lowe,

2000). Out-of-town, edge-of-town and even in-town, shopping centres have transformed the retail landscape (Fernie, 1998, Warnaby and Medway, 2004, Guy, 2013).

The point is that more than 54million (i.e. square footage from top 50 shopping centres - source: Retail Week Knowledge bank) square feet of shopping capacity (RetailWeek, 2015) has been added since shopping centres were introduced to the UK retail landscape. Leading the BCSC (2007) to comment that: ‘The amount of potential new space in the development pipeline is considerable and will be enough to provide for practically all of the projected sales growth over the next decade’.

What these centres offer is a shopping experience far superior to the traditional high street and consumers have been migrating to them. Retailers have also been drawn to these centres as they offer larger store capacity which is easier to operate and flexible logistics for product replenishment. The high street is increasingly constrained by pedestrian zones, parking charges, loading restrictions and ageing property (Findlay and Sparks, 2012). Contrast this with shopping centres which frequently include climate control; contain entertainment facilities; offer free parking; and have security attendants. Consequently, shopping centres are meeting the consumer requirement for convenience (usually in the form of easy parking), simplicity and safety whereas the traditional high street is lacking.

Furthermore, the large retail property developers such as Westfield, Hammerson, Land Securities etc. have been moving their portfolios into these prestigious shopping centre environments, leaving the secondary centres and high streets to other developers (Jenkins, 2015). What may not have been factored into such moves is the emerging realisation of some retailers, banks etc. that a smaller size of retail estate is now needed given the shifting patterns of consumer trends, particularly on the high street. Failure to realise this in all quarters is likely to lead to overcapacity further down the timeline. As Findlay and Sparks (2014, p15) explain: “Overall though, the basic fact remains: we have too much retail space and much of it is in the wrong format and in the wrong place”.

The amount of retail space also creates logistical problems for the supply chain. There has been tremendous change in the supply chain both physical and data logistics discussed below.

2.4.5.3. Supply chain

In the preface to their book (Fernie and Sparks, 2009) the advances of supply chain management are highlighted as Fernie and Sparks (2009, p xi) note: “it is increasingly hard to get over to students how much things have changed in the retail supply chain […] it is assumed that supply chains have always been at the forefront of retail innovation and have always delivered the goods. Nothing could be further from the truth. For a long time, the supply of products into retail outlets was controlled by the manufacturers and was very much a hit or miss affair.”

The late 1970s and 1980s saw a dramatic change in UK distribution and logistics with the centralisation of supply chains by the major retailers, particularly the grocers (Smith and Sparks, 2004, Fernie et al., 2010c). Since the early days of supply chain there has been a transformation of supply chain practice from ‘field to fork’ as noted below. “Retailers are the controllers of product supply in reaction to known customer demand. They control, organise and manage the supply chain from production to consumption” (Fernie, Sparks and McKinnon, 2010, p895)

The 1990s saw another transformation with quick response and efficient consumer response techniques employed throughout the supply chain. Driven by technology and data from consumer demand based replenishment systems. Again the UK grocers led the innovations in efficiency (Smith and Sparks, 2007). The most significant challenge to supply chain transformation comes from the emergent digital sales channel frequently referred to as eCommerce. Home delivery is nothing new and has been around in various forms for generations with most notably the growth of the mail order retail operations. The difference today is the customer demand for accurate, timely and cost effective delivery. The challenge of this last mile has been researched and reported (Lee and Whang, 2001, Fernie et al., 2010c). Supply chains are still adapting to eCommerce with ‘click and collect’ gaining momentum. Supply chain is one aspect of the physical dynamics that will continue to transform and remain a salient feature in any analysis of retail understanding (Fernie et al., 2010c). Digital retail as a salient feature is discussed as a separate point below.

2.4.6. Digital retail dimensions

The impact of technology has produced seismic shifts in consumer shopping habits and retailer business models representing a Paradigm shift. Digital retailing (‘dTailing’) is defined as ‘any digital interaction’ between people and organisations so it could include the consumption of free data or use of a communication channel. This is a broad definition as it would include the use of a search engine by a person with the search engine advertising products and services that may or may not be purchased by the person. The search engine is attempting to monetise its use and is ultimately selling something to someone. It therefore encompasses eCommerce, frequently referred to as e-tailing or on-line retail (Kalakota and Whinston, 1997), mCommerce, frequently referred to as mobile commerce or mobile retailing (Senn, 2000, Keen and Mackintosh, 2009), TV, phone, games, gaming and social media (Kaplan and Haenlein, 2010, Kietzmann et al., 2011). The latter aspects of retailing relate to selling via television channels, phone based sales, retailing using games and gaming channels.

All these channels are blurring at the edges, for example television programmes are now frequently watched via the internet on the move on tablets or smartphones (Ofcom, 2015), partly driven by technology convergence, for instance a mobile phone is not just for making telephone calls it has a multitude of uses such as shopping (European-Commission, 1997), and

40 also by organisations’ desires to reach more customers or subscribers, which is arguably facilitated by an online presence. The broad definition of dTailing, assumes that there is commercial intent in any digital communication, which may be saving time and or cost for a government agency. This digital communication may give the opportunity to sell to a subscriber or user of the digital platform. Therefore, dTailing seems to be the most appropriate term to capture all of this, taking the concept of uCommerce i.e. the combining of various types of electronic commerce (Watson et al., 2002, Galanxhi-Janaqi and Nah, 2004) one step further. The virtual world is the place where any non face-to-face interaction occurs in this hyperconnected world (Dutta and Bilbao-Osorio, 2012, Sargut and McGrath, 2011). This global transformation of technology enabled consumer behaviour is a fundamental ‘Paradigm’ shift that has started changing retailer business models and has yet to fully play out in the retail landscape.

The impact of digital technology and access to information (Hinshaw and Kasanoff, 2012) has had a profound influence on modern-day lives (Rotem-Mindali and Weltevreden, 2013) and attendant consumer behaviour which goes beyond internet shopping e.g. social media. Social media is now being used to sell directly to the consumer for example YouTube is a way to advertise the latest music and video releases where a clip can be watched and the album ordered. Even text messages drive consumption habits with the sharing of discount vouchers and adverts alerting consumers to sales and promotions. The most significant development from a digital technology impact is ‘big data’. Any and every aspect of a person’s life can now be captured digitally, monitored and mined. This is allowing for increasing levels of personalisation of products. Add to this the trend for machine to machine (M2M) interaction often known as the ‘Internet of Everything’, the advances in robotics and wearable technology e.g. health and fitness monitors, it becomes clear that digital technology has and continues to have a profound impact on consumer behaviour, retailers and organisations (McKinsey, 2011). Furthermore, this digital hyper-connectivity (Dutta and Bilbao-Osorio, 2012) has enabled and potentially encouraged the UK consumer to be amongst the highest global spenders ‘online’ with over 60% of the population gaining access on the move and have been termed ‘the smartphone society’ (Ofcom, 2015).

‘dTailing’ has enabled consumer ‘pull’ and the shift of power to the customer from retailers, manufacturers, and governments resulting in an informed ‘savvy’ customer. This has impacted retail change in a number of ways, of which nine key ways are shown in Figure 2.5. The consumer digital dynamics overlay diagram illustrates the complexity of the influences, and these are not exhaustive, which are likely to prove a fertile ground for continued research into consumer behaviour. There is plenty of ‘noise’ in this digital facet with many commentators, books and journals e.g. collaborative, sharing, robotics etc. Right now these are not mainstream (in the observational kaleidoscope) and hence not illustrated below. Taken individually at a micro level the individual changes are noticeable and discussed as examples in more detail below but when considered collectively the changes made by the digital hyper-connectivity

41 result in a Paradigm shift. Some of these individual changes are noted next with examples of change both from a consumer and retailer perspective and are neatly summarised in the quote below from the edited MIT book on Transforming Enterprise (Dutton et al., 2005). “deep processes unleashed by information technology continue to transform human activity. Markets, value chains, firms, transactions, business models, institutions, innovation, collaboration, standardization, trust, community – all have been and are being reshaped by information technology” (Kahin, 2005, p x)

Figure 2.5: Consumer digital dynamics overlay

Agile Complex

Impatient Ethical

Life Style Life Stage

Mission Mobile

Untrusting / disloyal

The consumer is agile in adjusting spending habits (GMA, 2007) and this can be seen in how they respond at the point of sale where vouchers for points and discounts’ are redeemed. The introduction of additional technology has enabled the production of vouchers, which can also be delivered to the mobile phone, so the retailer is no longer just relying on the loyalty cards but supplementing this and the consumer is utilising the vouchers for money off or additional points (Stahlberg and Maila, 2012).

The consumer is increasingly complex in their shopping patterns and research has been done on the rise of the purchasing power of the ‘older generation’ i.e. over 55 and that they behave a lot younger so marketing to them should be done on ‘perceived age’ (Myers and Lumbers, 2008) suggesting that retailers need to adjust their focus from the young to the ‘silver surfers’ who also have more disposable income and will use technology to research before making a purchase and that the products and services need ‘age adjusting’. There is also a complexity in the gender and age consumption of video games once seen as the preserve of teenage boys they are now consumed by females as well as men in older age ranges (Greenberg et al., 2010).

The consumer is determined by life stage, typically done by age group banding (Chen and Wellman, 2005, Moschis, 2007), which affects the way they consume technology and make consumption decisions. Even amongst the younger age groups discernible differences can be noted for brand loyalty between millennials and generation X (Pitta and Gurau, 2012). Consumer life style has been researched for many years (Plummer, 1974, Wells, 2011) and also has an impact on digital technology interaction and consumption decisions (Swinyard and Smith, 2003, Brengman et al., 2005). The Swinyard and Smith study was based on US households and researched online shoppers and non-shoppers and concluded: “On-line shoppers are younger, wealthier, better educated, have higher computer literacy, spend more time on their computer, spend more time on the internet, find on-line shopping to be easier and more entertaining, and are less fearful about financial loss resulting from on-line transactions” Swinyard and Smith (2003, p594).

Research into online shopping has continued (Soopramanien and Robertson, 2007) with further exploration of the characteristics and one of this researcher’s case studies discussed later in the thesis would suggest that the ‘wealthier and better educated’ element of the quote for USA shoppers researched in 2003 may not be relevant to the UK in 2015. Of the case studies undertaken (see Chapter 4), Retailer C targets consumer demographics classified as C, D & E and provides finance for products purchased. Retailer C has migrated from the traditional mail order catalogue operation to online only. What this means is technology impacts all lifestyle segments in consumer behaviour as well as all types of retail operations. A more recent trend is to see consumers shopping by mission (Walters and Jamil, 2003, Dennis and McCall, 2005) and the research suggests that there are gender differences with men more likely to shop by mission but also that there is a tendency to focus on events (mission shopping) such as birthdays, anniversaries, ‘back to school’ and so on. The evolutionary heritage of this research is also explored further with suggestions that in-store environments could be better arranged using technologies of sensory perception (Soars, 2009) such as smell, sound, colour etc. to change the shopping environment (Buber et al., 2007, Joye et al., 2011).

The use of the internet has been mentioned earlier under lifestyle and increasingly access and shopping is being done on the move (Funk, 2007, Lu and Su, 2009, Yang, 2010) to undertake research before making the purchasing decision (Grewal et al., 2004, Fernie et al., 2010a). The growth of price comparison websites is just one example of change affecting retailers as well as consumer behaviour.

The consumer is ethically aware (Harrison et al., 2005, Newholm and Shaw, 2007, Freestone and McGoldrick, 2008) which means that retailers need to change how they communicate their values as well as how they describe the products and services to the consumer. Examples of retailers that have embraced this into their business models are Body Shop and Lush with many of the listed retailers (e.g. M+S, Tesco, Sainsbury’s etc.) producing separate corporate responsibility reports.

The consumer is impatient with poor service (Zeithaml et al., 1991, Blodgett et al., 1995, Walsh and Godfrey, 2000) and this has been exemplified by the retailers’ response to checkout queues at the supermarkets. The large grocery retailers have changed the way they manage customer flow and invested in self scanning as well as self-service checkouts together with enhancing opening the number of checkouts in response to queue build up in-store.

Gaining consumer trust has implications for how retailers need to change to improve transparency given that consumers are sceptical of corporate marketing, willing to trust friends and family over business and governments (Sirdeshmukh et al., 2002, Trusov et al., 2009). The agile consumer will share instantly the news of the bargain buy or poor service via social media be it SMS, facebook, blog or twitter or all of them (Trusov et al., 2009, Jansen et al., 2009, Hennig-Thurau et al., 2003, Chen and Xie, 2008). The agile consumer wants transparency and good corporate citizenship from retailers and will boycott products and companies and galvanize support to protest publicly if the idea of ‘fairness’ is compromised (Maignan et al., 1999, Maignan and Ferrell, 2001, Carroll, 2003, Scherer and Palazzo, 2008). The boycotting of Starbucks for not paying enough tax in the UK is a good example of this. Although research on consumer needs in broad terms may suggest a wide and varied set of desires and expectations, in the digital space, it appears that consumers want their shopping experience to be: convenient (Wrigley and Lambiri, 2014); simple (Walters, 1977, Wolfinbarger and Gilly, 2001, Wrigley, 2014); and safe (Miyazaki and Fernandez, 2001) to do what they want when they want and from anywhere in the world. This has been enabled by the democratising digital technology. Indeed technology enables the delivery of personalisation and hence could be seen as an ideal segmentation tool.

The speed (Conner, 1993, Davis et al., 1998) of technology adoption has resulted in changed consumer behaviour that has put traditional retail business models at risk, 20 years ago forecasts would not have shown a computer company being the biggest seller of music. Apple now dominates the global music sales industry with its iTunes store. Similarly at that time, forecasts did not show the speed at which eBay, PayPal and Amazon would grow. What is known, is that retailers who have not adapted (Harford, 2011, Hinshaw and Kasanoff, 2012) to the changes in consumer demand are likely to fail, and in the UK the failure list is getting longer e.g. Virgin Megastores, Zavvi, Our Price, HMV, Borders, Waterstones, Clinton Cards, GAME etc. These retailers have had their products subject to digital change and did not move fast enough to adapt their retail business models. Although there is still some legacy retail store presence, the digital revolution has had a major impact on these business types. This Paradigm shift in consumer behaviour, initially overlooked by most retailers (with some still playing catch up with digital strategies like Morrisons, Aldi and Primark), has come from the democratisation of information through the digital and mobile technologies, as discussed above. Even though digital is seen as a seismic shift and mobile as a significant revolution, at the moment in the UK it accounts for about 12% of total retail sales (ONS, 2015b). Clearly this percentage varies from retailer to retailer depending on their strategic outlook, for example for

ASOS and Shop Direct Group it is 100%, and for Next it is about 38%, and for Harrods less than 2%.

The velocity of technological change and trying to understand the implications for retail business models is understandably confusing. Unlike in the 1990s where environmental scanning of consumer, socio-demographic and regulatory planning trends could suggest a shift was needed to move from ‘out of town’ towards town centre store development, and in spite of the power of consumer data from sources such as loyalty cards, todays retail market is difficult to read let alone predict. This has been identified as being a result of complex interaction effects between various trends and forces (technology, consumption practice, culture etc.) which has meant that: “changes which even the most sophisticated consumer facing retailers and service providers have found disruptive, are increasingly difficult to read” (Wrigley and Lambiri, 2014, p21).

2.4.7. Customer Lifetime Value

Another key component in working towards an understanding of ‘retail change’, discussed in more detail later in the thesis, is the notion of CLV. This CLV has become a key performance metric in other sectors such as telecommunications, hospitality, leisure and technology, yet has not entered the mainstream language or research agenda for the retail sector. Although some of the retailer loyalty schemes such as Clubcard (Tesco), Nectar (Sainsbury’s) and Advantage (Boots) track customers over their lifetime, an explicit use and understanding of CLV is absent. In fact Tesco are in the process of selling their customer loyalty monitoring division (Dunnhumby) suggesting its intrinsic value to the retail operations has diminished (Chesters, 2015) with the potential sale generating cash to assist with the Tesco turnaround programme. Most of the academic research has been around how to calculate CLV (Jackson, 1989, Berger and Nasr, 1998, Jain and Singh, 2002). For example Farris et al (2010, p167) use a simple discounted Cashflow of future income streams to calculate the value and define CLV as:

“The present value of future cash flows attributed to the customer relationship”

Whereas there are a number of different calculations and the primary debate has centred on the calculation method and underlying assumptions (Venkatesan and Kumar, 2004, Reinartz et al., 2005, Gupta et al., 2006, Haenlein et al., 2006, Neslin et al., 2006, Huang et al., 2012). For example different models focus on the calculation method for:  Customer acquisition, retention and expansion costs;  Customer churn using or not using real option analysis;  Arbitrary time horizon versus an infinite time horizon;  Constant margins versus variable margins;  Constant retention versus variable retention rates; and  Frequency rate for remodelling CLV.

Loyalty schemes have now been around for many years and provide useful insight into customer shopping patterns and behaviours. It is less clear if their use has transcended to the definitive metric of CLV in retail.

The airlines, hotel chains, credit and charge card businesses have been better at stratifying their customers, implying a degree of understanding the customer value to the business with the now familiar tiers of: black (American Express) or purple (Coutts), platinum, gold, silver, blue or red (Barclaycard, British Airways, Virgin Atlantic, Hilton Hotels). Retailers, particularly those without loyalty card information which is the majority of retailers, often do not know who their customers are let alone the CLV. However, when they do and mange this carefully, as will be shown in one of the case studies, then their performance demonstrates an underlying resilience. As the quote below indicates CLV is now extremely important.

“as marketing strives to become more accountable, we need metrics and models that help us assess the return on marketing investment. CLV is one such metric. The easy availability of transaction data and increasing sophistication in modelling has made CLV an increasingly important concept in both academia and practice” Gupta et al, (2006, p152)

So understanding not just customer behaviour but also CLV is critical in understanding the retail landscape.

2.4.8. Summary

In summary, today there are six salient facets observable in retail change. These are sufficiently important that they not only impact the retail landscape but also individual retailer performance and are summarised as:  Regulation;  Global influence;  Socio-demographic;  Physical retail dimensions (of space and supply chains);  Digital retail dimensions (of market space – dTailing); and  CLV. The first three (external or macro) retailers need to understand and respond to, whereas the next three (internal and micro) they need to rigorously manage.

It is worth repeating that these are interdependent and multi-relational forces making it complex to determine the impact each is having on each other and the retail landscape in general. Overlay the velocity of change affecting the retail sector and it becomes quite difficult to make predictions about success and distress without really understanding the impact of these facets. The retail winners and losers could be categorised by how well they have adapted to the changing themes discussed above. However this would to some extent be a historic reflection

46 of performance whereas the critical feature of this thesis is to work towards an understanding of current performance with a view to eventually creating better ways of understanding the reporting of retailer performance.

2.5. Conclusion

This chapter started by a review of retail models of change and discussed six concepts as well as cross referencing hybrid versions. In section 2.3.7 the models were assessed and the conclusion was that although each may have observable elements, too many exceptions to the rule render them as descriptive concepts rather than generalizable theories. As an example, the wheel of retailing would not have any place for luxury retail to emerge or exist. Furthermore, these models were conceived in the days of ‘bricks and mortar’ retail and can be considered ‘analogue’. Their relevance in today’s digital world is even less appropriate. The supporters of these models may point to mail order as a retail mechanism that was extant when the models were conceptualised and may also refer to on-line as just a digital version of this format. The digital age has created enormous disruption to what can be described as retailing as well as retailer business models. In so doing supporters of retail change models will have missed the fundamental difference in the ‘new economy’ where the aim of the digital retailer is not the immediate sales with a view to future profit and growing of tangible property assets to enhance the balance sheet i.e. traditional retailing as it was historically known. These new models are about acquiring subscribers (customers) to data mine and sell anything (e.g. any product from Amazon) and everything ( services such as Amazon Prime, Music downloads etc.) and on the way make profits from marketing activity. The subscriber adds intangible value and based on the subscriber base growth the new economy businesses can support substantial market valuations e.g. Amazon and eBay. It’s ‘retailing’ but not as we knew it.

What has been identified, referred to as deconstructing the kaleidoscope to examine these, are six salient facets of observable change that represent Mega trends and Paradigm shifts in the UK retail landscape. It has been noted that snap shot and single sided analysis cannot give a holistic view e.g. focusing on retail property vacancy rates as a metric, so looking long and deep at trends provides for better understanding. This has been done for the six salient facets of: Regulation; Global influence; Socio-demographic changes; Physical dimensions (of stores and supply chain); Digital dimensions (dTailing); and CLV. There may be different views of what the salient facets are or could be, for example, retail consolidation through the concentration of retailers or merger and acquisitions (‘M&A’). Also the innovation in retail, through technology in the physical dimensions of stores and supply chain maybe considered important. This latter aspect of technology shifts are noted from the technology enabled data management which is covered under dTailing. These other points are acknowledged as trends of potential increasing frequency but are seen by the researcher as the ‘norm’ of retailing i.e. they have been in play since the post war era and not sufficiently salient to make it into the kaleidoscope analogy.

The closest abstraction from the retail change models when applied to the observable six salient facets is the ‘external internal’ grouping of the six facets. This ‘macro micro’ grouping cannot really be claimed as a novel concept for the academic discipline of retailing as it has been borrowed from economics. So overall, the retail change models are seen as outdated rather than a help in understanding today’s retail landscape.

The point worth reinforcing in this concluding section is the interdependent and multi-relational nature of the facets and inherent driving forces making it complex to determine the impact each is having on each other and the retail landscape in general. Add to this the sheer velocity of change in a turbulent global economic environment makes understanding difficult and any predictions about the future prone to error. Nevertheless, getting some understanding of the current retail landscape is important to contextualise individual retailer performance. As mentioned in the introduction chapter understanding a retailer’s trading performance based on published information is increasingly difficult as Tesco and Woolworths illustrate.

Although not a theory or a model ‘deconstructing the kaleidoscope’ presents a new way of examining the current UK retail landscape. The next chapter moves from retail change to consider the literature on retail and business performance: measurement; metrics; and reporting.

Chapter 3: LITERATURE REVIEW

3.1. Introduction

This chapter picks up from the previous chapter, where models of retail change were discussed, with a review of literature relevant to the thesis. More specifically, literature on performance reporting, good (success) and bad (failure) performance, performance management, performance measurement and performance metrics is considered. The chapter is structured into the following sections:  Literature mapping, section 3.2, describing the academic disciplines and other sources of literature pertinent to the thesis objectives;  A critical review, in section 3.3, of accounting and finance given that financial information seems to have become the de facto standard for performance reporting;  An outline, in section 3.4, of governance and risk theory, that seemingly has promise but in practice is becoming rules bound and ineffective in putting into place performance management and reporting boundaries;  An analysis, in section 3.5, of the historical development of business failure prediction models, with an evaluation of why they are still fundamentally flawed as mechanisms for performance assessment;  Retail case study research on failure and success, in section 3.6, which reveals many insights but is only thematic and high-level and does not provide the appropriate metrics for understanding retailer trading performance;  Business and management research, in section 3.7, which is vast and varied and like retail case studies provides some insights but does not deliver the specific metrics to support the thesis objectives;  Performance measurement frameworks and models, in section 3.8, that provide some guidance yet fall short on specific metrics for the UK retailer; and  A summary that concludes in a need to directly question retail boards about what they see as important metrics, what they actually use and what they report. Each of the aforementioned is briefly introduced below.

The extant literature is reviewed to determine to what extent it helps in the understanding of retailer performance i.e. distinguishing between good and bad performance. A literature review map is created given that retail and business performance commentary comes from multiple academic, professional (including regulators) and practitioner disciplines.

In the next section a critical assessment of the current financial reporting requirements is undertaken. This involves the assessment of the numbers, including any metrics, and the accompanying narrative, which are current de facto performance reporting standards within the

ARA. Using examples including Tesco, Shop Direct and New Look the researcher demonstrates that published financial information is not enough as an indicator of retailer performance.

Governance theory and practice is then outlined. Followed by a review of risk theory, in particular the development of the company beta as a measure of company financial structure risk. This section shows that although the theories and models may have promise, their use in practice remains to be tested and furthermore, the mandated reporting rules do not deliver on the intentions of creating overarching boundaries for retail companies nor the related performance management and reporting thereof.

Given the proliferation of failure research, models and case studies, these are evaluated in a search for metrics that may help understanding of retailer performance. This section shows that the models are not as good as the model builders’ claim, in fact most are fundamentally flawed as business failure prediction models.

Similarly it is shown that neither research on learning from failure nor on business and management, which only provides high level insights into how to analyse a business or features of a business (for example, traits of good leadership), or case studies provide sufficient relevant UK metric information about how retail companies use and report their performance through non-financial metrics as key performance indicators. Theories and models of performance measurement and marketing metrics are then reviewed to assess their relevance as better ways to understand good and bad performance in retail. The general frameworks are discussed but the search for retail specific non-financial metrics proved fruitless.

In the summary of this chapter, a gap in the literature is noted that requires discovery of what UK retailers actually use as non-financial performance metrics and what they report and an assessment of the disconnect between what they do and what they say. This engagement with UK retail company boards is discussed in Chapter 4: Methodology and Methods where a multiple retailer case study strategy is outlined.

3.2. Literature review map

The literature on performance straddles many disciplines and is shown, in Figure 3.1, as a layered pictogram. The layers represent the academic and other disciplines going up from detailed specific research to more broader generalised themes as shown on the Y-axis and the bubble clusters represent the types of research. The continuum, although debateable whether or not it is a continuum, used by the researcher, assumes success and failure as two extreme aspects of business i.e. success to failure.

Figure 3.1: Literature review layer map

Consultants e.g. McKinsey, Bain / Brokers e.g. UBS, Morgan Stanley / Accountants e.g. BDO, PwC/ Lawyers e.g. DLA, Freshfields, Linklaters /Eversheds Practitioner Professional Research, trade & press e.g. Gartner/ Forrester/Kantar/Datamonitor /Mintel /Retail Week/BRC/ IGD/ FT/Bloomberg Regulatory Regulators e.g. Financial Conduct Authority, Competition and Markets Authority, Ofcom, ICAEW, Law Society Layer

www.insolvency .co.uk Companies Act; IFRS Successful:

Businesses; Leaders;Retailers Research is more section 3.7 Business: Business & broad and general Failure, & Performance Management Learning Frameworks & Success case Layer section 3.6 Metrics section 3.8 studies section 3.7 Failure case studies Retail section 3.6 Marketing Layer

Accounting & Finance section 3.3 Failure Prediction Governance Models Research is more & Risk Accounting section 3.5 & Finance detailed and specific section 3.4 Layer

Failure Success Company Voluntary Bankruptcy/ Administration/ Distress / Downturn Growth / Expansion / Profits Arrangement (CVA)/ Liquidation Receivership Chapter 11 (USA)

The academic research is grouped into bubble clusters on Figure 3.1 and cross referenced to the sections where these are discussed. The ‘success to failure’ continuum has some points marked along the axis. Certain failure points such as bankruptcy are defined by legislation and easier to annotate on the continuum whereas distress, downturn and expansion, as examples, of recognisable stages of a businesses’ development or deterioration are not so easy to mark up and are placed for illustration purposes. Furthermore the choice of the layers and the placement of the bubble clusters are for illustrating the multidisciplinary nature of the academic research in this area of interest. Finally the top layer serves to illustrate the multiple non- academic interest of commentators in this field of study. The subject matter has wide and varied perspectives, discussed in the sections that follow, and it is shown in the thesis, through the review of this literature, that none of it actually deals specifically with the research question and objectives. That is to identify: what retail performance metrics5 are used by retail boards to manage their performance; what retailers claim about their performance in the public domain; and to explore any disconnect between the two.

3.3. Financial information

3.3.1. Introduction

This section lays bare the over reliance of researchers on published financial reports which, the recent Tesco example (Chapter 1, section 1.1) proves, are subject to management misstatement be it deliberate or otherwise. Reliance on these for understanding can only give a high level and superficial post-hoc understanding of true trading performance. A critical assessment of the current financial reporting requirements is undertaken. This involves the assessment of the numbers, including any metrics, and the accompanying narrative, which are current de facto performance reporting standards within the ARA. Using examples including Tesco, Woolworths, Shop Direct and New Look the researcher demonstrates that published financial information is not enough as an indicator of retailer performance.

3.3.2. Accounting and financial information

Having suggested that there are weaknesses in accounting and financial information it seems appropriate to explain this thinking in more detail. There has been a plethora of research into performance from the Accounting and Finance discipline over the years which this researcher has grouped into three main themes noted below.

5 Retail performance metrics defined as any non-mandatory (mandatory meaning a legal or accounting rules disclosure requirement) metrics e.g. they maybe financial but are voluntary in calculation and disclosure such as LfL sales.

 The main theme for this thesis is categorised as corporate reporting (ICAEW, 1975, Cork, 1982, Smith, 1992, Hussainey et al., 2003, Beattie et al., 2004, Beattie and Thomson, 2007, Schleicher et al., 2007, ICAS, 2010, Aribi and Gao, 2011, Deloitte, 2012, Brennan and Merkl-Davies, 2013, ICAEW, 2013, ICAEW, 2014, IIRC, 2015). The underlying principles and inherent weaknesses are discussed below.  The second theme is around budgets and behavioural aspects of accounting (Argyris, 1952, Hopwood, 1974, Otley, 1977, Otley, 1978, Frow et al., 2005, Ariely, 2009b, Buttonwood, 2015). This behavioural aspect is pertinent to business failure models and is discussed in section 3.5.  The third theme has been grouped into management accounting and control (Schireson, 1957, Ferrara, 1964, AAA, 1969, AAA, 1974, Otley and Berry, 1980, Kaplan, 1984, Roberts and Scapens, 1985, Bromwich, 1986, Johnson and Kaplan, 1987, Antle and Demski, 1988, Simmonds, 1989, Scapens, 1990, Simons, 1990, Ward, 1992, Demski, 1993, Kaplan, 1994, Scapens, 1994, Simons et al., 2000, Johnson, 1992, Hopper et al., 2007, Berry et al., 2009, Scapens and Bromwich, 2010, Ward, 2012, Kaplan and Atkinson, 2015). This is broad with many inflections and is discussed in relation to performance and control frameworks in section 3.8

Accounting information is historical. The profit and loss account is a statement of the past 12 months’ performance and hence a look in the ‘rear view mirror’ and the balance sheet is a ‘snap shot’ of the assets and liabilities at the year end and is a summary of that current position. It is from such data that ratios are compiled (see section 3.3.4.) Yet, this historical published financial information seems unable to alert users of the accounts that retailers are in trouble in a timely fashion e.g. Woolworths where the accounts were signed in April 2008 and seven months later it went into administration.

The underlying concepts of accounting have some weakness (Johnson, 1943, Bowers, 1944, Ijiri, 1965, Ijiri and Jaedicke, 1966, Whittington and Tippett, 1999, Cañibano et al., 2000, Demski et al., 2002, Otley, 2002, Wyatt, 2005, Ahrens, 2008, Ashton et al., 2004, Ashton et al., 2009, Dahmash et al., 2009) which are frequently not understood or ignored and this is best explained by some simple examples. The first point to make is that the aim of accounting, with reference to the ARA, is to present the performance and position of a retailer or indeed any business or entity. Some of the roots of the principles used to calculate and present this information derive from the economic discipline but when reported only represent the legal entity not the entire activity. Consequently, some of the principles and conventions used in accounting do not reflect the economic and market reality. For example the convention of ‘prudence’, always taking the worst case scenario and adjusting the numbers to reflect and report this ‘downside’. Similarly the ‘matching’ principle, which at times is at odds with ‘prudence’, requires accruals to be made to reflect income and costs in the same time period even though cash may not have moved. The ‘valuation’ of assets both tangible and intangible has the greatest divergence and scope for

53 differing management judgements. For example if one is to consider intangible assets like brands, then what value (goodwill) should a brand have and should it appear on the balance sheet? Then there are also valuation challenges with employees and customers. They are both important but do not appear on the balance sheet yet markets accommodate this through the share price and market capitalisation values. Even valuing tangible assets can be problematic and subjective. This is particularly so for retail property portfolios. Many retailers do not value their entire store estate on an annual basis and some carry the store value at the historic purchase price in the books of account and hence the balance sheet. Take the most recent valuation of the store portfolio of Tesco where £4.725 billon has been written off (Tesco, 2015). Have the property values really dropped that much in six months since the half year accounts? The above examples are illustrations of weaknesses in accounting (Morris, 1997) and reporting and there are more detailed and technical issues as noted in the quote below. “unless an analyst is careful, it is extremely easy to draw incorrect inferences from figures reported in a company’s financial statements. Traditional textbooks do not generally examine the nature of accounting ratios: e.g. there are commonalities and interrelationships between them; there is an implicit assumption of linear proportionality; the statistical distributions tend not to be symmetric (i.e. they are not normal); […] and the means and distributions of particular ratios tend to vary between industries and even between different firms operating within the same sector” Morris (1997, p5).

The examples above serve to illustrate that rather than accept the reported financial information as a ‘true and fair’ representation of the retailer’s performance, care needs to be taken when interpreting the information presented.

3.3.3. Usefulness and predictive capabilities debate

Researchers and practitioners seem to fall into two camps when debating the usefulness and predictive capabilities of published financial information i.e. the ARA. The requirements for the ARA are defined by the Companies Act, EU regulation and International Financial Reporting Standards (IFRS) and the style and content by the regulating body i.e. the Financial Conduct Authority (FCA). Those researchers and practitioners that are in favour seem to be: Governments; reporting standard setters such as the International Accounting Standards Board (IASB); and some academics who place reliance on these reports as true representations of past performance and some use them as predictors of future performance (Beaver, 1966, Altman et al., 1977, Taffler, 1982, Agarwal and Taffler, 2008). Those more sceptical are other academics (Ijiri and Jaedicke, 1966, Horrigan, 1968, Morris, 1998, Shah, 1998, Ball, 2006), some commentators (Smith, 1992) and all the large and medium accountancy practices (Jopson, 2006, President-ICAEW, 2014) as noted in the quote below. ‘ “We all believe the current model is broken,” said Mike Rake, Chairman of KPMG “There are significant shortcomings to US GAAP [Generally Accepted Accounting Principles] and issues of concern with international financial reporting standards” he said “We are not in a very happy situation […] Corporate reporting has been largely untouched by the internet and digitisation, both of which have revolutionised the way companies operate [...] Current systems of reporting and auditing company

information will need to change – toward the public release of more non-financial information customised to the user, and accessed far more frequently than is currently done”’ Mike Rake, Chairman KPMG, (Jopson, 2006).

There is an explicit view amongst practitioners that ‘more frequent non-financial information’ may give a better view of a company’s performance, as noted in the quote above as well as more recent research from the Institute of Chartered Accountants in England & Wales and the quote from the President of ICAEW below. “We are on a journey and the journey's still happening right at the moment […] as you probably know the FRC is trying to develop some new combined guidance, on internal control and going concern, and I've been on the advisory group for that. It's likely to be a third exposure draft on that, probably in April or May [2014…] and there's still debate as to what the regime should entail.” Martyn Jones, (President- ICAEW, 2014).

So there is still plenty of discussion and debate about what is needed to measure performance and, where failure is concerned, academics appear to have been doing research and building failure prediction models based on information that the accounting profession deems ‘broken’. As such the reliability (Ijiri and Jaedicke, 1966, Dahmash et al., 2009), relevance(Francis and Schipper, 1999), timeliness (Davies and Whittred, 1980) and value (Cooper and Sherer, 1984) of the ARA has been consistently questioned, by accounting academics noted above, for many years (ICAS, 2010). Even the debate about the intellectual foundations of accounting is still raging in the 21st Century (Demski et al., 2002).

3.3.4. Ratio Analysis

When considering the use and limitations of ratios as a mechanism of financial statement analysis there is also much debate and a significant body of criticism, some of which has been noted above in section 3.3.2 (Ijiri, 1965, Ijiri and Jaedicke, 1966, Brainard and Tobin, 1968, Horrigan, 1968, Lev, 1969, Lev and Sunder, 1979, Whittington, 1980, Elgers, 1980, Barnes, 1987, Watson, 1990, Sudarsanam and Taffler, 1995, Martikainen et al., 1995, Whittington and Tippett, 1999, Ioannidis et al., 2003, Ashton et al., 2004).

Ratio analysis (a technique for financial statement analysis to make comparisons of a retailer: over time; with other retailers; and with industry averages), has its roots in balance sheet structure and credit analysis. As noted below it is dependent on the skill and experience of the research analyst. ‘From a negative viewpoint, the most striking aspect of the present state of ratio analysis is the absence of an explicit theoretical structure. Under the dominant approach of ‘pragmatic empiricism’ the user of ratios is required to rely upon the authority of an author’s experience’ Horrigan (1968, p294).

The quote above points out that the ratio analysis technique seems devoid of theoretical underpinning and is a mechanical mathematical exercise. Ratios should be used with care and with a deep understanding of their limitations (section 3.3.2) and interdependencies, see Figure

3.2 which shows the numerator and denominator relationship and the interdependencies inherent in the financial ratios i.e. where one number is composed of two other ratio numbers. Figure 3.2 Ratio inter-relationships

Consequently, any statistical model needs to be alert to the risk of ‘multicollinearity’ i.e. a situation in which two or more variables are very closely linearly related (Morris, 1997, Field,

2009). Ratios were designed to adjust for firm size and enable relative comparisons to be made: to other firms; to industry averages; across industries; across countries; and over time as already noted above, so using them as absolute measures or predictors of future events has serious limitations because of their inherent design flaws and purpose. This thesis does not assert that ratio analysis is worthless, quite the opposite. Used appropriately to compare retailers over a period of time can provide some insight and confirmation of industry trends (Evans and Mathur, 2013) from a ‘top down’ and high level perspective.

3.3.5. The current ratio and industry differences

Without understanding ratios in relation to the industry sector can also lead to misinterpretation. Take for example the Current Ratio. Standard accounting text books propose a minimum 1:1 ratio of current assets to current liabilities some recommending a 2:1 ratio as ideal (Atrill and McLaney, 2006, Dyson, 2010).

This would be fine if the firm was a stable company in stable economic times such as for example, “Widget & Co Limited” frequently used as a manufacturing company to illustrate and teach the use of ratio analysis. However, the same principles do not apply to ‘smart retailers’. To illustrate this difference, noted below in Figures 3.3 and 3.4, are examples of typical cash conversion cycles for manufacturing companies and ‘smart retailers’. Figure 3.4 clearly shows that a smart retailer uses creditors to finance the working capital and so the current ratio is likely to be less than one (<1) far from the ideal recommended of 2:1. So the point being made here is that using ratios without understanding their inherent limitations or industry specificity is likely to lead to misinterpretation of results.

3.3.6. Going beyond ratio analysis

This point, above, is further illustrated by looking at a specific example of a UK retailer such as Shop Direct Group. If a simple ratio analysis of the group was done, see Table 3.1 (directly sourced from the FAME database), this would suggest that the retailer is at risk of failure and indeed has been for many years (e.g. return on shareholders’ funds is negative and current ratio is less than 2 as suggested by standard accounting text books), yet it has not failed.

Further information going beyond the simplistic ratio analysis and ratio based failure models (see Table 3.8) reveal that the group is privately owned by the Barclay brothers. They tactically run the business as a ‘cash cow’ and strip any cash and assets from it, leaving debt, thereby applying pressure on the management team to perform and justify the use of cash. When justification meets the target investment returns criteria set, then cash is provided to progress the business proposition.

Figure 3.3 Typical cash conversion cycle manufacturing example

Stock turnover period Debtor days

Creditor days Cash conversion cycle gap funding required

Raw Payment for Sale of Cash materials raw materials finished purchased goods collected

Figure 3.4 Smart retailer cash conversion cycle example

Stock turnover Debtor period days Cash conversion cycle surplus funds generated

Creditor days

Products Sales Cash Payment for purchased collected stock

25 days 5 days 60 days

Table 3.1: Shop Direct Group ratio analysis from FAME database.

Ratios Shop Direct Shop Direct Holdings ------Group 30/04/2004 30/04/2005 30/04/2006 30/04/2007 30/04/2008 30/06/2009 30/06/2010 30/06/2011 30/06/2012 30/06/2013 GBP GBP GBP GBP GBP GBP GBP GBP GBP GBP

Profitability ratios

Return on Shareholders’ Funds (%) -0.45 -50.18 -7.70 -8.29 -51.60 -30.17 0.86 -119.88 -231.87 10.43 Return on Capital Employed (%) -0.44 -5.53 -3.78 -4.91 -23.33 -14.66 0.35 -40.73 -86.85 4.48 Return on Total Assets (%) -0.28 -3.13 -1.56 -2.39 -10.45 -4.26 0.21 -16.36 -29.33 1.46 Profit margin (%) -0.19 -2.04 -0.87 -2.15 -8.24 -5.21 0.14 -9.68 -14.33 0.77 Gross margin (%) 39.60 37.75 23.64 36.51 34.88 36.44 38.04 32.42 25.12 37.41 Berry ratio (x) 1.00 1.05 0.92 0.95 0.83 0.89 1.02 0.78 0.66 1.03 EBIT margin (%) -0.08 1.67 -2.18 -1.73 -7.16 -4.64 0.91 -9.25 -13.20 1.04 EBITDA margin (%) 0.15 4.38 6.13 1.00 -3.79 -0.88 4.95 -6.08 -9.62 4.59 Operational ratios

Net Assets Turnover (x) 2.33 2.71 4.33 2.29 2.83 2.82 2.46 4.21 6.06 5.81 Fixed Assets Turnover (x) 189.70 5.12 5.88 2.63 2.90 2.97 3.05 3.33 4.18 3.77 Interest Cover (x) -0.02 0.42 0.61 -0.30 -3.98 -3.52 1.04 -1.88 -2.71 1.23 Stock Turnover (x) 15.90 13.02 23.02 17.24 18.69 15.37 15.32 16.87 19.53 16.79 Debtors Turnover (x) 26.73 20.91 17.34 6.58 1.62 21.37 5.80 77.97 7.09

Debtor Collection (days) 13.65 17.46 21.04 55.44 225.48 17.08 62.93 4.68 51.47

Creditors Payment (days) 44.45 26.41 30.29 35.95 34.47 33.56 40.93 34.78 33.67 43.92

Table 3.1: Shop Direct Group ratio analysis from FAME database (continued).

Ratios Shop Direct Shop Direct Holdings ------Group 30/04/2004 30/04/2005 30/04/2006 30/04/2007 30/04/2008 30/06/2009 30/06/2010 30/06/2011 30/06/2012 30/06/2013

GBP GBP GBP GBP GBP GBP GBP GBP GBP GBP

Structure ratios

Current ratio (x) 2.77 1.61 1.19 1.13 1.02 1.02 1.28 0.82 0.77 0.74 Liquidity ratio (x) 2.51 1.34 1.05 1.00 0.90 0.95 1.05 0.66 0.61 0.57 Shareholders liquidity ratio (x) 37.77 0.12 0.96 1.45 0.82 0.95 0.68 0.51 0.60 0.75 Solvency ratio (Asset based) (%) 62.56 6.24 20.32 28.87 20.26 14.12 23.84 13.64 12.65 14.00 Solvency ratio (Liability based) (%) n.s. 6.65 25.50 40.59 25.41 16.44 31.31 15.80 14.48 16.28 Asset Cover (x) 2.28 6.98 5.18 4.67 7.22 3.45 6.70 14.83 14.80

Gearing (%) 13.88 969.46 194.84 110.38 194.84 414.55 181.19 296.88 281.87 183.52 Per employee ratios

Profit per employee (unit) -1,129 -1,931 -789 -1,964 -8,310 -5,650 167 -11,104 -20,958 2,499 Turnover per employee (unit) 602,922 94,589 90,362 91,506 100,837 108,540 117,469 114,679 146,283 323,884 Salaries/Turnover 3.55 19.23 23.04 24.45 23.01 21.34 19.79 21.22 18.71 7.32 Average Remuneration per employee (unit) 21,419 18,193 20,821 22,373 23,206 23,165 23,251 24,332 27,368 23,712 Shareholders Funds per employee (unit) n.s. 3,847 10,248 23,704 16,103 18,730 19,320 9,263 9,039 23,960 Working Capital per employee (unit) n.s. 16,445 5,512 5,291 910 1,962 9,342 -7,154 -10,831 -30,179 Total Assets per employee (unit) 402,412 61,687 50,431 82,105 79,483 132,650 81,032 67,883 71,460 171,118

The Shop Direct Group, therefore, is not a failing retailer. Indeed long before many other retailers, it made a strategic decision to close all the stores (Littlewoods and Index) and focus on online and catalogue retailing. They sold the properties during the ‘boom’ times unlike many retailers who are struggling with their real estate portfolios. For example Tesco has written off £4.7bn from its store real estate value in their 2015 ARA. This is a good example of going beyond a simple and standalone ratio analysis. Academic exercises that claim to predict business failure using ratios alone are likely to be erroneous and may also be a reason why many practitioners have not adopted business failure models which are discussed in section 3.5. On a similar theme, another UK retailer New Look is a good example of not needing to build statistical failure models to understand that the business has been in ‘financial distress’ (Altman and Hotchkiss, 2006) see Figure 3.5 and Figure 3.6. Looking at the balance sheet it is clear that the company has a negative ‘net worth’ of £288.9m at the 2011 balance sheet date. Figure 3.5 New Look Retail Group Consolidated Balance Sheet, 2011

NEW LOOK RETAIL GROUP LIMITED CONSOLIDATED BALANCE SHEET As at As at 26 March 27 March 2011 2010 Notes £m £m Non-current assets Property, plant and equipment 13 235.5 268.7 Intangible assets 14 741.4 729.2 Investment in joint venture 15 1.7 1.6 Available for sale finance assets 16 0.3 0.3 Other receivables 18 49.0 51.8 Income tax assets 1.9 - Deferred income tax assets 11 27.0 21.8 1,056.8 1,073.4 Current assets Inventories 17 149.5 126.3 Income tax assets - 05 Trade and other receivables 18 64.9 72.2 Derivative financial instruments 19 1.6 16.4 Cash and cash equivalents (excluding bank overdrafts) 20 191.4 206.3 407.4 421.7 Total assets 1,464.2 1,495.1 Current liabilities Trade and other payables 21 270.9 287.5 Financial liabilities 22 25.2 39.8 Derivative financial instruments 23 10.6 11.3 Provisions 28 5.7 10.5 Income tax liabilities 2.0 13.9 314.4 363.0 Non-current liabilities Trade and other payables 21 99.0 85.3 Financial liabilities 22 1,236.3 1,199.8 Derivative financial instruments 23 2.8 8.5 Provisions 28 11.3 13.4 Deferred income tax liabilities 11 89.3 94.1 1,438.7 1,401.1 Total liabilities 1,753.1 1,764.1

Net liabilities (288.9) (269.0) Deficit attributable to equity holders of New Look Retail Group Limited Share capital 31 10.4 10.4 Share premium 31 0.6 0.6 Treasury shares 31 (19.1) (14.0) Other reserves 32 8.3 25.9 Reverse acquisition reserve 32 (285.3) (285.3) Retained earnings 32 (3.8) (6.6) Total deficit (288.9) (269.0)

Reproduced from New Look Retail Group Limited ARA 2011 p50

Furthermore the Income statement shows that in 2011, it does not make enough operating profit (£98.4m) to cover the interest liability (£101.2m) from the debt burden. In this instance, statistical failure models seem unnecessary and unlikely to change the assessment that at a glance any professional or academic would come to i.e. the retailer is in financial distress. Figure 3.6 New Look Retail Group Consolidated Income Statement, 2011

NEW LOOK RETAIL GROUP LIMITED CONSOLIDATED INCOME STATEMENT For the financial periods 52 52 weeks weeks ended ended 26 27 March March 2011 2010 Notes £m £m Revenue 5,6 1,461.2 1,463.6 Cost of sales (673.6) (634.2) Gross profit 787.6 829.4 Administrative expenses (689.2) (695.4) Operating profit 7 98.4 134.0 Finance income 9 2.7 3.4 Finance expense 9 (101.2) (101.8) Share of post tax profit from joint venture 15 0.1 0.4 Profit before taxation - 36.0 Taxation 11 (2.9) (16.3) (Loss)/profit attributable to equity holders of New Look Retail Group Limited 32 (2.9) 19.7

Reproduced from New Look Retail Group Limited ARA 2011 p48

3.3.7. Summary This section clearly demonstrates that there is a tendency to look in the wrong place (historic financial records) at the wrong things (ratio analysis) to get a detailed understanding of a retailer’s true trading performance and future outcomes. This section has shown that:  professional bodies such as the ICAEW and the ‘big six’ accounting firms consider the ARA as inadequate and clarity of reporting to be work in progress;  accounting conventions and practices have inherent weaknesses;  financial ratios are interdependent and also do not have normal distribution properties; and  ratios (e.g. current ratio) should be used carefully and used in the context of industry specificity.

Overall, accounting and financial reporting in spite of their periodic updates to the rules, as a de facto standard has limitations and appears insufficient to get to an understanding of a retailer’s underlying trading performance. There is a requirement, therefore, to get to retail performance metrics used by retailers. The primary research for this thesis has concentrated on getting insight directly from retailers about what they actually do when measuring and reporting on performance see Chapters 5 & 6.

3.4. Governance, risk and company beta

3.4.1. Risk and governance

This section discusses the concepts, theories and reporting rules for governance and risks and demonstrates that as they stand they are ineffective in helping with the understanding of a retailer’s trading performance. Within the ARA there is a mandatory requirement to report on corporate governance, internal control and corporate risks. In the UK these overarching reporting of governance and risk processes have become more prescriptive through regulatory guidelines yet they still have not shed light on trading performance issues like the Tesco ‘scandal’ illustrates.

Corporate governance and corporate risk terms are frequently referred to in the same breath and sometimes, in the business environment, used interchangeably. Although the study of risk and governance could be separate areas for doctoral research, they are briefly discussed herein to set aside any view that the answer to the thesis question can be found in this body of academic work.

This section deals with corporate risk and governance which, as the quote below shows, “in the aftermath of the financial crisis, the Select Committees, thought we'd left town, and remember they had identified, where were the dogs that didn't bark. One of the criticisms was that Audit, didn't provide enough information on systemic risk” Martyn Jones, President ICAEW, 2014, is ‘top of mind’ for regulators and considers:  the theories that relate to corporate governance;  the practical development in governance and risk regulation the UK;  reporting of governance, risks and controls; and  the relationship of risk and the capital asset pricing model.

A key concept for corporate governance is the fact that a company i.e. a separate legal entity, exists and is distinct from its owners (typically shareholders) and managers (typically the board of directors). In an unincorporated business, for example a window cleaner, all three aspects could be embodied in one person and from a pragmatic day to day basis may not need to be identified. However under most Companies Acts around the world, the roles are separate and there are distinct legal duties of care (fiduciary duties) that each body has to each other.

3.4.2. Theories of corporate governance

The main theories (Clarke, 1998) relating to corporate governance (Clarke, 2004, Chambers and Weight, 2008) are noted below:  Agency theory (Arrow, 1951, Arrow, 1984, Ross, 1973, Lambert, 2009) assumes that managers will act opportunistically to further their own interests (Jensen and Meckling, 1976) before that of shareholders. Hence the need for a board of directors to ensure management behaves responsibly and implements controls to safeguard shareholder value.  Stewardship theory (Donaldson and Davis, 1991, Davis et al., 1997) assumes a degree of trust and that managers are motivated to succeed and will exercise responsibility and authority for the good of shareholders and in so doing will benefit themselves (Muth and Donaldson, 2002). This researcher asserts that this has underpinned the development of practical guidance for UK corporations, see Figure 3.7.  Stakeholder theory (Penrose, 1959, Evan and Freeman, 1993, Donaldson and Preston, 1995, Jensen, 2002, Freeman et al., 2010) takes a broader view of the purpose and impact of an entity and consequently the composition of the board and the benefits that entity should deliver and for whom. For example in Germany, employee participation on supervisory boards is a requirement reflecting a stakeholder approach.  Resource dependence theory assumes that the board will be conducive to ensuring that the organisation is connected to external agencies and environments to ensure access to resources and protect it from adverse change in the environment in which it operates (Davis and Cobb, 2010).  Managerial hegemony theory (Mace, 1979, Mace, 2009, Pettigrew and McNulty, 1995, Seal, 2010) assumes that the corporate entity as a separate body is irrelevant and that management will do whatever they want.  Political model theory (Lipsey and Lancaster, 1956, Pound, 1993, Turnbull, 1997, Gourevitch and Shinn, 2005, Roe, 2006) assumes that the allocation of corporate power and related privileges between the tripartite of company, directors and managers is determined by how government favour their various constituencies.

When we consider these theories and how they try and explain the relationship between the company, the board of directors and management it becomes clear that the theories are incredibly diverse and can be contrasted with the development of corporate governance rules discussed next.

3.4.3. Corporate governance rules and reporting

There are six theories noted above about corporate governance yet the practical development of rules in the UK, USA and other developed economies seems to have developed

independently and primarily as a response to dramatic corporate failures see Figure 3.7. In the UK, ‘tone at the top’ as referred to by the Cadbury report, i.e. the quality of UK plc boardrooms and how the directors conduct themselves and their businesses, has featured prominently in the press since the late 1980s when the UK experienced the collapse of the Bank of Credit & Commerce International (BCCI), Polly Peck and the Maxwell pensions scandal.

Figure 3.7: Corporate governance developments’ timeline.

Requirements and guidance: New / Revised Combined Code Combined Code Companies Act 2006 Cadbury Rutteman Greenbury Bribery Act 2010

Hampel Turnbull Higgs & Smith

92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15

Payments Comply for Internal or contracts th Control Explain EU 8 overseas Internal Directive Financial Internal Control Control & USA FCPA 2012 Directors’ roles Internal Risk Guidance by DoJ & responsibilities Control Management Consolidates all Consolidates USA previous all previous Sarbanes Companies Acts Guidance Oxley Act

One result of these failures was the introduction of the ‘Cadbury’ Code so called after the chairman Sir Adrian Cadbury who chaired the Committee that reported on “The Financial Aspects of Corporate Governance” (Cadbury, 1992). The setting for the report was noted in the draft for public comment in May 1992. ‘The country’s economy depends on the drive and efficiency of its companies. Thus the effectiveness with which their boards discharge their responsibilities determines Britain’s competitive position. They must have the freedom to drive their companies forward, but to exercise it within a framework of effective accountability. This is the essence of any system of good corporate governance.’ Sir Adrian Cadbury, (1992, p5).

The focus of the most recent changes to the reporting rules seem to centre on the narrative i.e. the words used in the ARA rather than the way the ‘numbers’ are presented. The Enhanced Business Review, the Transparency Obligations Directive, the Companies Act, Combined Code

65 and The Sarbanes-Oxley Act (Sox) in the USA (Sarbanes-Oxley, 2002) and Integrated Reporting (IIRC, 2015) are just some examples of this trend.

Good governance means many things to many people. Andrew Chambers (Chambers, 2002, Chambers and Weight, 2008) sums it thus: “It is invariably possible to point the finger at shortcomings in the corporate governance of an entity which has got itself into difficulties, not least because it is axiomatic in such a case that the system of corporate governance has not succeeded in delivering success. Corporate governance is now so broadly defined as to make it unlikely that deficiences in corporate governance would not be involved when entities decline or collapse – and this will appear so particularly with the wisdom of hindsight” Andrew Chambers, (2002, p1.)

Another useful view on corporate governance is given by Monks & Minow who provide the most comprehensive study of governance practice (Monks and Minow, 2003, Monks and Minow, 2012), albeit USA centric where they suggest (2003, p2) that: ‘..corporate governance is the structure that is intended to make sure that the right questions get asked and that checks and balances are in place to make sure that the answers reflect what is best for the creation of long-term, sustainable value.’

Retailers like most companies have to comply with governance and risk regulations and attest to this in their ARA, e.g. Tesco (Tesco, 2014a) . Yet the corporate failures and companies in distress, (see Chapter 2 Table 2.1), would indicate that the efficacy of governance reporting should be questioned.

Therefore good governance may be as much about ‘bottom up’ processes as well as ‘top down’ monitoring. Two good examples of ‘bottom up’ failures in retail are Ahold (De Jong et al., 2007) and Wickes (SFO, 2003), the latter resulting in the company eventually being taken over. In the case of Ahold, the overseas subsidiary company in the USA was recording ‘supplier income’ i.e. discounts and rebates and accounting for this incorrectly and hence overstating its profits’ position. In the UK case of Wickes it was a similar issue with overstating supplier income and manipulating documentation to ‘hoodwink’ the external auditors. The Tesco scandal relates to the same issue of supplier income.

All of these examples indicate a need for business processes to have integrity built into them and good risk management reporting. The majority of corporate governance statements for UK companies have become rule-bound (Abraham and Cox, 2007) and the same as each other, so it is hard for readers to distinguish who actually has good governance and who has not (see Tesco Annual Report 2014 p30 to 40).

3.4.4. Risk In its broadest context we deal with risk every day as noted in the quote below, from the book “Against the Gods” (Bernstein, 1996), and the theoretical construct can be traced to gaming theory, in other words the theory of chance i.e. probability (Kolmogorov, 1950) and financial structure risk in corporate finance theory (Modigliani and Miller, 1958).

‘The revolutionary idea that defines the boundary between modern times and the past is the mastery of risk…Risk management guides us over a vast range of decision making - allocating wealth, public health, waging war, planning a family, paying insurance premiums, wearing a seat belt, planting corn’ Bernstein (1996, p1-2)

In additional, the literature review on business failure prediction models in recent years (Balcaen and Ooghe, 2006, Calafell and Somoza, 2008) has evolved from claiming the ability to predict failure using financial ratios to classifying companies that have similar characteristics to companies that fail (Agarwal and Taffler, 2007). Taffler (Taffler, 1984) and more recently Altman (Altman and Hotchkiss, 2006) have acknowledged that what they are doing is essentially producing measures for risk of failure with their statistical models, although this is not what they had originally claimed as discussed in section 3.5.

The Turnbull report (ICAEW, 1999, Solomon et al., 2000) introduced the challenge, of measuring and reporting risk in the UK, to public companies. Although risk principles and certain reporting requirements have been in place for over ten years, the reporting of risk has only just begun appearing in UK retail companies’ ARA. In contrast risk reporting has been a feature of USA regulations for many years and companies are required to do this in their ’20-F and 10-K’ filing to the SEC (You and Zhang, 2009). Apart from the USA, where there is now a requirement to report risk in a specific way (Venkatachalam et al., 2001, Lin et al., 2010) to clarify what the market risks are for investors, the requirements in the rest of the world are less specific. Hence getting to an understanding of what is being claimed becomes challenging for most readers of these governance statements which provide little insight into performance.

3.4.5. Risk and the company beta

3.4.5.1. Capital Asset Pricing Model and calculation of beta

The calculation of the company beta, as a measure of risk, and the Capital Asset Pricing Model (CAPM) is referenced in corporate finance text books (Brealey et al., 2012). The theory of spreading the risks over a number of investments is well understood via Markowitz who has researched and published extensively on portfolio theory (Markowitz, 1952, Markowitz, 1970, Markowitz, 2012). He drew attention to the common practice of portfolio diversification and showed how an investor can reduce the standard deviation of portfolio returns by choosing shares in companies that do not move exactly together. His work has formed the foundations of most of the subsequent research on risk and returns from stocks and shares. Sharpe produced what is now known as the CAPM (Sharpe, 1964, Sharpe, 2012). The message is fairly simple that in a competitive market, the expected risk premium varies in direct proportion to beta, so all investments plot on a shares market line. A couple of drawbacks to this model are that it measures expected return yet only observed returns can be assessed.

Secondly a fully diversified portfolio will include private company shares, bonds, property, art work, fine wines etc. and most indices only include listed companies. The beta as a sole proxy for the expected returns has been further criticised. Fama & French (1993) undertook empirical studies in the USA testing returns from shares from 1963 – 1990 looking at market capitalisation, market to book value and beta and found that the market capitalisation and market to book value ratio were powerful predictors of returns and the beta made little difference (Fama and French, 1993, Fama and French, 2004).

The seminal work of Modigliani and Miller challenged the CAPM approach and contended that dividend policy and capital structure does not matter in ‘perfect capital markets’. Their premise was that the value of a business is determined by its underlying assets and not the cash flow streams relating to the financing structures. However the perfect capital markets assumption is not realistic and getting valuation for all assets can be problematic especially intangibles e.g. brand value (Veblen, 1908, Aaker, 1992, Wyatt, 2005) .

3.4.5.2. Retail company beta values

Looking at the theoretical CAPM and considering the two key components:  Market Risk; and  Company Specific Risk, and applying this to the retail sector the following results emerge, shown in Table 3.2 examples of beta values.

Table 3.2: Top 20 UK listed retailers - beta values

Company name Latest Latest beta Current market accounts Operating One month capitalisation date Revenue Last week Last avail. Year. £’000 £’000 Tesco plc 29/02/2012 64,539,000 0.47 27,019,967 J Sainsbury PLC 31/03/2012 22,294,000 1.13 6,552,777 Wm. Morrison PLC 31/01/2012 17,663,000 1.00 6,137,213 Marks and Spencer Group plc 31/03/2012 9,934,300 1.08 6,248,009 Dixons Retail PLC 30/04/2012 8,193,200 -0.14 1,049,483 Travis Perkins PLC 31/12/2011 4,779,100 0.08 2,656,629 Next PLC 31/01/2012 3,441,100 0.74 6,020,486 Debenhams PLC 31/08/2012 2,229,800 0.28 1,448,117 Burberry Group PLC 31/03/2012 1,857,200 2.03 5,449,357 Sports Direct International PLC 30/04/2012 1,835,756 1.40 2,287,931 JD Sports Fashion PLC 31/01/2012 1,059,523 0.89 345,254 HMV Group PLC 30/04/2012 873,100 11.49 7,853 Howden Joinery Group PLC 31/12/2011 853,800 0.88 1,110,046 Mothercare PLC 31/03/2012 812,700 0.32 297,848 Dunelm Group PLC 30/06/2012 603,729 0.67 1,327,995 Ocado Group PLC 30/11/2011 598,309 2.70 520,900 Asos PLC 31/03/2012 494,957 -0.13 2,162,448 Carpetright PLC 30/04/2012 471,500 1.61 453,440 Supergroup PLC 30/04/2012 313,800 2.54 459,744 Majestic Wine PLC 31/03/2012 280,304 -0.24 294,390

(Source: FAME database search results – 28/12/2012)

Considering the three retailers highlighted in Table 3.2 and comparing their relative position in the UK market place, they are similar in size using the operating revenue at between £800m to £900m. However, the beta values and market capitalisation are significantly different. Howden Joinery supplies kitchen and bathroom furniture to tradespeople who then fit these for consumers. It has a beta of less than one and a market capitalisation greater than its turnover. Mothercare retails children’s clothing and furniture and has a beta of less than one and a much smaller market capitalisation than Howden Joinery. HMV retails music, video and games on DVD and CD formats. Its beta is substantially greater than one and its market cap is substantially less than both Howden Joinery and Mothercare. If the relative strengths of the three retailers are considered, then Howden Joinery could be seen as the strongest (i.e. less risky investment) and HMV the riskiest investment. Since 2012 HMV has gone into administration although it has not become insolvent. So examining the company beta may give an indication of risk that has hitherto not been factored into the business failure models. What the beta gives, in theory, is the relationship of risk vis a vis the general market. it has been suggested by corporate finance researchers (Brealey et al., 2012, Moore, 2014) that Company beta as a measure of company risk could be a useful indicator that can signal potential distress yet it is notable by its absence from retail reports and all the failure models even though as an indicator it has been researched extensively in academia. In practice (see primary research Chapter 5) most retailers do not consider their beta to be important and do not publish it. This may in part relate to the technical assumptions that underpin it such as:  investors are risk averse;  investors have homogenous expectations;  capital markets are perfect;  investors can borrow and lend at the risk free rate; and  investors’ behave rationally.

Although these assumptions may be reasonable the main issue with the CAPM is that it is a single period model in that it is a snap shot of a point in time (Morris, 1997). Furthermore, it does not really capture all the risks relating to the company and the industry sector within which it operates. Although some organisations have attempted to get to a ‘Fundamental Beta’ to try and do this such as MSCI Inc. but their information cannot be validated and similar information is not available for the UK.

3.4.6. Summary

It could be argued that the beta is a way of noting financial structure risk and therefore may be akin to a leverage ratio such as the company gearing ratio i.e. equity to debt. This gearing ratio has been used by several failure model builders. Altman (Altman, 1993) touches on beta comparing it to his ‘Zeta’.

‘Much like the popular security return Beta measuring systematic (market-related) risk, a fundamental Beta is a relative measure of an individual firm’s risk characteristics […] we have found that the correlation (negative) between Zeta and the fundamental Beta of a company is extremely high’ Altman (1993, p221)

Apart from this reference and studies on share price movements (Morris, 1997) there is little evidence that retail company beta values have been built into failure models and tested for predictive accuracy. Similarly, true retail performance metrics e.g. LfL sales, customer numbers etc., appear to be absent from failure models developed that have been published in the public domain.

This section clearly demonstrates that there is a divergence between governance and risk theory and practical development of the rules as well as the practice of retailers when it comes to the disclosure in risk and governance statements. This section has shown that:  The theories of risk and governance do not provide sufficient detail to help with understanding actual performance;  There is a mismatch between governance and risk theory and regulation developments; and  The company beta only provides one aspect of risk i.e. financial structure risk.

Overall, risk and governance reporting, in spite of their periodic updates to the rules, appear insufficient to get to an understanding of a retailer’s underlying trading performance. There is a requirement, therefore, to identify retail marketing metrics used by retailers.

3.5. Business failure prediction models

3.5.1. Introduction

Predicting business failure has been researched for over 80 years in the USA and business failure prediction models have been around for about 40 years (Morris, 1997, Morris, 1998, Balcaen and Ooghe, 2006, Calafell and Somoza, 2008). Table 3.3 shows the key business failure prediction model research with five core types of model and the heyday of building and testing these being the 1980s and 1990s. To show the dependency on ratios as the primary variables in model construction, nine papers from Table 3.3 have been selected, on the basis that the authors were the first of that genres of model type or the models were specifically retail or UK focused and are analysed in Tables 3.4 and Appendix 2. An examination of the model types follows the Tables 3.3 & 3.4 presented below.

Table 3.3: Timeline of failure prediction models’ research (Odom and Sharda, (Charalambous et al., 2000) (Wu et al., 2010, Neural 1990) (Lin and McClean, 2001) Wu, 2010) Networks & (Lee et al., 1996) (Tang and Chi, 2005) Data Mining (Morris, 1997) (Virág and Kristóf, 2005) (Yang et al., 1999) (Wang et al., 2007)

(Zhou and Tian, 2008) (Xu and Wang, 2009) (Ohlson, 1980) (Keasey et al., 1990) (Charitou and Trigeorgis, (Zavgren, 1983) (Platt and Platt, 1991) 2001) Conditional (Zmijewski, 1984) (Sheppard, 1994) (Becchetti and Sierra, 2003) Probability (Gentry et al., 1985) (Platt et al., 1994) (Charitou et al., 2004) (Zavgren, 1985) (Lussier, 1995) (Kamath, 2006) Models (Keasey and Watson, 1987) (Morris, 1997) (He and Kamath, 2006) (Peel and Peel, 1988) (Mossman et al., (Aziz et al., 1988) 1998) (Gloubos and Grammatikos, (Bhargava et al., 1988) 1998) (Swanson and Tybout, 1988) (Altman, 1968) (Deakin, 1972) (Dambolena and Khoury, 1980) (Laitinen, 1991) (Shumway, 2001) Multiple (Edmister, 1972) (Taffler, 1982, Taffler, 1983, (Altman, 1993) (Bharath and Shumway, Discriminant (Altman et al., 1977) Taffler, 1984) (Altman et al., 1995) 2004) Analysis (Deakin, 1977) (Micha, 1984) (Morris, 1997, Morris, (Agarwal and Taffler, 2007) (Taffler and Tisshaw, 1977) (Ooghe and Verbaere, 1985) 1998) (Agarwal and Taffler, 2008) (Van Frederikslust, 1978) (Betts and Belhoul, 1987) (McGurr and (Bilderbeek, 1979) (Gloubos and Grammatikos, DeVaney, 1998a, 1988) McGurr and DeVaney, 1998b) Risk Index Models (Tamari, 1966) (Moses and Liao, 1987) (Morris, 1997)

Univariate Analysis (Beaver, 1966) (Morris, 1997) Decade: 1960s 1970s 1980s 1990s 2000s 2010s Adapted and updated from (Balcaen and Ooghe, 2006)

Table 3.4: Failure prediction models seminal papers’ analysis

Author Failure Definition Purpose Industry Trading Failure Model Sample Economic Cycle Variables Geography Type Size Time Accounting Retail Business / Other line / Finance Marketing Management (Beaver, 1966) “..inability to pay financial Provide an empirical 38 industries – mainly USA Univariate Paired sample matched by: Industry 1954 - Post WWII 30 Ratios – obligations as they mature..” verification of the predictive manufacturing Analysis and Asset size 1964 Growth Cycle with (table 1 p78) ability of Accounting Data / Public companies 79 failed minor downturns Financial Statements ‘Moody’s Industrial 79 successful Manual’ – only source (158 to 117 over five years data) available (Altman, 1968) Legally bankrupt, in receivership, Assess the quality of Ratio Manufacturing USA Multiple Paired sample matched by: Industry 1946 - Post WWII 22 Ratios or re-organising under the Analysis as an analytical tool Discriminant and Asset size 1965 Growth Cycle with Bankruptcy Act Analysis (MDA) 66 mutually exclusive groups of: minor downturns 33 bankrupt 33 viable (to 1966) (Taffler, 1982) Bankruptcy, receivership, ..identification of British Cross Industry UK 23 failed from London Stock 1968 - Consumer Ratios: voluntary liquidation and winding companies at risk of failure MDA Exchange 1973 Growth Cycle & Oil 58 P&L, B/S up 61 Solvent from London crisis downturn 44 Trend Stockbroker listing of which 48 Funds 45 considered good flow (Ohlson, 1980) Bankrupt and failed firms who Study forecasting relationships Industrial USA Conditional 105 failed from Wall St. Journal and 1970 - Inflation increasing Ratios: have filed for Chapter X and XI Logit Analysis 10-K 1976 and OPEC Crisis led 9 variables 2,058 from COMPUSTAT to global recession (not matched pairs) (McGurr and Chapter XI filing (voluntary Identify an accurate failure Cross industry to Retail USA Paired sample matched by: Retail 1989 - Longest period of Ratios used DeVaney, bankruptcy) prediction model for Retail MDA sector and Asset size 1993 growth in USA by previous 1998a) firms using mixed industry 56 failed 5 models models 56 non-failed being COMPUSTAT analysed (McGurr and Filing for bankruptcy under US Use of financial ratios to Retail USA Paired sample matched by: Retail 1989 - Longest period of 24 Ratios Number of DeVaney, Federal Bankruptcy Code (Ch. XI) discriminate between failed MDA sector and Asset size 1996 growth in USA employees 1998b) and non-failed retail firms 66 failed 66 non-failed COMPUSTAT (Bhargava et al., Bankruptcy filing under Chapter ..reconciling contradictions in Retail Canada USA Paired sample matched by: Retail 1972 - From oil crisis Ratios 1998) XI or Chapter VII protection or extant literature concerning Logit sector and Asset size 1991 recession to tech 8 variables deletion from COMPUSTAT reliability of different measures Discriminant 46 failed growth with 1980s dataset used to predict bankruptcy of Analysis 46 non- failed recession in retail firms. Population of 7,100 firms from between COMPUSTAT Manufacturing industry used as a control group for retail sector (Charitou et al., UK Insolvency Act 1986 ..examine incremental Industrial UK Logit Analysis Paired sample matched by: Industry 1988 - Early 90’s recession Ratios 2004) ‘insufficient assets to cover debts information content of and Neural and Asset size 1997 followed by growth 44 or inability to pay debts as they operating cash flows in Networks 51 failed cycle fall due’ predicting financial distress 51 successful and develop reliable failure prediction models.. (Agarwal and Financial distress, e.g. ..specifically explore whether Cross Industry UK Full population of non-financial 1979 - Early 1980’s 4 key Taffler, 2007) administration, receivership, a…..UK-based z-score model MDA firms available electronically and 2003 recession followed composite capital reconstruction, rescue driven by historic accounting fully listed on the London Stock by growth followed ratios rights issue, major disposals or data has true ex ante Exchange over 25 years: by 1990’s recession spin offs to repay creditors, predictive ability over 25 years Then growth government rescue, or since development. 232 failures acquisition as an alternative to 7,102 non- failed bankruptcy

3.5.2. The five model types

3.5.2.1. Univariate model

At the inception of prediction model building (Beaver, 1966) the key focus was stimulating debate on the empirical verification of financial information as a predictive tool. The information in published financial statements has been used as the primary variables in the form of ratio analysis to build, test and refine such models. Beaver’s (1966) was the seminal paper that re- started the debate about using accounting data and the usefulness of financial ratios. A failure prediction model was created using matched pairs of failed and non-failed firms in the USA. The conclusions were that the model could predict up to five years in advance of a company’s likelihood of failure. Beaver (1966) presents his misclassification error rate for year one as 13% and year five at 22%.

Yet businesses and retailers have continued to fail and the lack of commercial application for this type of research is discussed in this thesis on page 77. It is noted and Table 3.3 illustrates it clearly, that even though Beaver (1966) concludes that he has proven the empirical verification point through his model, subsequent criticism and avoidance of the model design would indicate otherwise.

All academic researchers, that have followed the modelling business failure prediction approach, have started from the same implicit assumption using historic published accounting information as the primary variables. As previously stated and repeated again for emphasis, by definition the profit and loss account is a statement of the past 12 months’ performance and hence a look in the ‘rear view mirror’ and the balance sheet is a snap shot of the assets and liabilities at the year end and is a summary of the current position. It is from such data that ratios can be compiled. However, if this published financial information, with all the mandatory requirements of International Financial Reporting Standards and the Companies Act (IASB, 2010, HMSO, 2009), was that good then one might expect many more ‘red flags’ about retailers in trouble than we have hitherto seen. Yet the demise of Woolworths, Comet, Focus, HMV, Phones4U and numerous other retailers in the UK had and continue to catch investors, creditors and employees by surprise. Discussion, on the appropriateness of taking historic published financial information and using it as a predictor of future performance was covered above in section 3.3. The governance and risk section 3.4 above further illustrates the inability of the ‘rules’ of statutory reporting in raising the ‘red flags’ as the quote (section 3.4.1) points out the Select Committee (President-ICAEW, 2014) had identified “the dogs that didn’t bark” i.e. the auditors and audit reports of the ARA. Implicit in this is the inadequacy of the current performance reporting regime in the public domain. Although the business failure model builders claim predictive ability, this thesis goes on to show that the models are flawed and the prediction claims are just academic exuberance.

3.5.2.2. Multiple discriminant analysis

Risk Index models are not discussed in this thesis and they have also been dismissed by most academics as ineffective and inappropriate. The lack of replication in Table 3.3 also illustrates this. To overcome the ‘univariate’ issue i.e. essentially taking one ratio at a time and analysing it in relation to the matched pairs of failed and successful firms to see if it can discriminate between the sample populations (e.g. working capital ratio is weaker in failed firms to non-failed firms), a model using Multiple Discriminant Analysis (MDA) was devised (Altman, 1968). The ‘univariate’ analysis was seen as susceptible to faulty interpretation and MDA was deemed the best solution to deal with all the statistical sampling and interpretation issues. A multivariate model attempts to explain a dependent variable in terms of several independent variables (example variables being, Sales, Profits, Total Assets, Working Capital etc.) with the discriminant analysis between failed and non-failed firms, see section 3.5.4 for a worked example. As can be seen from Table 3.3, it also prompted other researchers to develop MDA models.

The MDA paper (Altman, 1968) defined failure under the USA legal terminology of ‘bankruptcy’ and followed Beaver’s (1966) lead by using accounting ratios. Altman (1968) claims that the MDA model is better but it is still using historical published financial information. It does not consider retail companies and is manufacturing based. Altman concludes that bankruptcy can be predicted up to two years prior to the company failing and as one extrapolates further, the strength of the prediction diminishes. He contrasts this to Beaver’s (1966) claim that prediction can be made five years before failure. Although MDA became a popular method for building the models, their efficacy has been continually questioned mainly due to technical statistical model building critiques (Balcaen and Ooghe, 2006) but also for replication outside of the country population from which it originates (Agarwal and Taffler, 2007) and on its dependency of ratios as the only variables (Morris, 1997, Wu et al., 2010).

Some of the most cited papers in the Table 3.3, (Taffler, 1982, Taffler, 1983) examine failure prediction and its applicability to UK corporations (Taffler, 1984) in order to replicate the USA studies and test the business failure prediction models for efficacy and generalisability. Taffler, applies MDA to a cross section of UK firms and gets to a linear (z-score) model as opposed to a quadratic model. Taffler’s models are now in the right geography i.e.UK but not the relevant industry i.e. retail. Taffler, like Beaver (1966) and Altman (1968) is using the same variables i.e. financial ratios. He also adopted the legal ‘bankruptcy’ definition in the UK for failure. He concludes that separate models for manufacturing and retail companies need to be constructed (Taffler, 1984) after describing the outcome from his 1980 paper on distribution company failure model analysis. He also indicates that the models are not predictive but more a classification of businesses with similar attributes to failing firms. This affirms the point raised earlier that

published financial statements can only give a retrospective view of performance of individual firms and may have limited predictive potential beyond the next 12 months. Yet as shown in Table 3.4, Taffler goes on to build further models using similar variables with Agarwal (Agarwal and Taffler, 2007). This 2007 paper is UK based and is one of the most recent and again the last one reviewed in the MDA category of model building. It claims to prove that the z-score model works well with more recent data. It does start shifting the definition of failure from purely legal to encompass financial distress in a broader way.

3.5.2.3. Conditional probability models

The next step change in failure prediction model design began in the 1980s (Ohlson, 1980). This paper was concerned with forecasting accuracy and changed the basis of analysis to the econometric methodology of conditional logit models. The logit model is a type of regression model where the dependent variable measures the logarithm of the odds (i.e. the conditional probability) for a more detailed explanation see Chapter 8: Logistic Regression in “Discovering Statistics Using SPSS” (Field, 2009). The aim is to avoid all the pitfalls of statistical errors that MDA renders such as the risk of Type I and Type II misclassification errors as well as inappropriate assumptions of normal distribution curves for the variables being modelled plus the problems associated with establishing matched pair samples over time.

This was the next leap forward in method as shown by Table 3.3 and many other researchers came along to do more work of this nature. This was developed in the USA but did not cover the retail sector. The model avoided all the pitfalls of sampling methodology of MDA and appeared to produce higher levels of predictive accuracy see Table 3.6. The results indicate: “that four factors derived from financial statements which are statistically significant for purposes of assessing the probability of bankruptcy are: (i) size (SIZE)[using firm size as a factor]; (ii) the financial structure as reflected by a measure of leverage (TLTA)[using total liabilities to total assets as a factor]; (iii) some performance measure or combination of performance measures (NITA and or FUTL)[using either, net income to total assets or funds provided by operations to total liabilities, as a factor] ; (iv) some measure(s) of current liquidity (WCTA or WCTA and CLCA jointly)[using either, working capital to total assets or working capital to total assets and current liabilities to current assets, as a factor] ” Ohlson (1980, p123)

A more recent paper using logit analysis (Charitou et al., 2004) examines UK companies as shown in Table 3.4. It again only uses financial information and does not consider the retail sector. Balcaen and Ooghe (2006), as well as Morris (1997) summarise the: model types; issues with assumptions and the statistical relationships that are tested and presented as well as the fundamental flaws in these. Baclaen and Ooghe (2006, p71) go on to conclude that:

‘corporate failure prediction studies are subject to over-modelling”.

In other words there are too many models and model types that deliver no additional insight. Most of their paper is about the methods although they present a limited debate about the

quality of the key variables used in failure prediction models. They do discuss extraneous variables such as macro environment issues but incorporating macroeconomic variables in business failure prediction models is pointless because all companies are likely to be subject to these, within a given region. All businesses have to deal with this overarching economic environment.

3.5.2.4. Neural network and data mining models

The next trend of business failure prediction models has used modern computing techniques such as neural networks (Yang et al., 1999, Wu et al., 2010) and computing power to simulate interactions and undertake data mining techniques (Lin and McClean, 2001) searching for connections and correlations. Although the models are designed to draw upon any aspect of business activity and interaction, they rely heavily on financial information for the variables. The use of grouped matched pairs as a modelling technique used in univariate and MDA models is severely criticised (especially since there is an over sampling of failed firms) given the availability of information and computing power (Wu, 2010) and a focus on just the failed population using expert systems for insight is recommended by Wu.

3.5.2.5. Retail and UK

As illustrated in Table 3.4, most of the research models have been US centric, cross industry or manufacturing based. The fundamental principles of efficacy for any examination of business failure, and especially retail failure, adopted within this thesis are twofold:  first, any analysis should be industry specific; and  second, geographically specific. Research on retail failure models is scant (Bhargava et al., 1998, McGurr and DeVaney, 1998a, McGurr and DeVaney, 1998b, He and Kamath, 2006). These papers concentrate on the retail sector using similar (MDA and Logit Discriminant Analysis) techniques as shown in Table 3.4. However they are all based on companies in North America. They also use the narrow (legal) definition of failure and, like all before them, use financial ratios – note the empty column for retail marketing and business metrics in Table 3.4.

Although McGurr and Devaney (1998b) make a token gesture of using employee numbers they do not adjust for part-time workers and should arguably have used Full-Time Equivalents (FTE) in their model. However, as this thesis goes on to show, replicating these retail studies for the UK is unlikely to provide insight towards an understanding of retail failure due to the fundamental nature of ratio analysis and weaknesses in historic published financial information, already discussed above in section 3.3.

As noted above, McGurr and Devaney (1998a, 1998b) try to move from a cross industry model to a retail specific model using US data. Here these authors recognise the importance of the retail industry to the economy and the vulnerability to retail business failure. They cite resultant retail failure impact on unemployment as a major factor for the US economy. This is contrary to the view of Taffler who suggests that the UK retail sector is robust (Taffler, 1984). McGurr and DeVaney (1998a) seem to prove that mixed industry models lose their classification accuracy when applied to the retail sector. They examine the retail sector for the right reasons but in the wrong way since, like all other modellers, they use historic published financial information see Table 3.4 and Appendix 2. In addition their assessment of the current ratio maybe correct for simple ratio analysis of manufacturing firms but wrong when considering retailers, already discussed in section 3.3 with the illustration of the cash conversion cycle see Figures 3.3 and 3.4.

In fact there seems to be a proliferation (82 different types, Appendix 2) of financial ratios, some that just don’t make sense. It may be the accessibility of information and the computing power available nowadays that has encouraged researchers to engineer elaborate models with questionable variables. Taffler (1984) also discusses the commercial adoption of the various failure prediction models and notes that: ‘in the UK there has been to date less up take among commercial bankers and also professional accountants. This latter mirrors Altman’s US experience’ Taffler, (1984, p 223).

In short, he reveals that whilst the academic literature is replete with models they are not being applied in the real world. This may well be due to a perceived lack of efficacy in prediction models and/or a concern about the variables they draw on. In his book Morris critiques failure prediction models of all types and concludes that they are less useful than monitoring share price movements as indicators of corporate failure (Morris, 1997). He also considers accounting ratios and their weakness as the core components of failure prediction models. As noted above variables (such as working capital, sales, net income) used within business failure prediction models often seem divorced from those recognised by practitioners working in the retail sector. For example, poor performance in the following metrics would suggest the retailer is experiencing a ‘downturn’ in trading activity and is frequently a cause for concern about performance amongst retail boards, see Chapter 5 of this thesis.  LfL sales;  Customer footfall or the number of transactions processed;  Average basket spend i.e. the amount of spend per transaction; and  Sales ft2 (sales per square foot). So getting the right variables or metrics may be more important than the size and type of the model. However, the challenge for any researcher is that these retail performance metrics, noted in the bullets above, are not available in the public domain in the same way as historic financial accounting reports. So creating better and relevant models is not possible from secondary research sources alone as this meaningful information is not a mandatory reporting

requirement and therefore not reported by retail companies as Chapters 5 and 6 show. The literature on business failure prediction models discussed above, although primarily USA centric, has over time covered other geographic regions. As noted above some have considered UK companies (Taffler, 1984, Charitou et al., 2004, Agarwal and Taffler, 2007) and retail companies (Bhargava et al., 1998, McGurr and DeVaney, 1998b, He and Kamath, 2006).

However there is a gap with UK and retail combined. There is no published failure prediction models for UK retail that have a high classification accuracy. Most of the UK retail research appears to be exploratory and discursive as noted below in section 3.6. Also as is apparent from Tables 3.4 and Appendix 2, no one has incorporated retail industry relevant metrics into the failure prediction models although it has been done for other industries such as oil and gas (Platt and Platt, 1991, Platt et al., 1994).

One aspect of this thesis was to evaluate whether or not industry specific (Porter, 1979, Platt and Platt, 1991) i.e. the retail sector, geography specific (Porter and Solvell, 1999) i.e. the UK, and non-financial (i.e. retail marketing) metrics (Platt et al., 1994) combined would give a better view of retail company distress than just financial ratios alone. The challenge is to gather the metrics and apply them to a new style of risk distress model. This style of model has been attempted in general terms through expert systems and neural network techniques but they all suffer from the difficulty of obtaining relevant metrics and tend towards the inherent model weakness of using historic financial information. Due to the lack of availability of the required metrics, a UK retail distress model is not central to the thesis. However, should information start to become readily available this could form a good area for further study.

3.5.3. Results of failure model studies

Summarising the predictive claims of some key business failure models it can be seen that although much is claimed the relevance and use to management, employees, government, shareholders etc. is somewhat limited. It can be seen from Table 3.5, that the different business failure prediction models, have a wide and varied classification accuracy. The two retail examples (McGurr and DeVaney, 1998b, He and Kamath, 2006) show classification rates less than the manufacturing and cross industry models. Table 3.5: Predictive accuracy claims of 11 model types summarised Years prior to failure Model Year one Year two Year three Year four Year five (Beaver, 1966) 87% 79% 77% 76% 78% (Altman, 1968) 95% 72% 48% 29% 36% (Deakin, 1972) 87% 90% 82% (Altman et al., 1977) 96% 85% 75% 68% 70% (Ohlson, 1980) 96% 95% 92% (Taffler, 1982) 96% 61% 48% 35% (McGurr and DeVaney, 1998b) 79% (He and Kamath, 2006) 80% 65% 70% (Agarwal and Taffler, 2007) 96% (Wu et al., 2010) 89% (Wu, 2010) 98%

The one study that appears to stand out is that of Ohlson, whose methodology is dismissed by Taffler (1982, p342) as noted below: ‘A conditional logit approach is used by Ohlson, 1980, although his results were disappointing.’

Taffler (1984) also determines that his linear regression method is better than Altman’s quadratic approach whereas Altman (Altman, 1993) in part two of his book, chapters 8 to 12, criticises every other methodology and promotes his Zeta model as the best method. McGurr and Devaney on the other hand consider Deakin and Ohlson’s models to perform the best (McGurr and DeVaney, 1998a) although the results for retail companies are not as good when compared to the manufacturing or cross industry model studies and McGurr and Devaney (1998a) cite three possible reasons as: bias due to the distribution of the cross industry firms; retail firm ratios being different; and, the population used in the replicated study being from a different population to the original studies. Wu (2010) does an analysis of companies in Taiwan and Wu et al do an analysis of various models using USA data and both determine that neural networks including other information, i.e. not solely reliant on financial ratios as variables, produce better models. Although the linear z-score model is considered resilient and the best method for assessing UK listed companies by Agarwal and Taffler (2007), the limitations of all models is articulated by them whereby they acknowledge that some academics may be putting the wrong interpretation on the purpose and use of these models (Agarwal and Taffler, 2007). ‘The technique quantifies the degree of corporate risk in an independent, unbiased and objective manner […] Z-scores are not explanatory theories of failure (or success) but pattern recognition devices […] so it is only misunderstanding of purpose that elevates the z-score from a simple role as a measurement device of financial risk to the lofty heights of full blown theory of corporate financial distress […] Separate models need to be developed for analysing the financial health of unlisted firms […] A final point relates to the continued misunderstanding of the specific nature of z-score models which can only be appropriately applied to the population of firms from which they were developed. As such, it is totally wrong and potentially dangerous to seek to apply the very accessible Altman (1968) US model in market environments such as the UK”. Agarwal and Taffler (2007, p298/9)

The last sentence in the quote above is somewhat odd given that Altman has claimed the successful application of his model in other countries (Altman and Hotchkiss, 2006), in spite of its inherent limitations, as discussed earlier.

It can be expected that the debate will continue with academics producing bigger and different models to deliver the desired outcome to justify the methodology being pursued – ‘the Pygmalion affect’. It seems that the main focus of the model builders is on the modelling techniques rather than the relevance and efficacy of the variables. Consequently the focus of debate should shift from the ‘measurement device’ to both the variables being used and the decision context.

There has been some shift in thinking with the production of business failure prediction models that incorporate ‘other market data’ (Marais et al., 1984, Zavgren and Friedman, 1988, Peel and Peel, 1988, Zavgren et al., 1988, Shumway, 2001, Platt et al., 1994, Wu et al., 2010, Wu, 2010).

However, they by and large, do not take into account how the information may get used i.e. the decision context (Libby, 1975, Keasey and Watson, 1991). It is generally understood that ‘what gets measured gets done’ (Peters and Waterman, 1982) so a focus on the variables and for what they may be used has been a sensible next step in the research and this has also been recommended by some academics. ‘There seems little doubt that additional studies utilizing existing methodologies are likely to produce increasingly poor returns in terms of improving either our understanding of firm failure or in achieving significantly superior predictive accuracy […] If the literature on distress/failure prediction is to progress further, then more explicit and formal modelling of the economic interests and decision processes of the firm’s major stakeholders will probably have to be undertaken.’ Keasey and Watson (1991, p100).

The above quote tries to move the debate but still discusses the model building whereas the real issue is the variables used. Model building as a process has to make assumptions. The major criticism of the assumptions is that frequently they do not reflect reality and behavioural aspects so rarely deliver reliable outcomes. There is an increasing body of criticism about economic and finance models (Buttonwood, 2015, Thaler, 2015) and most of the same criticisms are equally valid for business failure prediction models. As the quote above indicates, decision processes are important and decisions are made by people (e.g. board members, shareholders, investors etc.) yet the models to date do not factor this into the model design. People tend to be biased and are frequently irrational (Ariely, 2009b, Ariely, 2009a) in their decision making and therein lies the conundrum of predictive model design. Not factoring in this vital element in model design is likely to deliver weak and ineffective models. So the claims by many of the academics who have suggested that business failure can be predicted by their models due to the statistical significance scores, in reality is seen by this researcher as ‘academic exuberance’.

3.5.4. Testing UK retail data

In the discussion above it has been shown that the business failure models are fundamentally flawed and to close out the debate and prove the point, some have been applied to a sample of UK retail companies. The models selected are based on the ability to follow the model in the published information. For example, Altman’s 1968 model is replicable whereas Taffler’s 1984 model is not and neither is Ohlson’s 1980 model (although Zavgren (1985) uses the Ohlson method and this is replicable). The failure models, shown in Table 3.6 below, present conflicting results and cannot be relied upon for meaningful understanding given the criticism of the statistical methodology discussed earlier.

The models noted have been applied to historic financial published information of an iconic UK retailer with a long and successful history, Harrods. These calculations have applied the coefficients in the original research papers to Harrods data whereas the technically correct statistical method would be to calculate the coefficients from the Harrods population sample.

However as this is a unique single company analysis and matched pairs sampling has not been used this cannot be done, hence the use of previously published coefficients to facilitate model replication.

Table 3.6: Eight business failure and risk models applied to Harrods

AUTHOR 2009 2010 2011 2012 2013 Beaver (1966) RoA -0.25 0.03 0.05 -0.01 -0.003 Beaver (1968) CA/CL 0.43 0.42 0.45 0.40 0.45 Altman (1968) Zeta = -1.42 -0.35 -0.09 -0.08 -0.04 Altman (1993) Zeta = -0.96 0.03 0.20 0.25 0.28 Deakin (1977) I = -4.39 -0.49 -0.22 -1.06 -0.96 Taffler (1984) z = -17.29 -16.84 -14.82 -16.11 -15.41 McGurr & DeVaney (1998b) Z = -59.79 220.74 209.42 240.35 217.89 Zavgren (1985) P = 32% 33% 27% 29% 28%

All the results show, apart from McGurr & DeVaney 2010-2013 highlighted, Harrods is doomed to failure even the Zavgren model suggests a 27% to 33% likelihood of failure in all the years. However, reality is a long way from these ‘predictive’ models’ claims of generalizability.

A similar set of unreliable results is noted in Table 3.7, where the highlight indicates safe and the remainder indicate failure, when the eight models are applied to the financial ratios of Shop Direct Group (SDG). Table 3.7: Eight business failure and risk models applied to Shop Direct Group

AUTHOR 2009 2010 2011 2012 2013 Beaver (1966) RoA -0.04 0.001 -0.021 -0.029 0.013 Beaver (1968) CA/CL 1.02 1.28 0.82 0.77 0.74 Altman (1968) Zeta = 1.71 3.06 4.02 3.46 3.09 Altman (1993) Zeta = 0.82 1.92 0.97 1.09 3.02 Deakin (1977) I = -1.31 -0.28 -3.09 -4.36 -0.13 Taffler (1984) z = -3.46 0.75 -6.63 -8.93 -3.86 McGurr & DeVaney (1998b) Z = 266.8 567.6 -98.5 114.7 150.2 Zavgren (1985) P = 33% 59% 33% 30% 34%

In taking the testing one step further, it has been suggested by more recent research (Wu, 2010) that the entire distressed population should be used. Hence, a population of distressed UK retail firms has been determined. This has been done by taking the Mintel UK Retail Rankings for 2005 and comparing it to the same ranking for 2010. Where retailers are no longer in existence in these rankings in 2010 e.g. Safeway, mfi, then these form the population together with known retailers that have been reported as failing in the financial press or been taken over due to distress e.g. HMV, GAME. The list of 194 retail companies is presented in Appendix 1. Given that this is a list of ‘failed’ retailers the logic would be to expect a prediction classification accuracy of at least 95% using any of the models. Taking the last available year, from the FAME database, of each of these distressed retailers (this is where the greatest accuracy claim is made by all the model builders) and applying Altman’s 1968 model, with the Z-score formula as noted below:

Z = 1.2X1 + 1.4X2 + 3.3X3 + 0.6X4 + 1.0X5

Where:

X1 = Working capital / total assets;

X2 = Retained earnings / total assets;

X3 = Profit before interest and tax / total assets;

X4 = Market capitalisation / book value of debts;

X5 = Sales/ total assets.

The results gave a very poor level of classification accuracy at 37% as shown in the Table 3.8 below. A Z-score of 3 or more should be classified as ‘safe‘, whereas scores of 1.8 or less should be classified as ‘failure candidates’ p14 (Argenti, 1983).

Table 3.8: Z-score model applied to the 194 UK retail ‘failure’ population

Classification Number % accuracy Failure (<1.8) 72 37 In between 37 19 Safe (>3) 85 44 Total 194 100

To further illustrate the results, the top ten ranked (from the failure population sample) retailers in 2005 are given in Table 3.9 below with their classification scores. Only one out of the top ten (Somerfield) is classified accurately. It is acknowledged that the Altman 1968 model was designed for the USA manufacturing sector so the illustration above can be criticised for not being correctly adapted for: retail and for the UK and for not using matched pairs methodology.

Table 3.9 Top ten retailers from the 194 ‘failure’ population classification score

Retailer Classification Retailer Classification score score Safeway Stores 3.8 Comet 3.6 Somerfield Stores 0.8 TJX 4.5 Woolworths 2.9 Superdrug 2.7 John Menzies 4.2 Wickes 2.8 Game Group 4.4 TJ Morris 6.3

Note: 3+ = safe; 1.8 or less = failure

Nevertheless, any finessing (or as noted in practitioner world ‘fudging’) of variables, methods etc. is unlikely to deliver a 95% accuracy. Any finessing will just become retro fitting the model to get the desired outcome.

The purpose of this section is not to present dozens of results of models applied to UK retailers as this is not the central tenant of this thesis. The purpose was to give some illustration to prove the models are fundamentally flawed when applied to UK retailers and this has been done.

3.5.5. Summary

To summarise this section, a number of key points have emerged:  statistical failure prediction models using only financial ratios are essentially risk indicators whose predictive accuracy is questionable;  the trend towards using other market data appears to improve the classification accuracy of the failure models yet they are still mainly reliant on historic published financial information;  UK retail failure model research has not been published and exploratory tests have proved the models are inappropriate;  although market data has been used, retail marketing metrics have not been used to enhance the predictive accuracy of models;  further work is suggested with prediction models and the decision context;  none of the models to date have explicitly used a measure of market and company risk based on corporate finance theory i.e. the Company beta; and  none of the models have accounted for the behavioural aspect of company actions given that retailers are run by people.

3.6. Failure theories and retail research

3.6.1. Introduction

An examination of the literature shows that whilst there is considerable work on broad business success and failure (see section 3.7), the specific nature of retail failure tends to be under examined, with a few exceptions (see section 3.6.3). Retail case study research on failure and success reveals many insights but is only thematic and high-level and does not provide the appropriate metrics for understanding retailer trading performance. Similarly it is shown that neither research on learning from failure nor on business and management, which only provides high level insights into how to analyse a business or features of a business (for example, traits of good leadership), or case studies provide sufficient relevant UK metric information about how retail companies use and report their performance through non-financial metrics as key performance indicators.

3.6.2. General theories implying failure

Retail change has already been discussed in Chapter 2. In this section the focus is on failure. There are a number of theories that implicitly and explicitly deal with the life cycle (i.e. change)

of a business and / or a retailer and from these it can be deduced that at some point a business will come to an end. Retail success literature is discussed in section 3.7.

General theories of failure could be deemed as by-products of other general theories such as: free market competition (Smith, 1776); employment (Keynes, 1937, Keynes and Moggridge, 1971) ; business cycles (Schumpeter, 1980); evolutionary economics (Boulding, 1991, Hodgson, 1998); organizational decline (Whetten, 1980, Cameron et al., 1988, Sull, 2005); and retail change already discussed in Chapter 2. These all suggest that failure is a normal part of the business life cycle including the effect of economic and market forces, although Argenti (1976) did identify corporate defects such as bad management, creative accounting and unresponsiveness to change and then sequenced ‘defects’, ‘mistakes’ and ‘symptoms’ (Argenti, 1976). As referenced above, the general theories acknowledge the notion of business failure but then say very little about it. Retailers that demonstrate ‘adaptive resilience’ and how this manifests in their performance management and metrics allowing them to avoid distress and thereby failure becomes an interesting area for study and is discussed under section 3.6.6.

3.6.3. Business and retail failure research

The academic ‘failure’ literature can be grouped into three broad categories:  failure prediction models (Beaver, 1966, Altman, 1968, Ohlson, 1980, Taffler, 1984, Agarwal and Taffler, 2007, Wu et al., 2010) already discussed above in section 3.5;  Case study research (Burt et al., 2002a, Pal et al., 2006, Pal et al., 2011); and  understanding and learning from failure (Arino and De La Torre, 1998, Sull, 1999, McGrath, 1999, Shepherd, 2003, Edmondson, 2011). Most of the early research into business failure has its roots in the credit scoring of companies and is focused on bankruptcy and was essentially USA based (McLaughlin, 1927, Warren, 1935). This research history, relating to failure model development has been discussed above in section 3.5. The other two categories are primarily case study based and are mentioned in the sections that follow but it is fair to say at this point that none of this specific company based case study research gave retail boardroom metrics that could have been used to assess the research objectives.

3.6.4. Success and failure continuum

The literature mapping of success and failure research, see Figure 3.1, shows considerable publications in accounting and finance from a regulatory perspective (HMSO, 2009), defining the rules for corporate reporting (IASB, 2010), bankruptcy and liquidation (Cork, 1982).

Success and failure, for this thesis, have been placed on a continuum and known stages of failure have been annotated on this continuum in Figure 3.1. Moving from success at one end to

failure, a retailer can be deemed to go through a number of stages the first being a ‘down turn in performance’ followed by ‘distress’ accompanied frequently by ‘financial distress’.

The term ‘failure’ means many things to many people and has been used liberally not just by the media but also academics (see the definition discussion in section 3.6.6.) For instance there is much media coverage about the failure of M+S and Tesco yet they have not failed (in the bankruptcy sense) nor are they in ‘financial distress’, although some commentators may argue otherwise, but it could be argued that they have been suffering a ‘down turn’ in performance, widely reported in 2014 (Gutherie, 2014), which in early 2015 (Rankin, 2015) is reported as turning around. There is only one clear and unambiguous narrow definition of ‘failure’ and that is when a retailer has gone bankrupt. When this occurs it is too late e.g. Woolworths and Phones4U, such that an early warning broader definition of failure, ‘financial distress’ and distress may be helpful together with the ability to predict retail failure.

3.6.5. Underlying theory in failure research

Examining the underlying theory inherent in the failure research two broad categories emerge:  Normative (deductive); and  Positive (inductive). The normative theories are more general conceptualisations and typically deductive whereas the positive theories are based on empirical evidence, hence inductive and aim to answer the question why businesses fail.

Failure models at best have a tenuous link to theory via models such as shareholder option pricing, gamblers ruin as a measure of risk, cash flow, market value, CAPM as a measure of risk and agency models. Whereas, case studies do a better job than failure prediction models at linking to theory such as ‘notion of blame’ (Pal et al., 2011) but case studies may then be difficult to generalise.

There has been a wave of research on failure particularly with start-up ventures and small business (Drucker, 1998, Shepherd, 2004, Landier, 2006) and this focuses on learning from failure and considers failure to be the norm with success as the exception (Ghosh, 2011). This is understandable when considering start-up small businesses where there is a high failure rate. In 2011, 261,000 firms employing fewer than 100 people were formed in the UK and 229,000 died (ONS, 2014a). From his venture capital fund research, Ghosh (2011) redefines failure to include firms where the investment funds for the start-up are not returned i.e. the cash return on investment (CROI) is less than one (CROI < 1).

3.6.6. Defining retail failure

Most of the academic literature, particularly failure prediction models, take a simple approach to the definition of failure i.e. bankruptcy. This is understandable as it is a finite and easy to recognise event. Overall, the literature shows definitions ranging from legal (i.e. the Bankruptcy Code in the USA Chapters 9 to 15 and in the UK, the Insolvency Act 1986) being relevant for the various terms such as: liquidation; receivership; company voluntary arrangement (CVA); and reorganization, to deterioration in business performance.

The legal definitions being favoured by the accounting and finance based research (Altman, 1968, Ohlson, 1980, Taffler, 1982, Morris, 1997, McGurr and DeVaney, 1998b, Charitou et al., 2004, Agarwal and Taffler, 2007) whereas the broader definitions are used by the retail and marketing researchers (Burt et al., 2002a, Mellahi and Wilkinson, 2004, Pal et al., 2006) as well as business and management researchers (McGrath, 1999, Sull, 2005, Harford, 2011, Edmondson, 2011).

There seems to be no real consensus or precise definition of ‘retail business failure’. In their review of international expansion strategies and the case of Marks and Spencer, Burt et al appear to use the term failure appropriately given that the retailer has been divesting a significant part of its business and an element of strategy would appear to have failed (Burt et al., 2002b) at the time. More recently, 2014 and 2015, M+S is back to international expansion so it is clear that M+S, the company, has not failed. A view could be taken that the international strategy has not failed either, merely been adapted.

Consequently, a distinction needs to be made between the retailer (corporate entity) failing e.g. Woolworths and an element of strategy or operations failing e.g. M+S international expansion. Nevertheless, there is still an element of opaqueness in determining when a retailer reaches or goes beyond a tipping point from which it cannot recover. There is a lack of consistency in the use of the term ‘failure’ by academics, analysts and commentators with a tendency to sensationalise their reporting by referring to ‘failure’. The legal definition of bankruptcy is clear cut as the entity has failed and there is legal evidence for this and can be termed a ‘narrow’ definition. The stage that typically precedes bankruptcy requires a ‘broader definition’ and this is discussed below. Prior to this is a stage of ‘financial distress’ which can be terminal or temporary (Altman and Hotchkiss, 2006). What precedes ‘financial distress’ and sometimes runs in parallel is ‘organisational decline’(Cameron et al., 1988) or as referred to in this thesis as ‘distress’. Organisational decline has been defined as: ‘a two stage phenomenon in which, first, an organisation’s adaptation to its domain, or microniche deteriorates, and second, resources are reduced within the organisation” (Cameron et al, 1988, p6).

In contrast, Sheppard and Chowdhury (2005) determine that failure is about the failure of strategy and in particular the response of management once decline has set in (Sheppard and Chowdhury, 2005) to successfully execute a turnaround.

A broader definition of failure has been stated as:

“we propose that an organization fails when its ability to compete deteriorates as a consequence of actual or anticipated performance below a critical threshold that threatens its viability” Mellahi and Wilkinson (2010, p533).

So it is more than bankruptcy, it’s about the inability to ‘adapt’ (Harford, 2011) and there is a ‘threshold’, as quoted above, from which there is no return. Having defined failure, in a broader way as above, then ‘success’ given the continuum in Figure 3.1 could be seen as the opposite i.e. competing and performing above a critical threshold. In practice this is usually seen as meeting or exceeding business plans.

To clarify, this research considers the stages that can lead to end of life for a retailer as fourfold:  ‘Narrow’ legal based definition of failure (i.e. bankruptcy including various stages e.g. CVA, Chapter 11 etc.);  ‘Broad’ definition of failure as defined above by Mellahi and Wilkinson (2010);  ‘Financial distress’ (Altman and Hotchkiss, 2006);  Distress (organisational decline defined above (Cameron et al, 1988)).

To illustrate these four stages UK retail examples are presented. In the ‘narrow’ definition Woolworths is a clear example with GAME and HMV also. The latter two have gone through a CVA process and now may be considered to be in the ‘distress’ stage as they seem not to have broken away from this ‘end of life’ position towards success. Within the broad definition would be examples of ASDA in 1992 and Tesco in 2014 resulting in a change of management to effect a turnaround strategy. New Look have been in Financial Distress for some time (see Figures 3.5 and 3.6) and in 2015 moved into the broad definition of failure due to changes in management and ownership. Although Burt et al, (2002b) use the term failure correctly in relation to M+S international strategy, it is only an element of strategy that has failed not the entire corporate business so the retailer should have been viewed as in distress in line with the definition above of organisational decline.

The search for retail performance metrics as predictors of retail failure would suggest a broader definition, as failure may not just be about bankruptcy. To illustrate this for retail, an example would be a week on week, month on month, year on year downward trend of customers visiting the retail outlet or website. This retail performance metric (i.e. customer numbers) could be one of a series of early warning indicators that could predict potential retail business performance, in terms of success and distress. This is explored in the thesis’ primary research.

3.7. Business and management research

3.7.1. Introduction

This section briefly touches on business and management research where, as the examples below illustrate, people talk about performance and even measuring it but are not very good at articulating how this can be done. For example, much is written about successful organisations and successful leadership (Drucker, 1954, Drucker, 1964, Drucker, 2001, Handy, 1995, Handy, 2002, Handy, 2015, Peters and Waterman, 1982, Peters and Austin, 1985) but these types of commentators make generalised statements and haven’t defined the specific metrics to monitor successful performance. For example, in ‘Managing for Results’, Drucker (1964, p183) refers to ‘the key decisions’. ‘Decisions are made and actions are taken at every step in this analysis of a business and of its economic dimensions. Insights are ‘bled-off’ and converted into tasks and work assignments. At every step of the analysis there should be measureable results […] But for full effectiveness all the work needs to be integrated into a unified programme for performance.’

These are generalised statements and Drucker (1964) doesn’t actually identify what the metrics are to get the measurable results. Similarly with the views of Charles Handy (2015, p7) in his most recent collection of essays there are general observations and comments such as: ‘to move forward in many areas of life it is sometimes necessary to change radically, to start a new course that will be different from the existing one, often requiring a whole new way of looking at familiar problems, what Thomas Kuhn called a paradigm shift […] The real problem is that the change has to be initiated while the first curve is still going. That means that those who have been in charge of that first curve have to begin to think very differently about the future, or, more often, let others lead the way up the new curve.’

So referencing a change of leadership, in the quote above, in an organisation and think about the business in a different way does not really help because what is needed are metrics to facilitate the thinking or understanding any of the ‘curves’. When considering books or comments from retail leaders past and present, then similar unmeasurable themes emerge (Honeycombe, 1984, Powell, 1991, Jones, 2005, Leighton, 2008, Woodhead, 2012, Leahy, 2012). For example, Terry Leahy talks about management style and describes his tenure at Tesco in ‘ten words’ such as ‘Trust’ (Leahy, 2012) with no real prescription on how to define and measure trust. The point is that, it doesn’t matter how good the qualitative observation is when made, if it doesn’t translate into measurable performance metrics then decisions become difficult to explain and justify.

3.7.2. Marketing measurement frameworks

There is marketing research (Aaker, 1991, Ambler, 2000, Ambler, 2003, Farris et al., 2010) and retail marketing research (Walters, 1977, Walters and White, 1987, McGoldrick, 2002) but even McGoldrick’s (2002) iconic text book just repeats general ideas such as Porter’s competitive forces on p137 and 147 or a ‘SWOT’ analysis p139 or a ‘balanced score card’ p150. When referring to the monitoring of financial performance (in McGoldrick, 2002, Chapter 6), which is pertinent to trading performance statements, it is just a repeat of traditional accounting measures. He also makes references to the ideal current ratio as 2 to 1 on p216 accepting other researchers’ inappropriate understanding of retail finance. However, it is noted that McGoldrick (2002) was not trying to convey what retail boardrooms should use as metrics merely informing students of various ways performance can be considered. The marketing and retail marketing research does not express what a successful retailer should measure at boardroom level although there are some productivity measures noted, such as direct product profitability which is of little practical use due to the cost apportionment challenges e.g. how much of a CEO’s cost is apportioned to a tin of baked beans?

The literature review of marketing metrics (Ambler, 2000, Ambler et al., 2004, Arikan and Peppers, 2008, Farris et al., 2006, Farris et al., 2010, Srinivasan and Hanssens, 2009, Wind, 2009, Zahay and Griffin, 2010, Rust et al., 2010) has shown an abundance of metrics (see Farris et al (2010) for their 119 metrics). However, the calculation and use of these marketing metrics by retailers is not easily determined from publicly available records thus hindering the understanding of retailers’ performance. A more definitive guide to measuring marketing performance has been produced by Farris et al (2010). This work begins to define more clearly what a metric should be and how it should be calculated at a general business level and thus provides some guidance that may be applied to retail specific metrics, for example the calculation of CLV. The work is based on surveys completed in the USA across industry by marketing managers so to some extent represents one way (a marketing emphasis) to consider business decisions. Its relevance to UK retail is untested. The top ten metrics listed by Farris et al (2010) are: 1. Net Profit 2. Margin % 3. Return on Investment 4. Customer Satisfaction 5. Target Revenues 6. Sales Total 7. Target Volumes 8. Return on Sales 9. Loyalty 10. Annual Growth %

Of the 119 metrics listed it would appear that there is a substantial mix of what could be termed financial metrics (e.g. net profit, sales) alongside marketing metrics (customer satisfaction, loyalty). Not all of these are published in the ARA. Furthermore, this list does not contain retail

specific metrics e.g. LfL sales, sales per ft2, which sometimes can be gleaned from some ARA of retailers. How these marketing metrics compare to those used by this thesis’ case studies is discussed under the primary research findings presented in Chapter 6.

3.7.3. Retail measurement frameworks

Specific literature on retail performance (Berry and Lusch, 1996, La Vere and Kleiner, 1997, Berry et al., 1997, Arnold, 2002), being based on case study approaches aimed at understanding what makes a successful retailer, tends to focus on qualitative assessment producing valuable insight but unmeasurable or ‘soft’ metrics such as: Leadership; Quality; Service; Empowerment etc., things that are difficult to quantify objectively. Berry et al (1997) do reference specific metrics used by Walgreens such as: Sales; Earnings; Sales ft2; prescriptions filled. Berry et al (1997) accept these metrics as used by the retailer but do not assess them any further yet conclude that success is driven by ‘Leadership with Heart’ or a retailer that is ‘Fun to Shop’. These ‘anecdotes’ may be their observations of successful retailers but are of little help in getting to identifiable performance metrics. None of the research to date gives detailed guidance on what retail performance metrics should be used at board room level and how they could be reported to create trust and transparency about a retailer’s trading performance.

3.7.4. Performance measurement theory

As noted by Neely (2002 pxii), there is no agreement on what is important and relevant when it comes to performance measurement. Every researcher seems to follow their own agenda thereby hindering the development of a cohesive model or theory (Neely, 2002). ‘A significant problem is that there appears to be no agreement as to which are the important themes and theories in the field. Everyone writing about the topic has his/her own preferred references and freely cites them. While this diversity is appealing, it also hinders development, because it makes it almost impossible for generations of researchers to build upon the work of others.’

As the quote above indicates, it is difficult to discover any one theory that is relevant and appropriate to explain the problem being researched in this thesis i.e. performance metrics that help understanding of UK retailer performance.

3.7.5. Strategic control

The study of strategic control (Goold, 1986, Goold and Quinn, 1990, Simons, 2013) as a separate discipline emerged from management accounting (Scapens and Bromwich, 2001, Scapens and Bromwich, 2010) and the behavioural school of thought (Hopwood, 1974, Hopwood, 2007). Effective strategic control is likely to be influenced by reliable and trustworthy performance metrics so identifying these metrics is important. Most of the research suggests a

‘top down’ view of the organisation and a longer term perspective i.e. more than just the next twelve months budgetary cycle. This is good in the sense that it deals with board room level metrics. The ideas vary from one simple articulated goal (Collins and Porras, 1996) to creating levers (Simons, 1994) for strategic control systems. The challenge for all of this thought is that the speed of change in the UK retail environment is unprecedented making long term planning a questionable exercise, see Chapter 2. Couple this with the lack of specific retail performance metrics renders some of this academic literature and research as ‘directional’ but limited in implementation.

3.8. Performance measurement and metrics

3.8.1. Introduction

This section outlines the status of the literature review on performance measurement and metrics. Performance measurement is used as a term in the thesis as it refers to the idea that management should measure, monitor and manage performance and this has been articulated famously by Lord Kelvin (Kelvin, 1883) as noted below: “When you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind”

Measures; indicators and metrics are defined, based on the Oxford English dictionary (UoO, 2005), below.

A measure is defined as: ‘a standard unit used to express the size, amount or degree of some aspect of performance’.

An indicator is defined as: ‘a statistic which provides specific information on the state or condition of specific aspects of performance’.

A metric is defined as: ‘a measuring system that quantifies a trend, dynamic or characteristic of performance’.

Although the three definitions above may appear similar they are treated as quite different in this thesis and the following example illustrates the difference. ‘Sales’ are considered a metric and they are measured in units of ‘pounds sterling’ making the ‘pounds sterling’ a measure. When sales are compared to the prior year for year on year (‘yoy’) performance and a decline is noted the ‘negative yoy’ is the indicator. As shown in the quote below, Ferris et al (2010, p1) metrics are critical for understanding performance.

“In virtually all disciplines, practitioners use metrics to explain phenomena, diagnose causes, share findings, and project the results to future events. Throughout the worlds of science, business, and governments, metrics encourage rigor and objectivity. They make it possible to compare observations across regions and time periods. They facilitate understanding and collaboration”

3.8.2. Financial measurement frameworks

Financial performance measurement and reporting has become the de facto standard for all businesses including retail due to the statutory requirements of the Companies Act (HMSO, 2009). An entire industry, in the form of the accountancy and auditing profession, now exists in the production, verification, reporting and regulation of this financial information (President- ICAEW, 2014).

The thesis shows most of what is produced within this arena of ARA is nowadays done specifically for external publication and not for management use. Management use internal reports and metrics yet choose not to share this information in the public domain. This practice of selective performance metric disclosure is explored in Chapter 6. The ARA is historic and backward looking and the process that produces it has been described by the accountancy profession as ‘broken’ in Chapter 1. The reliability of what is reported (Smith, 1992, ICAS, 2010) has been continuously questioned over the years with Tesco being the most recent example resurfacing the debate into the public arena.

Argenti (1976) raised the issue of ‘creative accounting’ as a defect and contributor to corporate collapse in his seminal paper almost 40 years ago. Yet 50 years of regulation development seemingly have not dealt with the challenge of making these published reports simple and clear conveyors of performance. The underlying theory, development and limitations of accounting have been discussed above in section 3.3. The key message from this section is that published financial information alone is insufficient to provide a good understanding of a retailer’s performance.

3.8.3. Business measurement frameworks

Turning to academic literature, some models and theories emerge (Hatch and Cunliffe, 2013) the most cited and taught in strategy classes is the work of Porter (Porter, 1980, Porter, 1985) but these are broad frameworks that do not funnel down to specific and relevant metrics for the retail sector. For example ‘the value chain’ concept is discussed with the firm being a series of functions with these functions divided further into value chains. For example, sales and marketing and then within this promotion (Porter, 1985, p46) but it falls short of mentioning specific metrics for measuring promotion. What does emerge from some of this type of literature is that any performance measurement analysis should be industry focused and geographically

specific for greater relevance be it competitor analysis, town centre development or retailer trading performance (Porter, 1974, Porter, 1979, Porter and Solvell, 1999, Wrigley, 2014).

3.8.4. Balanced Score Card

The most cited business performance measurement framework in the performance measurement literature is the work of Kaplan (Kaplan, 1984, Johnson and Kaplan, 1987, Kaplan, 1994, Kaplan and Atkinson, 2015) and more recently the balanced score card and strategy maps (Kaplan and Norton, 1992, Kaplan and Norton, 2001, Kaplan and Norton, 2004). The balanced score card attempts to create a framework for monitoring the entire organisation and suggesting the use of a ‘dashboard’ (Euske et al., 1993, Lebas and Euske, 2002, Neely et al., 2005) and this framework is further developed to try and deliver more detailed guidance on strategy and the linking of strategy to performance measures (Kaplan and Norton, 2000). These and other frameworks (Otley et al., 1995, Otley, 1999, Mellahi and Wilkinson, 2004) still fall short of providing specific metrics that could be used by retail management. The concept of the balanced score card can be seen in the public domain in very general terms in the performance measurement tools once used by Tesco. “To manage Tesco in a balanced way we use a management tool called the Steering Wheel. It is divided into four quadrants – Customer, Operations, People and Finance – which, in turn, are divided into several segments, each with a set of Key Performance Indicators (KPIs) which are based on demanding but achievable targets. Performance is reported quarterly to the Board, and a summary report is sent to the top 2,000 managers in the company to cascade to staff. The remuneration of senior management is shaped by the KPIs, with bonuses based on a sliding scale according to the level of achievement on the Steering Wheel” Tesco, (2005, p8)

Figure 3.8 Tesco wheel, source: adapted from Tesco Corporate Responsibility Review 2005, p9

We trust and respect each other

By 2014, the Steering Wheel, Figure 3.8, has been replaced by three big ‘scale for good’ ambitions (Tesco, 2014b) and although there seem to be KPIs attached to these three

ambitions, Tesco appears to have lost the balanced score card. Sainsbury (JSainsbury, 2014) also use (Figure 3.9) what appears to be a balanced score card in wheel-like format.

Figure 3.9: Sainsbury’s wheel, source: adapted from J Sainsbury plc 2014, p0

Operational excellence

Growing space Great food & creating property value

J Sainsbury plc Compelling Developing general new business merchandise and clothing

Complementary channels and services

Our values make us different

Implicit in both wheels above are detailed performance metrics but these details are not clear from the reports although it is assumed these details can be obtained from within the company.

3.9. Summary of literature review

This chapter has provided a review of the diverse academic disciplines that hint at performance, success and failure. Section 3.2 put this literature on to a map to provide a picture of the sources of commentary that may give some insight into how to understand a retailer’s trading performance.

In section 3.3, accounting and finance literature is critically reviewed and it is demonstrated that the ARA may not provide an accurate reflection of a retailer’s performance even though it has become the de facto standard for performance reporting in the public domain. In the words of the President of the ICAEW when questioned by the Select Committee, “where were the dogs that didn’t bark”. This is in reference to the banking crisis, the risk reporting and the auditors of these banks. A clear recognition that the current process is ‘broken’ consequently any use and interpretation of accounting information should be handled with an understanding of its weakness. This point has been aptly and publicly articulated by the former Enron, Chief Financial Officer in a recent speech in London:

“There may be a fundamental difference between a company following the rules and a company presenting a true picture of its financial position” Andrew Fastow, (Sheppard and Hume, 2015)

Similarly, section 3.4 gives an outline of governance and risk theory as well as the practical developments in the UK. The conclusion from this review is that the reporting of governance and risk is becoming ‘rules’ bound, has little to do with theory e.g. the absence of Company beta in any reporting, the theory and practice are not aligned and neither illuminates performance metrics that could help in understanding business performance let alone retailer performance. There is much written on both business success and failure with an abundance of research applied to the construction of business failure prediction models. The review of the literature on failure prediction models shows that they are fundamentally flawed as they are reliant on historic published financial information in the form of the ARA and the ratios derived therefrom. Section 3.3 has already demonstrated the inherent weakness in accounting and finance information. Hence the models are predicated on the wrong variables yet researchers have continued to develop different model types using the same inherently weak variables. These models are also weak particularly since the models do not accommodate behavioural aspects of decision making. Certainly the reliance and acceptance of these types of models within the ‘practitioner world’ is limited. They are therefore inadequate mechanisms for performance assessment and as for predictive ability seen as academic exuberance.

The case study research on business success and failure as well as retail success and failure provides some insight although most of this is either operational and at a detailed level or high level dealing with, for example, ideas of leadership traits. Hence the research gap is addressed with the objectives of this thesis, with the three objective statements repeated below:

 To identify what retail performance metrics are used by retail boards to manage their performance. In so doing: o To identify any commonalities amongst the performance metrics used by retail board directors. o To determine whether or not retail performance metrics change over time.  To identify what retailers claim about their performance in the public domain; and  To explore any disconnect between the two objectives above i.e. the connectivity between the performance metrics retail boards’ use and those they publicly report.

Chapter 4: METHODOLOGY AND METHODS

4.1. Introduction

The chapter structure is shown below by section heading and each of these headings are outlined in this introduction.  Research objectives related to methods;  The pragmatic philosophical world view;  Challenges to the approach and methods;  Multiple case study design in three phases;  Applying grounded theory;  ‘Dimensional’ analysis and the explanatory matrix; and  The inductive and interpretive presentation of findings.

This chapter begins with a restatement of the research objectives linked to the methods. It places this research as tending towards pure research on the “business and management research continuum” (Saunders et al, 2011, p11) that ranges from ‘basic’ research to ‘applied’ research. Applied research is aimed at finding solutions to organisation or management problems whereas basic or pure research is aimed at understanding and expanding knowledge of the processes of business and management.

The next section sets into context the epistemology and ontology inherent in this research design. It explains the pragmatic philosophical approach adopted for this research (Dewey et al., 1917). This approach recognises there is more than one way to see the real world and more than one way to undertake the research. The pragmatic approach emphasises the research problem as critical over and above the methods used and accepts that many and varied methods can be used to obtain data about practical business issues (Saunders et al., 2012, Creswell, 2013).

The section on conducting the research plan in three phases explains the methods and techniques used at each phase. The phases are essentially a breakdown of the research process into three sequential and logical time and activity periods to demonstrate planning rigor and a disciplined approach to data collection. A mix of methods (Creswell and Clark, 2011) is used with the quantitative elements, such as preliminary survey information and secondary data analysis, supporting the preparation and execution of semi-structured interviews with retail board members and their resultant qualitative data (Cassell and Symon, 2004).

This chapter then outlines the limitations and challenges in conducting the research for this study, as well as briefly explaining the rationale for avoiding other potential methodological

approaches that could have been employed (Sayer, 2010, Scapens and Bromwich, 2010, Smith, 2011). There is also a section on the researcher’s background and the implications for conducting a project of this nature.

The next section then describes the various approaches to case study research, such as replicated multiple case studies (Yin, 2009), one of which typically adopts a positivist approach to develop testable hypothesis to contribute novel insight by examining differences (Eisenhardt, 1989, Eisenhardt and Graebner, 2007) and has been followed by many researchers (Langley and Abdallah, 2011). Although this exact approach is not used in the current study, there are similarities, not least in using multiple cases and multiple participants within cases. In essence, therefore, it will be shown that the current study amounts to a replicated interview method within and across the six retail company cases, with the retail company forming the unit of analysis. This particular approach aims to capture what retail board members actually consider important in terms of performance metrics and within that the specific sub-group of non-financial (termed by this researcher as ‘retail performance’) metrics and gain an understanding of which ones they commonly use. The underlying research strategy (see Figure 4.2 for terms used and their meaning in this thesis) is a particular style of an ‘informed’ grounded theory approach (Glaser and Strauss, 1967, Gioia and Chittipeddi, 1991, Thornberg, 2012) noted below.

Within grounded theory there is a particular style of analysis that aims to give rigour to the analytical process called ‘dimensional’ analysis (Schatzman, 1991), this is what has been used for this thesis. There is a detailed description, section 4.9, of the use of ‘dimensional’ analysis (Schatzman, 1991, Robrecht, 1995) as this is the basis upon which the analysis develops the concepts. Using an explanatory matrix, constant and interactive comparison of data collected is undertaken to develop salient dimensions that can form multiple novel concepts or viewpoints. One of these viewpoints then becomes the dominant perspective from the research as it more closely represents the understanding of the research problem (Kools et al., 1996).

The last section of this chapter sets out the way in which the inductive and interpretive findings are presented. It explains how the first part of the findings involved a summary of those retail performance metrics that were communicated by case study participants, as described in their own words and terms. These data are then structured into a first order thematic analysis to arrive at Focus Areas (FAs) where performance metrics are concerned, and then further subjected to a progressive abstraction leading to second order analysis and additional insights (Gioia and Chittipeddi, 1991, Langley, 1999, Langley et al., 2013, Gioia et al., 2013). Thus the case analysis reflects the pragmatic philosophy and fits within an overall interpretivist tradition.

4.2. Research objectives related to methods

The purpose of the thesis is to work towards getting a better understanding of UK retailers’ trading performance through their use of non-financial performance metrics. Chapter 3 clearly showed that what retail companies publish as their performance in the ARA are not mutually consistent i.e. retail performance metrics and the ARA, and for this reason difficult to rely on the ARA for the purposes of performance comparability. Two examples of Tesco and Woolworths were used to illustrate this fact. This indicates that there is no one best way to undertake research on retailer performance and that the methods used will impact upon what can be discovered in this respect.

According to Saunders et al, (2012) all business and management research can be placed on a continuum according to their purpose and context see Figure 4.1 below. As mentioned in the introduction to this chapter, this continuum ranges from pure at one end to ‘applied’ at the other end. Applied research has been described as focused on solving a business or management issue and frequently conducted within an organisation with a view to providing recommendations.

Figure 4.1 Pure and applied research continuum (from Saunders et al, 2012, p11)

Research Continuum

Pure Applied research research

Purpose: Purpose:

• improve understanding of • expand knowledge of processes of business and management particular business or • results in universal principles management problem relating to the process and its • results in universal principles relationship to outcomes relating in solution to the problem • findings of significance and value • new knowledge limited to the to society in general problem • findings of practical relevance and value to manager(s) in organisation(s) Context: Context: • undertaken by people based in • undertaken by people based in a universities variety of settings including • choice of topic and objectives organisations and universities determined by the researcher • objectives negotiated with • flexible time scales originator • tight time scales

Pure research is aimed at understanding a phenomena and expanding the body of knowledge and results in conceptual frameworks, models or theories. Both pure and applied research, Saunders et al (2012) indicate a need to contain the potential for taking some form of action in the realm of business and management activity. On the basis that the continuum runs left to

right and reflecting on the continuum, this research project positions left of centre towards pure research since the topic and objectives have been chosen by the researcher and the findings benefit both the research community and retailers.

The research objectives are presented in Table 4.1 below aligned against the methods and analysis techniques and where in the thesis the findings can be located.

Table 4.1: Objectives, methods and analysis

Objective Methods Analysis technique To identify what retail Semi-structured interviews First order analysis performance metrics are used with case retailer board presenting findings in the by retail boards to manage their directors supported by voice of the participants performance. survey data and company- using NVivo 10 to code provided information. metrics In so doing:

to identify any commonalities Case study within and across First order thematic amongst the performance retail companies and analysis into aggregated metrics used by retail boards; company annual reports FAs using NVivo 10 to and aggregate coda. A track back 10 year historic to determine whether or not view of the case study Second order analysis retail performance metrics company with current and abstraction from FAs change over time interview transcripts and using dimensional company provided historic analysis. Review of coda information for exemplar on NVivo 10. retail companies.

To identify what retailers claim Semi-structured interviews First order analysis about their performance in the with case retailer board presenting findings in the public domain directors supported by voice of the participants survey data and company- using NVivo 10 to code provided information. metrics A track back 10 year historic Second order analysis view of the case study and abstraction from FAs company with current using dimensional interview transcripts and analysis. Review of coda company provided historic on NVivo 10 information for exemplar retail companies.

To explore any disconnect Semi-structured interviews Second order analysis between the two objectives supported by survey data and abstraction from FAs above i.e. the connectivity and company provided using dimensional between the performance information and annual analysis. Comparative metrics retail boards’ use and reports. and iterative analysis those they publicly report using NVivo 10.

4.3. The pragmatic philosophical world view

This section presents the epistemology and ontology that underpins this research. The pragmatic worldview adopted can be traced to the school of thought that believed doing scientific research in a structured and rigid quantitative process only added to incremental learning (Dewey et al., 1917) whereas a new way of approaching research would allow for ‘new’ and more creative knowledge and this is illustrated by the quote below. “Intellectual advance occurs in two ways. At times increase of knowledge is organized about old conceptions, while these are expanded, elaborated and refined, but not seriously revised, much less abandoned. At other times, the increase of knowledge demands qualitative rather than quantitative change; alteration, not addition.” (Dewey, 1917, p3).

This pragmatic approach which has been developed by social scientists recognises there is more than one way to see the real world and more than one way to undertake the research. The pragmatic world view emphasises the research problem as critical rather than the methods used and accepts that many and varied methods can be used to obtain data about practical business issues (Creswell and Clark, 2011, Saunders et al., 2012, Creswell, 2013)

4.3.1. Ontology and epistemology

There has been and continues to be a debate about the ontology and epistemological philosophies that underpin research because they impact the choice of methods for conducting the research and the conclusions that can then be drawn therefrom. Often this debate is conveyed by opposing views of positivist or interpretivist philosophies, which frequently feed through to quantitative versus qualitative methods.

The positivist philosophy, rooted in the physical and natural sciences, adopts an objective and rational approach focusing on observable phenomena and aims to be value free, whereby reality is external to the researcher in the search for truth. The research process is typically incremental and highly structured within a research style until there is a paradigm shift (Kuhn, 1962, Kuhn, 1977, Kuhn, 1990). This rational approach to research typically creates hypotheses for testing theories and models through experimentation, observation and data collection. Such a philosophy is also adopted by many business and management researchers (Saunders et al., 2012) . An alternative philosophical perspective is that the nature of reality is perceived by the researcher and hence socially constructed (Sayer, 2010). The researcher is influenced by experience, knowledge, language, culture and political systems and therefore takes a subjective view of observable phenomena. Any conclusions are inevitably value bound and there may be limits to generalizability of conclusions. The important recognition, however, is not the philosophical nature of the enquiry but the ability to reflect on the choices of method and accept and understand the methodology that underpins them.

100

Saunders (2012) succinctly summarises the interrelationships between research philosophies and other methodological choices using a ‘research onion’ diagram. This has been adapted to illustrate the choices made for this thesis in Figure 4.2 below. A research philosophy rooted in a pragmatic world view and interpretivist understanding, feeds through to an essentially inductive approach. The data collected and the analysis has been based primarily on qualitative methods supported by some quantitative tools. The main strategy adopted has been multiple case study research using informed grounded theory and this is discussed in more detail later. The individual techniques employed involve semi-structures interviews, survey work and secondary data interrogation. Figure 4.2 demonstrates the ‘backward’ nature of the entire research process in choosing the techniques and tools used to conduct the primary research imply a dependency on the strategy adopted which in turn is influenced by the methods. The methods are shaped by the approach taken which reflects philosophical viewpoints. Some researchers may wish to represent the process in the opposite direction. Figure 4.2: Research philosophy’s influence on method choice

Techniques Strategy Methods Approach Philosophy

Replicated semi- structured interviews Multiple retailer case Qualitative with a Inductive approach Pragmatic world with board directors studies using mix of tools & view & interpretivist and retail experts Informed Grounded techniques understanding supported by some Theory with a survey tools and particular emphasis retail company on first and second internal and order analysis using a published external dimensional matrix for information. conceptualisation of Covering a cross- the findings section of six retailers over a ten year time horizon

Based on Figure 4.2, it becomes apparent that the view taken considers that the ‘objective’ aspects of retailer performance and related metrics as less important than the way in which the retail board members focus in on those specific metrics they consider important. This emphasises the research philosophy of a pragmatic world view and interpretivist understanding.

101

In the words of Saunders et al. (2012, p137): ”The challenge here is to enter the social world of our research subjects and understand their world from their point of view […] Not only are business situations complex, they are also unique. They are a function of a particular set of circumstances and individuals coming together at a specific time”

The quote above neatly summarises the research situation whereby the aim is to understand what retail board members consider as important metrics for the retail company when they come together at board meetings. Such an interpretivist view comes from the intellectual tradition of phenomenology, which considers the human aspects of making sense of the world (Cassell and Symon, 2004).

Four research paradigms were proposed by Burrell and Morgan (1979), based on a two by two matrix of subjectivist to objectivist and radical change to regulation see Figure 4.3. The interpretivist paradigm falls within the subjectivist regulation quadrant and from a philosophical world view this is about how humans attempt to make sense of the world. This proposed paradigm view implies a degree of exclusivity in that one approach means that the other is foregone (Burrell and Morgan, 1979). For instance a subjective view undertaken implies that the researcher has foregone any objectivity. This is a narrow and pessimistic view of social science research (Roberts and Scapens, 1985) that can be improved upon to reflect the context within which the research is being undertaken (Ahrens, 2008).

Figure 4.3: Two dimensions and four paradigms for social theory analysis

Radical change

Radical humanist Radical structuralist

Subjectivist Objectivist

Interpretive Functionalist

Regulation

102

As noted in the quote below, more recent research methods and techniques, particularly in interpretative management accounting (Ryan et al., 2002), overcome this narrow viewpoint. “Methodological debates in accounting frequently emphasise the distinction between objective and subjective research. A growing body of interpretative management accounting studies, often based on fieldwork, is continuing to develop approaches that seek to overcome that distinction by exploring the various ways in which accounting can become part of the contexts in which it operates” (Ahern, 2008, p292).

Notwithstanding the potential confusion that can arise from not taking a particular stance e.g. positivist and quantitative, the pragmatic worldview adopted allows for multiple perspectives to tackle the research question, and in this instance there is an appreciation of the social construction of phenomena common within an interpretivist understanding and tradition. However, this does not mean that objectivity in the research process is completely ignored nor has the rigor and independence of the researcher been put aside. As noted under the challenges section below in 4.4.6, there is a recognition of the researcher’s position in the entire research process.

4.4. Challenges to the approach and methods

4.4.1. Introduction

There were a number of ways the research question (i.e. working towards an understanding of retailers’ trading performance) could have been approached in terms of research design. The research gap shown in Chapter 3 indicates that the subject matter can straddle many academic disciplines such as accounting, management, strategy, marketing and so on. Given that, as shown in the previous chapter, historical published financial information does not adequately convey a retailer’s trading performance, then finding ways to get to metrics that might help in this understanding was the logical next step.

4.4.2. Analysing published information

Corporate disclosure and in particular voluntary disclosure of metrics such as LfL sales could be conducted through secondary research. Studying published information such as trading statements and the ARA may give some insight but this does not necessarily get to what companies are actually using as performance metrics, only what they publicly disclose which may be a carefully managed disclosure (Merkl-Davies and Brennan, 2011, Merkl-Davies et al., 2012, Brennan and Merkl-Davies, 2013). This type of positivist empirical research is popular amongst accounting researchers (Ryan et al., 2002, Beattie et al., 2004, Beattie and Thomson, 2007) although historically done extensively in the USA where scores were given and published they were subjective and have now been abandoned. As noted in the quote below

103

there is a need for metrics within the process (Beattie, 2002) for getting to non-financial metrics when reading the narrative statements in the ARA. “A difficulty in conducting research into voluntary disclosure is the need for metrics to measure salient attributes, such as amount or quality” (Beattie, 2002, p105).

As the quote suggests ‘salient attributes’ are required to validate voluntary disclosure such as the ‘amount’ of times a reference has been made (e.g. the number of times the word customer has been used) or some form of ‘quality’ assessment. Although analysing ARA has become popular, little has been done with retailer ARA and on retailer non-financial information disclosure within ARA and trading statements, in particular. Hence, a review of a set of retailer ARA was done and specifically used to create a template (King, 2006) for content analysis and coding interview comments. However, as noted earlier and confirmed by the primary research in Chapter 5, this type of secondary research can only give a superficial understanding.

4.4.3. Model building

Using ratio analysis to create failure models was shown in Chapter 3 to be ineffective in understanding a retailer’s performance and is described as ‘academic exuberance’. This positivistic approach to testing quantitative data to predict retail failure has therefore been dismissed because these models are fundamentally flawed in their construction (Morris, 1997, Balcaen and Ooghe, 2006) and didn’t work when applied to UK retailers as shown in Chapter 3, Tables 3.6, 3.7, 3.8 and 3.9. Adding non-financial metrics such as LfL sales, customer numbers etc. to these retail failure prediction models was considered as a potential way to improve them. The aim being to improve their accuracy by reducing the Type I and Type II errors (i.e. Type I implies there is a genuine effect in the population, when in fact there is not and Type II implies there is no effect in the population when in reality there is) (Field, 2009). However, such non- financial metrics are not readily available in the public domain, as noted in the Chapter 3. In addition, financial and economic models have been criticised (Buttonwood, 2015, Thaler, 2015) for not reflecting reality so extending this criticism of models ( in that exclusive reliance on normative theory leads to making systematic, predictable errors in forecasting models) also applies to business failure prediction models. People are not rational, or at the very best are predictably irrational (Ariely, 2009b) and the business context and related operational processes are complex which means that aspects of behavioural: economics; accounting; political environment etc., need to be included into failure prediction models for them to be credible in the real world.

4.4.4. Surveys

One approach that could have been used was surveys (Fowler, 2009) as a means to collect data on metrics used by retailers, similar to the research on marketing metrics which gathered data from marketing managers in the USA (Farris et al., 2010). The challenge with surveys into

104

retail boardrooms is to get to the right person and then get the response. This is discussed under the phase one section 4.5.2.

4.4.5. Focus groups

Another method considered but not used was focus groups and specifically Delphi panels (Strauss and Zeigler, 1975, Preble, 1984, Rowe and Wright, 2001, Okoli and Pawlowski, 2004, Duboff, 2007). A similar challenge to survey response was considered as board members indicated a reluctance to give up the time and to participate in a group forum. Even the use of electronic on-line communication was considered too much time by board members, from whom ‘soundings’ were taken. Two out of the ten board members were however prepared to have their company participate in a confidential case study.

4.4.6. Researcher experience and influence

In the preface to the thesis a brief biography of the researcher is given. It is acknowledged that having spent almost ten years in senior management roles at Tesco and ASDA some of which were in Finance that there may be a degree of embedded subjectivity inherent in the thinking. When this is coupled with the fact that the researcher is a Fellow of the Institute of Chartered Accountants there is an additional risk of financial training bias. In addition, the last ten years as an experienced partner with Deloitte and Head of the UK Retail practice meant becoming familiar with and advising most of the UK retailers. All this knowledge and experience is difficult to extract out of the mind when conducting the research activity. As noted above, being mindful of potential bias is important and although it can be viewed as a disadvantage, there are also considerable advantages from this experience and knowledge such as experience of retail metrics and financial information used by some retail companies.

4.5. Multiple case study design in three phases

4.5.1. Introduction

The three phases of the research carried out in this study are illustrated in Figures 4.4 and 4.5 below. The research was sequenced in this way to assist the planning and timing of the work and ensure that it was tackled in an integrated and disciplined manner. The core part of the research was through a multiple case study strategy, using semi-structured interviews with board members in retail companies. This was supplemented with additional information in the form of reports and documentation from the case companies. However a certain amount of planning and preparation preceded this in phase one as noted below.  Phase one, in this preparation stage some primary and secondary data collection and analysis has been conducted to inform the case studies’ design.

105

Figure 4.4: Three phase research design for this thesis

Phase 2: Qualitative data collection and qualitative data analysis supported by preparatory survey questionnaire and company specific information

Pilot survey of key performance metrics

UK retail Integration and Qualitative data ‘failure models’ triangulation of collection Qualitative data Review of application test analysis findings published ARA

Phase 3: Integration and triangulation of Phase 1: Primary & secondary data collection and qualitative and quantitative results within case

analysis to inform case study design studies and between case studies

106

Figure 4.5: Detailed research process steps for this thesis

Phase 1 Interim step Phase 2 Phase 3

Procedures Procedures Procedures Procedures • Identify UK retail population and • Use Phase 1 products to frame • Case selection using replication • Triangulation of data from distressed retailers therein the research tools for Phase 2 logic multiple sources for overlays and commonality • Run failure models on distressed • Establish a case study database • Preparatory questionnaire for population to test model efficacy individual board members. • Interpretation of results using: • Establish case selection st • Identify non-mandatory reporting protocols • Qualitrics toolkit for survey design • 1 Order Analysis; and nd metrics in ARA • Re-design 'exploratory test and execution • 2 Order Analysis • Prepare 'exploratory test survey' survey' as a preparatroy • Interviews with each board member • Reflections on objectives using open design questionnaire, questionnaire for interviews • Request for information documents using on-line Qualtrics toolkit, for • Design a semi-structured • Audio recording of interviews retail board members to collect data interview process • Transcribe audio recordings on day to day metrics used • Establish a 'wish list' of internal • Coding and thematic analysis • Survey 40 board members to comapny documents assess appetite for Delphi panels & • Confirmatory validation of coding • Determine public record sources case study candidates • NVivo qualitative software Products for further information • Comparison of day to day metrics • Discussion and implications with external published information • Preparation of results for each • Ratio and risk analysis from FAME case study as a working paper database Products • Analysis figures, tables, Products • Interview protocol matrices & chapters • Pilot questionnaire including: telephone & • Further research suggestions email contact and Products • Data identifiying metrics in • Gaps interview agenda. use • Audio record of interviews • Call to Action • Cross section of retailers • Ratio analysis within retail • Text data (inteview transcripts, to approach for multiple failure models board papers and screen shots case study research. from systems used) • Non-mandatory metrics in • Identify distressed failed ARA • Cross grid analysis of data retailers patterns and thematic matrix

107

 Phase two, where qualitative data was collected and analysed. This analysis was supported by company information such as management accounts and financial reports plus survey responses that facilitated the participant interviews.  Phase three, was a triangulation of data from participants’ interviews within each case study as well as across the multiple cases. This was compared with the artefacts provided by the participants and gathered in the secondary research. This enabled the findings presented for first order analysis (Gioia and Chittipeddi, 1991, Langley, 1999, Langley, 2007) in Chapter 5 and second order analysis (Gioia et al., 2013) presented in Chapter 6 (Langley et al., 2013). This first order and second order informed grounded theory process is explained in more detail in later sections, 4.7 onwards.

4.5.2. Phase one

The initial planning phase consisted of three core pieces of preparatory work:  Gathering secondary information on the UK retail company ‘population’;  Creating a survey tool for the semi-structured interviews; and  Creating an initial template to assist and speed up the coding of case study data. Testing failure models against company published financial information was done and has been presented in Chapter 3 and discussed in the ‘challenges’ section 4.4.3.

4.5.2.1. UK retail population

The UK retail company population was obtained from Mintel’s Annual UK Retail Rankings Report (UKRR) (Mintel, 2011). This report lists, by sector, the top 692 retailers by turnover, with a turnover of £6m or more and an aggregation is shown in Table 4.2.

Table 4.2: Top 5 to top 200 UK retail population by market share

Retailers Market share %

Top 5 39.4 Top 10 52.4 Top 25 64.8 Top 50 75.2 Top 100 83.8 Top 200 95.5 Source: UKRR Mintel, 2011

Company and financial information was also sourced from Retail Week Knowledge Bank and the FAME database for the top 200 retailers representing 95% of the retail market share by turnover. The UKRR list was used to judgementally select a cross section of 40 retailers that had also attended two Manchester Business School retail events (“High Street Bye or Buy?” February 2012 and “MBS Retail Review with BDO Retail Trading Update” February 2013).

108

Attendees were approached during the events (and emailed) with an initial survey to assess appetite for participating in a case study for gathering retail performance metrics actually used.

4.5.2.2. Survey design

The seven question survey was designed using the ‘Qualtrics’ toolkit available on the MBS portal. The survey used is given in Appendix 3. It was designed with the first page as two ‘open’ questions, as shown below, and only these were visible to the participants on the survey portal before they progressed to the next five questions. Q1: Please list the top five metrics (any type of measure) you use to measure performance of your retail company Metric 1 Metric 2 Metric 3 Metric 4 Metric 5

Q2: Please list the top five metrics (non-financial and/ or non-mandatory) you use to measure performance of your retail company. These may include derivations of financial measures such as EBITDA or Like-for-Like Sales. An example of non-financial might be customer visits / numbers. Metric 1 Metric 2 Metric 3 Metric 4 Metric 5

The purpose of open questions was to try, as much as possible, to avoid leading the participant. Although giving three explanatory examples might imply the participants are being led to a certain degree, it is certainly not an exhaustive menu to choose from as used in many other research surveys (indeed later findings, see Chapter 5, demonstrate that retail board directors have some very independent views and are not easily led). In addition restricting the possible responses to five was to focus attention onto what was considered most important.

The next question gave 12 areas for the participant to do a relative ranking and the fourth question asked about when the metrics are reported and consequently discussed. The next question was aimed at assessing appetite for further participation in the research via questionnaires, interviews or workshops, and the penultimate question sought permission to use the information publicly or if requested it would be kept confidential. The final question was open, giving an opportunity to provide further information. Overall the survey was timed as about 15 minutes in duration to complete.

The ‘exploratory test survey’ was tested on three senior academics prior to emailing 40 participants with the ‘hyperlink’ to the survey for completion. Ten responses were received providing a first step in gathering metric data. The survey also appeared to concentrate management attention and elicit valuable metric information. However, a 25% response rate may be considered good but getting to a response number of say 500 (in order to do a meaningful statistical analysis), using straight line extrapolation, would imply the survey

109

population would have to be 2,000. Getting access to this number of up to date email names and addresses of retail board directors was not possible. The ten responses did however confirm that the survey would be a good tool to identify case study participants and inform some of the lines of enquiry for semi-structured interviews within those cases.

4.5.2.3. Template analysis

Using template analysis (King, 2006) in the context of this research, was to start the process of thematically analysing and coding the transcripts from the interviews as well as other information provided. The aim was to increase objectivity and coding reliability across the multiple case studies. A template is a flexible technique that allows for tailoring as the coding develops and works well within the NVivo software (Bazeley and Jackson, 2013) used to code the data. In this instance, the process of creating a ‘first pass’ coda is in line with interpretative phenomenological analysis. Hence, template analysis is particularly useful when the interviews of different people are to be compared within and between case studies.

A template coding structure, that will evolve once the data are analysed, was prepared from a narrative analysis of an ARA. Given that the coda should actually develop and evolve from the data, the purpose of the template coding structure was to ‘kick start’ the coding and analytic process. Therefore, preparing more coding structure templates from large numbers of narrative reports of many retail companies over long periods of time was not necessary i.e. do one and let it evolve from the data analysed. To get to this ‘first pass’ code structure, before the interviews are coded, a set of ARA was selected. The creation of this template coding structure specifically focused on a ‘watershed’ document. The ARA of Woolworths Group plc 2008 was used as it was the last one before Woolworths went into administration. The process adopted was to screen key narrative statements for any metrics these mentioned (Hussainey et al., 2003, Schleicher et al., 2007, Abraham and Cox, 2007). The statements used were those of the: Chairman; Directors’; Chief Executive; and Finance Director, as they are the primary narrative statements in an ARA made by board members to communicate performance to shareholders.

Discourse analysis techniques have been used on the narrative statements (Dick, 2006), above mentioned, by colour coding paragraphs of text for positive, negative and neutral comments to ascertain the level of positivity in these as a simple indicator of the complexity underpinning the reasons for the narrative communication style and also any potential hubris (Brennan and Conroy, 2013, Brennan and Merkl-Davies, 2013).This also provides further evidence of the unreliability of annual report and accounts for understanding retailer performance as discussed in Chapter 3, because within six months of publishing the accounts, Woolworths Group plc went into administration (Burden, 2014, Manning, 2014). In addition, the narrative was analysed line by line for any potential metrics mentioned and a list compiled (this original template coding structure is given in Appendix 4). Reference to financial metrics was simpler to identify, often

110

relating to terms like ‘sales’, ‘profit’ and ‘debt’. Reference to non-financial metrics was more difficult to determine, but reference to customers and their satisfaction was typically one area of identification. The results, summarised in Table 4.3, presented 154 references to potential metrics which can be categorised into a number of possible themes (see Chapter 5 for thematic categories and explanations) where management have focused their attention.

Table 4.3: Summary of template coding structure

CODA / reference Chairman’s Directors’ Chief Finance Total report report Executive’s Director’s report report Page number 03 04-09 10-15 16-19 03-19 Number of pages 1 6 6 4 17 Financial metrics in themes: Profit & loss account 10 23 19 52 Balance sheet 4 2 8 15 Cash flow 1 1 Ratios 1 9 5 14 Total 15 34 33 82 Retail metrics in themes: Customer 1 2 5 8 LfL Sales 1 4 5 People 1 1 2 Store properties 3 3 Buying & products 8 9 17 Business model 1 4 5 Risk & regulation 3 1 4 Brands 5 7 12 Digital 2 2 4 Supply chain 1 4 5 Market share 2 5 7 Total 3 25 43 1 72

TOTAL 18 25 77 34 154

As discussed in Chapter 3 and also in Chapter 6, the trust that can be placed in this ARA as a ’true and fair’ view and signalling ‘red flags’ is questionable since they were signed off in April 2008 and the group went into administration in November 2008. This also highlights the challenge of the ARA in fully explaining a retailer’s performance. Management have complied with the rules and signed off and in the words of the former Enron Finance Director: “There may be a fundamental difference between a company following the rules and a company presenting a true picture of its financial position” Andrew Fastow, (Sheppard and Hume, 2015).

4.5.3. Coding confirmability testing

The template coda has undergone an independent validation exercise from a member of the Manchester Business School faculty. This inter-coder reliability testing exercise was performed by Dr. Daniel Hampson, Lecturer in Marketing and he has been completely independent of this research project. The validation process involved the independent recreation of the template coding structure using the same Woolworths Group plc 2008 ARA as well as the re-coding of

111

the Finance Director transcript from the interviews. The results were then compared to validate and confirm the coding quality and consistency of method for the thematic analysis.

In addition, there was further validation of the template coding structure, through discussion of the results with specific individuals who represented a panel of retail expertise e.g. global head of retail of the largest professional services firm in the world, see section 4.5.4.3 for the experts consulted.

4.5.4. Phase two

4.5.4.1. Introduction

The second phase of the research consisted of identifying, i.e. purposive controlled sampling, the case studies and arranging for them to participate in the research and identifying, i.e. purposive controlled sampling, the board members for interviews and then collecting the data. As shown in Figure 4.1, the strategy chosen was case study research which was inductive in approach, qualitative in method and covered six retail companies. Typically when conducting this type of research there are potential data quality issues in relation to: reliability; researcher bias; generalisability and validity. In subsequent sections, the process, methods and techniques used are described and have been designed to mitigate these potential issues. Yin (2009) has a two-fold technical definition of case study research. This is given in the quote below. “it investigates a contemporary phenomenon in depth and within its real life context, especially when the boundaries between phenomenon and context are not clearly evident […and] copes with the technically distinct situation in which there will be many more variables of interest than the data points, and as one result relies on multiple sources of evidence, with data needing to converge in a triangulation fashion, and as another result benefits from prior development of theoretical propositions to guide data collection and analysis” (Yin, 2009, p18).

For this thesis, the case study strategy follows the logic of design and data collection techniques proposed (Yin, 2009) but differs in the ‘prior development of theoretical propositions and hence the method of data analysis’, in other words this research design has no preconceptions about the data with the analysis following what emerges from the data.

Replicated multiple case studies typically adopt a positivist approach to develop testable hypothesis to contribute novel insight by examining differences, the ‘Eisenhardt Method’ (Eisenhardt, 1989, Langley and Abdallah, 2011). Although this exact approach is not used in the current study, there are similarities, not least in using multiple cases and multiple participants within cases. In essence, therefore, the current study amounts to a replicated interview method within and across the six retail company cases, with the retail company forming the unit of analysis. This particular approach aims to capture what retail board members actually consider important in terms of performance metrics and within that the specific sub-group of non-financial (termed by this researcher as ‘retail performance’) metrics and gain an understanding of which ones they commonly use. The underlying research strategy (see Figure 4.2 for terms used and

112

their meaning in this thesis) is a particular style of an ‘informed’ grounded theory approach (Glaser and Strauss, 1967, Gioia and Chittipeddi, 1991, Thornberg, 2012). Within grounded theory there is a particular style of analysis that aims to give rigour to the analytical process called ‘dimensional’ analysis (Schatzman, 1991), this is what has been used for this thesis. There is a detailed description, section 4.9, of the use of ‘dimensional’ analysis (Schatzman, 1991, Robrecht, 1995) as this is the basis upon which the analysis develops the concepts. Using an explanatory matrix, constant and interactive comparison of data collected is undertaken to develop salient dimensions that can form multiple novel concepts or viewpoints. One of these viewpoints then becomes the dominant perspective from the research as it more closely represents the understanding of the research problem (Kools et al., 1996). Hence the ‘Eisenhardt Method’ (Eisenhardt, 1989, Langley and Abdallah, 2011), aspects of which Yin (2009) refers to, is not used because the aim of the research objectives is to first understand what retail boards actually do. The better strategy for this thesis is termed as the ‘Gioia method’ (Langley, 1999, Langley and Abdallah, 2011), involving grounded theory research, as presented in their seminal paper on ‘sensemaking and sensegiving’ (Gioia and Chittipeddi, 1991). Therefore, this thesis has an inductive approach which is based on a multiple case study strategy with a qualitative method using a mix of tools and techniques such as replicated semi-structured interviews facilitated by a briefing preparatory survey and a template coda to be adapted and updated from the data collected (see Figure 4.1). All the analysis being done through the use of Qualitative Data Analysis Software (QDAS) i.e. NVivo 10. This analysis tool allows for all data sources to be uploaded for analysis. The reason for applying these tools and techniques was to create a degree of consistency in method within and across the retail case studies.

4.5.4.2. Case selection

Rather than undertaking just one case study and to overcome some of the criticisms of qualitative research, the style adopted is a multiple retailer case study strategy covering six retail companies. Yin (2009) refers to the robustness of replication logic in this approach as follows: “The ability to conduct 6 to 10 case studies, arranged effectively within a multiple- case design, is analogous to the ability to conduct 6 to 10 experiments on related topics; a few cases (2 or 3) would be literal replications, whereas a few others (4 to 6) might be designed to pursue two different patterns of theoretical replications (Yin, 2009, p54).

Each carefully selected case was analysed using a replicated method (i.e. using the same tools and techniques) to seek out similarity in findings, whereby the researcher sought commonality in the metrics actually used by the retail board directors. This design differs from other replicated multiple case studies where a theoretical framework is created first and then tested across the case studies (Eisenhardt, 1989, Eisenhardt and Graebner, 2007). There is an element of predetermination and a more deductive method in this latter approach of defining a framework and then checking for its existence. Whereas this thesis adopted a more grounded

113

and inductive approach in a search for whatever the participants share when they focus on the subject of metrics. Even though an initial pre-defined template coda was constructed, its purpose was to speed the data analysis, recognising that the template would change and evolve as the coding exercise progressed representing an ‘informed grounded theory’ strategy (Thornberg, 2012, Charmaz, 2014). This is somewhat different to a ‘pure' grounded theory strategy which would suggest no previous preparation or even literature review just direct interviews and observations to explore what emerges (Charmaz, 2003, Charmaz, 2014). This replication of the interviews with participants within the unit of study and across units (i.e. replicating the same process with different board directors within the retail case companies and across different retail companies), allowed for triangulation, adding weight to the validity of the outcomes in data collected and data analysed.

Yin (2009) recommends six case studies as a good number for undertaking a replicated multiple case study approach as shown in the quote above and this guidance has been followed. The cases were selected to cover: on-line only, store based only and ‘clicks and bricks’ retailers; old and established as well as new retailers; products and service retailers, capturing as much variety of retail within a small selection of retail case studies. The retail case companies are presented in section 4.5.4.3. The case or unit of analysis was the retail company and four revelatory (i.e. selected for their specific characteristics that may reveal specific features - e.g. an on-line-only retailer) cases were selected from the original survey sample (Yin (2009) suggesting 2 or 3) plus one failed retailer and one in distress to make up the six case studies (Yin (2009), suggesting 2 or 3) so the make up for this research design is four plus two, retail companies. The retailer in distress actually went into administration during the primary data collection period resulting in what could be described as four successful and two failed retailers.

4.5.4.3. The six case study retail companies

The retailer case studies have been referred to as A, B, C, D, E and F with a synopsis below:  Retailer A: a relatively new company established in 2006 operating exclusively on-line selling branded ‘Premium & English Heritage’ clothing, footwear and accessories. The founding couple have recently raised external funding for expansion and added two board members who are also minority shareholders. The company is (still) privately owned.  Retailer B: a long established (1849) and iconic premium luxury retail department store that describes itself as selling the finest merchandise, ‘everything to everyone everywhere’. The primary source of income is from one site and there is an on-line presence which accounts for less than 2% of the business. The company has been publicly listed and ownership has changed over the years. It is currently privately owned and was acquired in 2006 by a sovereign wealth fund.  Retailer C: a long established (1890) ‘home shopping’ retail group that has grown through public listing, mergers and acquisitions and changed the business model over time. The company currently operates exclusively on-line via numerous brand names and is privately owned.  Retailer D: a long established (1869) ‘service’ focused retail chain with a nationwide network of branches (Total Group 1,102). Owned by a family trust and run by the third generation as a public

114

company. Currently expanding the business model to include other ‘service’ offerings within the retail fascia and through adding different retail brands.  Retailer E: a retailer established in 1996, whose activity was the retail distribution of mobile telephone connections, handsets and associated products. It operated both a nationwide branch network (717 stores) and on-line business. It had been privately owned, was a public company and most recently owned by a private equity group. It went into administration in 2014.  Retailer F: a long established, in 1909 in the UK, ‘variety store’ retailer that has been privately owned, listed then acquired by other listed entities and then re-listed. Has operated a nationwide branch network (818 stores) and an on-line business. Went into administration at the beginning of the most recent recession in 2008.

4.5.4.4. The process

The primary purpose of the research is the examination of metrics used to support retail board members in understanding their company’s performance. So the aim was to capture and gain insight from 17 interviews with Board members, see Table 4.4 below, of the six retail companies. The number of interviews by company and participant are given below and it should be noted that the interviews were framed on the basis of what metrics the retail company uses and of these, what the participant favoured and were actively discussed at the board meeting. Interviewees were board members from different functional backgrounds. In addition, individual retail expert interviews were also conducted to aid validation and triangulation with:  A Global Head of Retail at a ‘big four’ professional services firm;  The President of the Institute of Chartered Accountants;  The President of the Institute of Insolvency Practitioners;  The Head of Mergers and Acquisitions for Retail at a ‘big four ‘professional services firm.  The Head of Consumer Insight from a ‘big four’ consulting practice.

In total they gave 14 hours and 8 minutes of comments plus various corporate documents.

Qualitative research was used as it captures temporally evolving phenomena in rich detail. The data comprised:  Surveys and other secondary research i.e. ratio analysis in the form of failure prediction models discussed earlier in Chapter 3 as a part of the phase one planning.  In vivo: meetings; semi-structured interviews in real time (with a reflection on history – for failed retailers) supported by survey questionnaires;  Artefacts: management reports provided by the retail company; published information – ARA sourced from FAME database, press and media coverage from Retail Week and Financial Times alert service; and other company provided documents.  In vivo: meetings where unstructured interviews were conducted with individuals that represented a panel of experts.

115

Table 4.4: Summary of retail company and participant interviews

Participant Retail company A B C D E F Total Chairman 1 1 Managing Director 1 2 3 Marketing Director 1 1 Finance Director 1 1 2 1 1 6 Strategy Director 1 1 Operations Director 1 1 Customer Director 2 2 Risk Assurance Director 1 1 Financial Controller 1 1 Total 4 5 3 2 1 2 17

A copy of the agenda for the meetings is presented in Appendix 5. The detailed planned procedures and outputs from the research steps is shown in Figure 4.5. Although the original plan was a three phased approach, there was an interim step between phases one and two to ensure consistency of approach to all the case studies and interview participants.

4.5.5. Phase three

The data has been analysed using a QDAS toolset NVivo 10 (Bazeley and Jackson, 2013) as a repository for all information, including the transcripts from the recorded interviews. When discussing the use of the NVivo toolset Bazeley and Jackson (2013) note: “For some methods, most notably grounded theory, your initial analysis will involve detailed, slow, reflective exploration of early texts – doing line-by-line coding, reading between the lines, identifying concepts and thinking about all of each concept’s possible meanings as a way of ‘breaking open’ the text” (Bazeley and Jackson, 2013, p72).

All the data, residing in NVivo as a repository, was developed and structured by progressive abstraction starting with first order ‘open’ coding of interview comments (paragraphs) and assigning ‘terms’ such as: ‘net promoter score’; guest shop; mystery shopper etc., where such specific metrics or performance indicators are mentioned. These are the participant ‘voiced’ concepts, in line with a grounded theory approach (Strauss and Corbin, 1994). Then the first order analysis was refined to second order themes using axial coding (Kendall, 1999) at a higher level of abstraction such as: customer (i.e. the customer-related metrics ‘noted in bold’ above were aggregated under this term); sales; margin; people etc., these terms are all explained in Chapter 5. These then represent the themes i.e. FAs of the first order analysis. Figure 4.6 shows a transcript extract and initial reading with Figure 4.7 giving the process of metrics’ assimilation, moving from a abundance of metrics into large buckets (categories) two of which overflowed and required smaller groups which were then aggregated to FAs. This aims to represent how NVivo was used. The figures have been simplified for illustration as the process was complex and non-linear.

116

Figure 4.6: Transcript excerpt – a first reading

Transcript excerpt Understanding the participant’s voice

“The fundamental challenge with [F] Retail was actually-- Commenting on poor Business Model implying the retail offer was not probably, it was never a phenomenally good business model7 full-stop for an awful long time9 - probably. Although it was right “for an awful long time”. Goes on to explain another flaw in the high up in the FTSE7,8 at one point in all those things, but I business model – organisation structure. think the start of the end9 was when [Group Top Co]7,8 demerged which was - I can't remember when - late '90s9. Organisational structure changes imposed - not working for F Retail and What happened then was the freeholds5,6 got sold. I can't explaining the structure of Group Top Co imposed changes: Cash remember what the name of that company7 was, it was well stripped out and rentals burden imposed i.e. Business Model in the form 9 before I got there . But basically the crux of it was through of organisational structure creating the “massive problem”. 7,8 7,8 different structures et cetera. [Group Top Co] took the cash1 and [F] Retail took the rent2 - which was never going to Voicing when the first ‘tipping point’ took root – “start of the end” work for [F] Retail really. Clearly the rentals2 would have been a big drain on cash1 - which made individual store P&Ls3 suddenly change. Now there is a view that each Time references to past – FTSE listing; late 90s changes – demerged; store5,6 should be self-sufficient3 and therefore we know, you cash stripped out “before I got there” should be charging2 into the branch accounts3, a commercial rent2. And I accept that view - not sure I would do it - but let's Analogy with another retail company M+S where store estates are accept that view. But if you would be doing a market rent2 “winding down” adjustment3 for properties5,6 like that, or if you were doing an 2 1 7,8 internal rent , the cash is still within the group . So if Management challenges and problems explained through Metrics, anything horribly went wrong, you'd still got a cash1 situation themes and tools: that says, 'Actually I can still spend3,4 money1 on that store5,6 et cetera. This case, it was a real rent2, there was no cash1, The paragraph using NVivo can be and so what that meant is, we couldn't invest4 in the store 1. Cash coded to all eight potential themes chain6. A rental2 burden3 and not only that, but the rents2 that 2. Rents and metrics without losing any of were put in place, were called Horizon Rent2 deals. I think we 3. Branch accounts – store P&Ls the participant’s story telling called them. I think for 800 locations5 there are over 600 4. Investment enabling later reflection and landlords5,6 - very difficult to negotiate, and it was a massive 5. Number of stores 6 subsequent: problem, and the store estate just was winding down and 6. Store property estate st 1 Order Analysis; and down and down in terms of quality. There are other retailers 7. Business Model today that I look at that are very similar - Marks & Spencer's nd 8. Organisation structure 2 Order Analysis being one. And so the start of the rot was then9, if not before.” 9. Time / point in time

Note: The superscript references are to the terms numbered 1-9 in the second column

117

Figure 4.7: Process of metrics’ assimilation

Customer numbers (footfall); Customer numbers (clicks); Customer conversion i.e.

average weekly transactions; Net Promoter Score; customer satisfaction index; customer complaints; ‘Rant & Rave’; mystery shopper i.e. guest shops or DVD audits; customer feedback card; Trust Pilot Review; Google Analytics; Loyalty programme;

CLV model, Total Sales i.e. Turnover or Income; Income by geography, branch (store), brand, department, category, segment, SKU. Sales by year, month (four weekly months for some retailers), week, day. Sales including VAT, excluding VAT. LfL Sales. Sales v prior period. Staff numbers (FTE); Staff Turnover percentage; Staff Retention percentage; Staff engagement e.g. Happy Index, feedback and discussion forum / portal i.e. ‘yammer site’; training completions Gross profit margin; Operating profit margin; Net profit margin; Buying gross margin; Commercial gross margin; Buyers net margin; Contribution by branch; Margin components, waste, markdown, stock loss, shrinkage, Intake margin; Exit margin, Margin by day, brand, product, own v concession Daily cash balance; daily cash flow; cash performance, cash generated and cash drivers; cash by, period v prior period v prior year v forecast ; Borrowings i.e. debt owed; bad debt write-off; Debt analysis i.e. leverage or gearing ratio; Metrics Bond; Borrowing headroom; interest cover; Credit Insurance Sales density Non 2 sales or profit £ per ft brand contribution; store contribution; number of stores; new store, daily sales, contribution,Financial profits; store size; store rents; long leases; sale and buy Financial backs; store network coverage; Owned stock holding by number by value ; concession stock holding; Stock by, department, brand, category, SKU by channel, on-line, mobile, tablet, new v repeat v organic v PPC (pay per click) v SEO (search engine optimisation); Google analytics dashboard Brands for resale numbers; Brand mix; corporate brand values, ROCE (return on capital employed promotions; Data Protection Act Customer, consumers, Buckets stock turnover i.e. number of days or percentage; stock sell-through; AgedSales, Stock Profit, by, season, Return price, consumer discount; behaviour, returns percentage; footfall, Mark down rate; carry forward lines; stock accuracy; known stock loss, unkownTime stock loss; stockon availability Investmen i.e.t, out of stock Store offer; Category authority; product attribute; churn, disloyalty, brand mix; product mix; seasonal product; open to buy; supplier income i.e.Pensions, rebates; Gross supplier relationships;abandonment, Private bounce, v public ; Group structure by division; Margin, Fixed Assets, conversion, CLV, Customer Ownership by equity house Cash, Cash Flow, Numbers sovereign wealth fund; family; public listing; culture and control; stores Debtand on -line; new business by segments by geography, Sales growth by Distress signs, Beta Customer Groups segment; CAGR (compound annual growth rate); profit Creditgrowth, Crunch Pay to sales ratio or wage percentage; overhead costs; operations cost Sales growth; staff , numbers, Market Share and market shareReputation by segment, Damage Availability; Service level People Margin Quality, Self-Audit Debt Cash Distribution centre cost, Unit handling cost; Unit delivery cost; delivery time windows Named Propertycompetitors e.g. Net –a Porter, Other Retail Sins, Culture Stock control Carphone Warehouse, Express, Gifts, Sales by channel, on-line, mobile, Productstablet, new and v Buyingrepeat v organic v PPC (pay per click) v Competition Risk and regulation SEO (search engine optimisation); Google analytics dashboard Brands for resale numbers; Brand mix; corporate brand Commission Business Model Return on investment values, ROCE (return on capital employed Competition Digital Focus ROI (return on investment) by channel, by capital project spend Profittype, by store; payback Growth period; IRR (internal rate of return) by green, amber, red Supply chain Areas Competition Brands Net profit; EBITDA (earnings before interest, tax, depreciation and amortisation); Profit beforeCost base Tax Regulatory risk such as MMC scrutiny, FCA compliance, EU directivesMarket e.g. share weee dynamic risk management process; HealthCompany & Safety beta not used TCF (treating customers fairly); Risk registers; Data Protection Act 118

4.6. Applying the grounded theory research strategy

The first step in applying the grounded theory research strategy is to upload all the information into NVivo and then start the process of coding. The coding process applies to all the documents and a sample of a transcript (paragraph) coding for illustrating the process is presented in Figure 4.6. It can be seen that the former (the retail company no longer exists) retail board member recalls an important issue that in his view started ‘F Retail’ on the road to failure i.e. a flawed business model. The Figure attempts to show the first reading of the transcript and initial abstraction of metric information as well as the ability to code the entire paragraph to multiple reference points in NVivo.

This then allows for later reflection when considering each reference point for any further meanings and also makes the full excerpt of the transcript available for first order analysis and second order analysis and these steps are explained in more detail later. Figure 4.7 shows the next part of the process taking the multitude of metrics gleaned from the data and first putting them into four buckets for metric collection: financial; non-financial; time; and other. The financial and non-financial buckets were uninformative and overflowed so required segmenting into smaller groups for further analysis. There is then an iteration process as the metrics in these smaller groups are aggregated into meaningful FAs.

It can be seen in Figure 4.6 that some metrics such as net profit margin can be coded into more than one bucket and that the FAs will have multiple metrics attached to them. It is also a moot point whether or not, for example, The Data Protection Act (DPA) is a metric per se. However, in the context of the interview, DPA compliance failures were considered sufficiently important to merit comment. Hence the coding process did not ignore the voice of the participants and this was put into the bucket termed ‘other’, then grouped under a broad category of ‘risk’ and finally appears in the FA of ‘risk and regulation’. So the aforementioned traces the path of one metric item from interview comment to FA.

4.7. First order analysis

Reference was made to individual case narratives (used as working papers) prepared by the researcher as each retailer case study was conducted. These had been drafted to support the analysis as well as the questionnaires and documents provided by the retailer. Essentially an iterative process of interviewee comment comparison, within each unit of analysis i.e. the retailer company as a case study and between cases, until the themes emerged. They were aggregated until a manageable set were reached. “[from the long list] reduce the variety to a shortlist of about 20 that provides the Exec of a large company with a broad enough picture of market performance.” (Ambler, 2003, p119).

119

Typically, when working with larger quantitative datasets, a cluster analysis technique can be used to get to a manageable set. Guidance was taken from the NVivo training workshops and as the quote above suggests that 20 is a good number (Ambler, 2003). In spite of following guidance, getting to 20 can still be seen as a subjective decision, even if an intuitive form of cluster analysis was used, shown in Figure 4.7, whereby agglomeration of the data occurred using QDAS. The NVivo toolset allows for iteration, aggregation and changing the relationships i.e. parent to child or child to parent. The method of open and axial coding (Kendall, 1999) using NVivo software aimed for 20 as a good manageable number. As noted in the quote below 20 also became a point at which conceptual saturation was reached. “The purpose of grounded theory methodology is to generate theory through the process of constant comparison. Data are analysed as they are collected, through the process of coding […] As similarities and differences in the codes are conceptualized, a coding scheme reflecting theoretical constructs is refined by clustering codes together to make categories. Conceptual saturation is reached when no new categories are generated from the open codes.” (Kendall, 1999, p746).

There has been a number of coding movements back and forth to make sense of the comments, metrics and themes. To some extent there is an element of knowledge and experience of the researcher in divining the information and a risk of researcher bias but that is also a part of the creative moment of bringing together all the rich information to gain insight (Langley, 1999) as noted in the quote below. “As a story or problem is revealed to the researcher, the dimensions of the problem have no form until the researcher takes a perspective, or viewpoint, on the information” (Robrecht, 1995, p174).

The individual metrics, having been tagged with the FAs, moved into the various FAs. The data is presented in Table 5.1 of Chapter 5. The presentation of the findings pushes the idea of giving voice to the participants (Gioia et al., 2013) as a part of the qualitative analysis and reporting, by liberally using their words in quotes to illustrate the themes. This not only gives them a voice but also acts as confirmatory evidence that the theme title, e.g. customer, is actually what they talked about. Using the multiple voices from within and across the case studies adds weight to the validity of the thematic analysis.

The first order analysis of FAs presented in Chapter 5 is the first of the dimensions (see section 4.9 explanatory matrix) but it is recognised that with a different lens, a different researcher may come to a different viewpoint as noted below. “the researcher strives for the capacity to entertain multiple perspectives on a given situation. Each perspective gives a different configuration to the data; it tells a different story. The chosen perspective becomes the theme that configures the story, making the phenomenon understandable” (Robrecht, 1995, p174).

This vulnerability of the method may limit extrapolation and generalisation of the findings. Nevertheless, what this grounded theory strategy has revealed, directly from the data, is what these retail boards actually consider sufficiently important to merit their time and attention. The natural process (Schatzman, 1991) of continuous comparative assessment of the data did

120

deliver 20 designations that have been considered salient and termed FAs in this thematic analysis. This natural process is, of course, part and parcel of the researcher background and understood to be a strength in the grounded theory approach (Strauss and Corbin, 1994).

The next step was to devise or select a mechanism for summarising the metrics to a vital few that can be used for gaining an understanding of performance against each FA identified. Again this requires judgement and experience and is a final creative step in the qualitative process. Comparison with academic literature, secondary information published by retailers and metrics mentioned by the case study participants have been used to give sense to the second order analysis and framework abstraction given in Chapter 6 (Gioia and Chittipeddi, 1991, Gioia et al., 2013). This process is explained in more detail below.

4.8. Second order analysis

The second order analysis takes the next logical step of understanding, using the dominant themes expressed by the participants and gleaned from the overall data i.e. FAs, the researcher attempts to derive a conceptual framework to put the retail case study stories into a more theoretical perspective by means of this second order analysis. This method of conceptualisation, being very difficult to convey (Kools et al., 1996), is explained below. “One of the dilemmas in explicating the process of grounded theory method is the intrinsic difficulty in describing the constant, dynamic interactional relationship between the researcher and the data.” (Kools et al, 1996, p317).

Qualitative research is often criticised for lacking scientific rigor in its process. In this thesis a dimensional analysis (Schatzman, 1991) strategy has been used to inject rigour into the analysis process. To help explain the linkages between the data and the development of a number of concepts, dimensional analysis is described and illustrated below. Using this method, five concepts emerge. One of the five concepts has ultimately percolated to the researcher’s primary viewpoint that, in the context of performance metrics, retailers are on a journey towards becoming ‘trust intelligent’ through transparency of disclosing non-financial metrics and is discussed in Chapter 6.

4.9. Application of an explanatory matrix

4.9.1. Dimensional analysis and the explanatory matrix

As with most qualitative research, the data gathered was complex and made more so given that this research was conducted over a number of years with different board members across different retailer types. A critical aspect of dimensional analysis of data is the reconstruction of multiple components of complex phenomena. This involves giving labels to data e.g. customer,

121

business model etc. and then expanding these into various attributes which include dimensions and their properties. The aim of this process is to get the data conceptualised in a multitude of ways by identifying the parts of the whole (Schatzman, 1991), such as the 20 FAs, and providing a broader view of the complexity. “As a methodological strategy, this data expansion serves to illuminate the plethora of dimensions and corresponding sets of properties unique to any phenomenon […] data are collected and scrutinized until a “critical mass” of dimensions is assembled, which represent emerging pathways that possess some explanatory power” (Kools et al, 1996, p317).

The first order analysis using axial coding, gave voice to the participants as recorded in the themes identified in Chapter 5. These themes were then subject to the first dimension of second order analysis which rendered the 20 FAs using the explanatory matrix as shown in Figure 4.8. The centre piece of the matrix is a sequence in method (‘Context’, ‘Conditions’, ‘Processes’, ‘Consequences’) to interrogate the data (Designations i.e. terminology ascribed upon the data groups) via simultaneous and interactive analysis to arrive at ‘Dimensions’ and their ‘Properties’ (Robrecht, 1995). So designating the data allowed for the first dimension to emerge as the 20 FAs. From these dimensions one perspective becomes the predominant viewpoint. These dimensions can be seen as ‘higher order’ findings as they have been abstracted and interpreted from the first order analysis.

Figure 4.8: Explanatory matrix for dimensional analysis

Perspective

Dimensions

Context Conditions Processes Consequences

Designations

DATA

Adapted from Kools et al, 1996, p318

122

4.9.2. Explaining the procedural steps

Given below is the method adopted in a step-by-step explanation of each term in Figure 4.8 and relating them to the detailed analytics used. It should be noted that the simplification of the explanation below may imply a linear process but it was not linear, as mentioned in the quote below, it was, iterative, simultaneous and interactive. “Analysis does not usually proceed according to a prescribed process with distinctive phases or stages.” (Kools et al, 1996, p317).

Each of the matrix components are explained below:  Data: all the complex data, including transcripts of the interviews, was loaded into NVivo as a comprehensive repository.  Designations: these represent the labelling of the data into coherent and manageable categories.  Context: This sets the boundaries of the research and were focused on getting to an understanding of performance metrics actually used by retail boards. It is important to keep front of mind the origin of the research question that understanding a retailers trading performance in the public domain is difficult to discern hence the search for non- financial (retail performance) metrics that may improve understanding.  Conditions: these are the most salient aspects of interpretation that ‘cluster up’ to form the more important dimensions. There were five salient dimensions as noted below.  Dimensions: the designations were categorised into five dimensions and these are discussed in Chapter 6.  Processes: These reflect the intended or unintended actions that are framed by the conditions. In the context of this research the impact is on what metrics are actually used internally and what attitude is taken to making disclosures externally.  Consequences: These are the outcomes which show that retailers are not as transparent as they could be about their performance and some, particularly privately owned companies, choose to be minimalist in their disclosure. This explanatory matrix is the central part of ensuring rigor in the analytical process for the grounded theory strategy.

4.10. Inductive and interpretive presentation of findings

The findings from the first order analysis are presented in Chapter 5 and from the second order analysis are presented in Chapter 6. The inductive approach and the specific grounded theory strategy adopted, lends itself to a different way of presenting the findings. Instead of using the positivist style of: Results; Discussion; and Conclusion, where clear demarcations are the order of the day and objectivity and rationale rule. The findings are acknowledged as subjective and hence presented in the order of abstraction with the chapter headings referenced as Chapter 5:

123

First Order Analysis and Chapter 6: Second Order Analysis. The point here is that the findings, discussion and to some extent method steps are inter-linked and have occurred through constant comparison and conceptualisation so neat linear reporting dilutes the richness of the story.

4.11. Summary

In summary, this thesis is based on a qualitative research method supported by a mix of tools and techniques (as shown in Figure 4.1) that comprise elements of quantitative data collection and analysis. For the most part it is a multiple case study strategy searching for common metrics used by retail companies, as explained by retail board members’ replicated semi- structured interviews, to manage trading performance. An inductive approach with implicit interpretivist bias as a best means for understanding a socially constructed management phenomena i.e. retail metrics used within retail organisations. Given that these metrics have been explicated then they, as shown in Chapters 5 and 6, present a better way of understanding a retailer’s trading performance than has been hitherto available.

124

Chapter 5: FIRST ORDER ANALYSIS

5.1. Introduction

This chapter outlines the findings from the retail company case studies undertaken. Six retail company case studies were completed: two failed retailers; and four that can be described as successful. For conformity, and ease of writing, all case study interview comments and information have been treated as confidential.

What is presented first is the retail performance metrics the retailers actually use. Here the abstraction (see figure 4.6 in Chapter 4, which shows the process of reading the data and analysing it) sought to identify commonalities amongst the cases given that most retail organisations use metrics of some sort (Ambler, 2000, McGoldrick, 2002, Farris et al., 2006). What can be seen (an extract given in Table 5.1) is a abundance of metrics’ data. A mechanism is therefore employed to sift these retail performance metrics, thereby creating clusters annotated good, better and best for easier comprehension.

The next section in the chapter presents a thematic analysis of interview data from the cases. From this process retail performance metrics considered important by board members have been aggregated into ‘Focus Areas of retail performance measurement’ (FAs). Each FA is briefly discussed and interview quotes used to evidence this discussion and give voice to case participants.

5.2. The metrics overview

In the previous chapter, Figure 4.7 illustrated the way the retail performance metrics were assimilated. Figure 5.1 presents the retail performance metrics actually mentioned by interview participants and contained in company provided information. Even though the retail performance metrics listed are all relevant, as they have been referred to by retail board members as worthy of their attention, a mechanism for aggregation or clustering is arguably required to make them more easily comprehensible.

5.3. Sifting the metrics

5.3.1. Introduction

When there is a abundance of metrics in use, getting to an understanding of which metrics are more important and most useful should be imperative for retail managers, as they only have limited time available. Consistency in how retailer performance metrics are calculated is also important,

125

Figure 5.1: The abundance of metrics’ data extract

Customer numbers (footfall); Customer numbers (clicks); Customer conversion i.e. average weekly transactions; Net Promoter Score; customer satisfaction index; customer complaints; ‘Rant &

Rave’; mystery shopper i.e. guest shops or DVD audits; customer feedback card; Trust Pilot Review; Google Analytics; Loyalty programme; CLV model, Total Sales i.e. Turnover or Income;

Income by geography, branch (store), brand, department, category, segment, SKU. Sales by year, month (four weekly months for some retailers), week, day. Sales including VAT, excluding VAT.

LfL Sales. Sales v prior period. Staff numbers (FTE); Staff Turnover percentage; Staff Retention percentage; Staff engagement e.g. Happy Index, feedback and discussion forum / portal i.e. ‘yammer site’; training completions Gross profit margin; Operating profit margin; Net profit margin; Buying gross margin; Commercial gross margin; Buyers net margin; Contribution by branch; Margin components, waste, markdown, stock loss, shrinkage, Intake margin; Exit margin, Margin by, day, brand, product, own v concession, Daily cash balance; daily cash flow; cash performance i.e. cash generated and cash drivers; cash by, period v prior period v prior year v forecast; Borrowings i.e. debt owed; bad debt write-off; Debt analysis i.e. leverage or gearing ratio;

2 Bond; Borrowing headroom; interest cover; Credit Insurance Sales density i.e. sales or profit, £ per ft ; brand contribution; store contribution; number of stores; new store, daily sales,

contribution, profits; store size; store rents; long leases; sale and buy backs; store network coverage Owned stock holding by number by value; concession stock holding; Stock by, department,

brand, category, SKU; stock turnover i.e. number of days or percentage; stock sell-through; Aged Stock by, season, price, discount; returns percentage; Mark down rate; carry forward lines;

stock accuracy; known stock loss, unkown stock loss; stock availability i.e. out of stock Store offer; Category authority; product attribute; promotions; brand mix; product mix; seasonal product;

open to buy; supplier income i.e. rebates; supplier relationships; Private v public; Group structure by division; Ownership by equity house, sovereign wealth fund; family; public listing; culture and

control; stores and on-line; new business by segments by geography, Sales growth by segment; CAGR (compound annual growth rate); profit growth, Pay to sales ratio or wage percentage;

overhead costs; operations cost growth; staff numbers, Market Share and market share by segment, Availability; Service level, Distribution centre cost, Unit handling cost; Unit delivery cost;

delivery time windows Named competitors e.g. Net –a Porter, Carphone Warehouse, Express, Gifts, Sales by channel, on-line, mobile, tablet, new v repeat v organic v PPC (pay per click) v SEO

(search engine optimisation); Google analytics dashboard Brands for resale numbers; Brand mix; corporate brand values, ROCE (return on capital employed); ROI (return on investment) by

channel, by capital project spend type, by store; payback period; IRR (internal rate of return) by green, amber, red, Net profit; EBITDA (earnings before interest, tax, depreciation and amortisation); Profit before Tax Regulatory risk such as MMC scrutiny, FCA compliance, EU directives e.g. weee, Data Protection Act, TCF (treating customers fairly); Risk registers; dynamic risk management process; Health & Safety Note: Company beta not used

126

for retail managers and those interested in company and broader sector performance, because it allows for comparability over time both within and between retailers.

Yet this is not always the case. For example, a metric that is consistently used by retailers is LfL sales, yet it is inconsistently calculated, as discussed in Chapter 3, and inconsistently presented (i.e. retailers may present it in the half year report but not in the next quarterly trading statement), raising questions about its quality and credibility as a measure of performance (Deloitte, 2006, ICAEW, 2013, Smith, 2015). Similar reservations about other metrics are revealed when they are critically evaluated. An example is the metric for sales density i.e. sales ft2. Retailers calculate the square footage of this metric using different methods (Deloitte, 2006). Some exclude store rooms, service points such as tills, lifts and travellators and even car parks, whereas other retailers may include these in their footage calculation. Similar differences apply to the sales part of the sales ft2 figure, where: some retailers use a VAT inclusive figure whilst others do not; some include the gross sale whereas others net off discount vouchers and loyalty points; and some include the on-line sales picked at store (Deloitte, 2006). In summary, as captured in the quote below, there is a lack of consistency in the definition and calculation method for most of the metrics used in the retail industry. “Other sectors have found it possible to create a standardised basis for calculation for their most significant key performance indicators. If greater transparency could be achieved as a first step, it might open the way to achieving greater consistency as well. A standard method of calculating like-for-like sales would create a far more reliable tool for decision making, both internally and for external stakeholders” (ICAEW, 2013, p11)

5.3.2. A sifting matrix

Although many performance metrics are used and have been documented (Walters, 1977, Walters and White, 1987, Fernie and Sparks, 1999, Ambler, 2000, McGoldrick, 2002, Farris et al., 2006, Fernie et al., 2010c), very little attention has been given to their innate quality and resilience. Quality and resilience are subjective terms and consequently difficult to define objectively. Nevertheless, properties or attributes of quality and resilience may be noted and these are discussed below. A ‘sifting matrix’ developed for the purposes of this thesis, and presented in Figure 5.2, can act as a framework for clustering the retail performance metrics used by retailers into quadrants of good quality vs. poor quality, and high resilience vs. low resilience. Using a traffic light warning principle, it can be seen in Figure 5.2 that two quadrants (top left and bottom right) are colour coded amber, indicating potential concerns with either the resilience or quality of metrics respectively. One quadrant (bottom left) is colour coded red, emphasising issues with both metric quality and resilience combined. The top right quadrant is colour coded green, indicating metrics of both higher quality and resilience. Figure 5.3 focuses in on this green sector of the matrix, emphasising that the very best performance metrics with the greatest utility to retailers may be those with the highest levels of quality and resilience combined. The dotted line is for illustration purposes to show how the grouping of retail performance metrics into clusters from good to better to best can aid comprehension about the

127

quality and resilience of these retail performance metrics. Both Figures 5.2 and 5.3 represent conceptual proposals or matrices, but they are a useful means of starting to think about how retail performance metrics might be classified and organised, and represent a commencement point for any debate about retail performance metric utility. Figure 5.2: A sifting matrix for retail performance metrics

Quality High

Amber Green

High Resilience Low

Red Amber

Low

Figure 5.3: Sifting matrix - the focus quadrant

Quality

High

Resilience High

128

Not only is quality subjective, as noted in the quote, but the same applies to resilience shown on the horizontal axis. “it must, however, be emphasised that no definitive set of quality attributes and weightings of those attributes exist, since quality is subjective and context- dependent” (Beattie et al, 2004, p230).

Accepting the subjective nature and context dependency noted above in the quote, assessing retail performance metrics in terms of their quality could be done by reference to attributes previously applied to determining quality information. These attributes were originally defined for accounting information (AAA, 1974) and are now enshrined as fundamental in this area, as well as being applied to the provision of information in for example information technology (Miller, 1996). Thus, according to the American Accounting Association (AAA), the attributes of quality information are eightfold, as detailed in the first eight rows of Table 5.1. Row 9 of Table 5.1 also includes an additional attribute of ‘cost-efficient’, on the basis that there is always a cost benefit trade-off in providing information (Lanzetta and Kanareff, 1962, Porter and Millar, 1985).

Table 5.1: Attributes of good quality information for retail performance metrics 1 Relevance / mutuality of objectives 2 Accuracy / precision / reliability 3 Consistency / comparability / uniformity 4 Verifiability / objectivity / neutrality / traceability 5 Aggregation 6 Flexibility / adaptability 7 Timeliness 8 Understandability / acceptability / motivation / fairness 9 Cost-efficient Adapted from American Accounting Association Report of the Committee on Concepts and Standards (AAA, 1974)

There is a danger of falling into a ‘positivistic trap’ of attempting to score or put ticks against the attributes listed in Table 5.1 to create a schema for assessing a metric’s quality. Indeed, the same point applies to resilience discussed below. Take for instance the attribute ‘relevance’ noted in the Table 5.1 above. This in itself is subjective, as something that is relevant to one person, e.g. earnings per share (EPS), may have a high degree of relevance to a shareholder but little relevance to an employee. So creating any form of scoring mechanism for attributes of quality in relation to information or indeed retail performance metrics is problematic. What is being suggested is that ‘informed judgement’ needs to be applied with cognisance of the attributes of good quality information to determine the assignment of a metric to a particular position or ‘cluster’ within the sifting matrix.

As noted, the sifting matrix is aimed at not just positioning a retail performance metric in terms of its quality, but also in respect of its resilience. These attributes, shown in Table 5.2, are essentially the properties that contribute to an understanding of resilience (Vaishnavi et al.,

129

2007, Gallopín, 2006) and can help determine, judgementally, whether or not resilience is low or high.

Table 5.2: Properties of resilience applicable to retail performance metrics 1 Robustness e.g. usefulness over time, constancy, stability 2 Calculation bases e.g. fixed assumptions v changing assumptions 3 Degree of independence v interdependency e.g. numerators & denominators 4 Use of complete data v samples 5 Simplicity in calculation 6 Rapidity , speed of availability e.g. dashboard type indicator 7 Adaptability .e.g. relevant to different retail models

Consequently, resilience can be considered in terms of aspects such as: usefulness over time; the calculation bases – can the underlying assumptions of the bases be changed to reflect a different number, where a ‘no’ means high and ‘yes’ means low; interdependency on other numbers e.g. numerators and denominators (i.e. ‘yes’ means low and ‘no’ means high); the use of approximations; the use of sample data i.e. incomplete populations; simplicity in calculation; and speed of availability.

Using these attributes as contributing properties to quality and resilience, the metrics may be sifted and clustered on the matrix presented, as shown purely for illustrative purposes in Figure 5.4, using the experience and judgement of the researcher. The clusters in the top right hand of the quadrant can be deemed as the ‘best’ retail performance metrics with those closer to the axes as ‘good’ and those in between as ‘better’. A different person may cluster the same retail performance metric differently based on their experience and retailer being assessed. The point of the matrix is it acts as a mechanism for starting the debate about the quality, resilience and essentially the use of retail performance metrics to convey trading performance. In other words, a discussion to be conducted in recognition of the matrix presented, rather than just accepting a retail performance metric as a given. In short, the matrix allows for a way of thinking about retail performance metrics and indeed, how they might compare with each other. Thus, when considering ‘net profit’ as a retail metric, it has an important role as making a profit is part and parcel of the mission of a retailer so could be considered high on quality. Whereas sales, as a metric, can be made at a loss so would be relatively lower than net profit on the quality axis. The opposite relationship applies when considering resilience as net profit is arrived at after multiple adjustments dependent on the application of management judgement and is harder to validate. Whereas sales typically come out of the EPoS data therefore is simpler and quicker to calculate and is also mapped on Figure 5.4 for illustration. The effectiveness of the sifting matrix is dependent on the judgement and experience of the user. Nevertheless, it does provide a framework for clustering of the metrics enabling management to assess the relative importance ranking of the metrics given that they are all important. For stakeholders, they can evaluate what is being presented when trading performance is reported.

130

Figure 5.4: Sifting illustration of 20 relatively ranked retail performance metrics

Quality

High Average weekly transactions’ spend

Average weekly Net profit Cash LfL sales transactions’ generated number Business Sales per ft 2 model Wage Operating profit percentage margin Staff Availability CAGR retention EPoS ROCE Gearing Sale s Market Stock share turnover Competition Brand values Supplier income

Regulatory investigations

Resilience High

Note: each of the retail performance metrics mapped in Figure 5.4 are discussed later in this chapter. Sales are show for illustration purposes.

5.4. Thematic analysis

There was broad consistency amongst the retailers about the FAs, see the prioritised list below. These FAs have been derived from a thematic analysis (King, 2006, Cassell and Symon, 2004) of case study interviews until a manageable number (i.e. 20) of FAs were aggregated or distilled as a first order analysis (Langley, 1999, Gioia et al., 2013). The FAs represent areas of retail performance management and reporting to which board members appear to apply their attention, direct their retailer’s resources and target performance metrics. The relative ranking of the FAs noted below whereby customer is at the top and growth at the bottom represents the quantity and variety of metrics found and this can be seen in Tables 5.7 and 5.8. However, even though there was evident ranked or prioritised consistency in the FAs discussed across interviewees, the metrics used to measure these demonstrated a tendency to vary depending on the retailer. For example customer as a FA, is measured and monitored in a abundance of

131

ways like: mystery shopper scores6; net promoter scores7; customer engagement survey8 results; google analytic9 metrics; CLV10; customer engagement11 metrics; and customer numbers (footfall12, clicks13 and transactions14), these terms are also discussed in subsequent sections. This prioritised list for FAs emerging from the case interview data, is discussed in more detail by FA and summarised against the metrics mentioned at the end of this Chapter in Tables 5.7 and 5.8.

 Customer  Sales  People  Margin  Cash  Debt  Property  Stock control  Buying and products  Business model  Risk and regulation  Profit  Return on investment  Brands  Digital  Competition  Supply chain  Market share  Cost base transformation  Growth

6 Commercially available agency provided service (or conducted in-house) based on pre-defined attributes (by the retailer or agency) to anonymously test instore customer service. 7 Commercially available agency operated customer feedback scoring process typically involving asking a customer how likely they would be to recommend the retailer to friends or family members and ranked against a peer group 8 Commercially available or in-house operated either online, mobile SMS or face to face survey with predefined questions to obtain customer views of the shopping experience. 9 Online analytics, paid for service of varying levels of analytical information provided by Google monitoring various aspects of online behaviour on the retailers website e.g. country of origin, page dwell time. 10 CLV is customer lifetime value defined in Chapter 2. 11 Customer engagement metrics refers to the reported information against a predefined target based on customer engagement survey result such as a customer satisfaction index this is commercially available or can be in-house operated. 12 Footfall is a way to estimate customer numbers entering or close to the store, usually provided by third party agencies based on sample populations of customer traffic flow. 13 Clicks represent customer numbers through the calculation of visitors to a retailer website this can be unique or returning customers. 14 Transaction count is a system generated number from the checkout tills that represents customers actually served.

132

5.5. Customer

The term ‘customer’ dominated the interviewee discussions and is presented as the first relatively ranked FA. Its prevalence is given by the quote below. “We have focused ourselves very much as a customer-centric organization, and therefore everything we do is in and around customer.” Managing Director, B.

Discussions centred on a number of metrics used for the FA of customer and these are examined below.

5.5.1. Customer numbers

Customer numbers, is a key metric and focus for all retailers and its importance is presented in the two quotes below: “a key measure that we challenge ourselves on, customer numbers, because in the end that’s the key driver of sales” Strategy Director, C.

“Customer numbers […] we talk about that in the trading meeting, management meetings and board meetings […] I report on customer numbers”, Marketing Director, A.

Monitoring customer numbers is easier if a business has an online presence as the individual ‘clicks’ are registered via tools such as Google analytics. As one interviewee explained: “[customer numbers] It's through Google Analytics […] it's an industry-wide used tool […] in terms of an indication and a like for-like comparison, it's useful to us.” Managing Director, A.

Conversely, not every retailer that has a physical store base has the ability to monitor customer numbers via accurate footfall counts, although many rely on sampling data from third party agencies such as CACI, LDC or Experien’s Hitwise division. A reluctance to rely on footfall as a pure metric of retail performance arguably relates to the potentially poor quality attributes of footfall data, particularly in terms of accuracy and precision as well as data completeness (see Tables 5.1 and 5.2 above). One interviewee pointed out the challenge here: “Customer numbers […] it's important. It's not something that we focus on. If you did floor clickers through this store, it would be completely random […] because there are people who just come in and look. So, our conversion rate is very small […] Because lots of people just come in, as a tourist attraction” Finance Director, B.

Because, some respondents considered footfall counts unnecessary or unreliable, they preferred to measure customer numbers through the ‘average weekly transaction numbers’ generated by the point of sale (‘PoS’) equipment: “Customer numbers, I look really at the average weekly transactions […] I'd be looking [at] your strike rate […] So that's the bit that I emphasized rather than on footfall” Operations Director, D.

In short, the ‘hard’ metric for customer numbers, in a physical and on-line retail space, is the average weekly transaction count number. Observers, including academics, unfamiliar with retail operations in detail, may sometimes confuse this term with customer conversion (Farris et

133

al., 2010). As a customer makes a purchase at the checkout, a customer count is registered in the system. These hourly, daily, weekly counts can be aggregated to deliver this ‘hard’ metric. Whereas customer conversion, is really a measure of shoppers entering a store or web-site and then making a purchase so is expressed as a percentage. The ‘transaction count number’ represents the customers actually served. This is frequently used by retailers to determine the number of customers they serve because the customer has actually purchased rather than just ‘window shopped’ whereas estimating footfall through sampling is not as reliable. It can be seen that this metric (transaction count number) has a number of quality and resilience attributes (Tables 5.1and 5.2 above) such as completeness, accuracy, simplicity, robustness, speed and can be validated.

5.5.2. Customer engagement and satisfaction

A number of interviewees identified customer engagement and satisfaction as important to measure, and this can therefore be viewed as another key retail performance metric relating to the FA of the customer. Like the measurement of customer numbers discussed above, there was variation in how this metric was used. Net promoter score was clearly one way to measure customer satisfaction and engagement: “Net promoter score […] where we're trying to benchmark our performance against those of our peers and competitors.” Finance Director, C.

Mystery or guest shopping scores were also important for gauging customer engagement or satisfaction for organisations: “We do well over a 100 guest shops every month. And therefore our real focus is on- what is the journey that the customer sees?” Managing Director, B.

One retailer also expressed customer engagement through their CLV modelling process which can be categorised under this FA as it is also used to generate incentives for customers as noted below: “the most exciting part of customer lifetime value is the work we're doing on predictive customer lifetime value […] All of that is within the model, that then kicks out a number for a given customer or customer segment […] we're starting to be able to differentiate our levels of investment. For example, receiving a £30 coupon instead of a £20 incentive” Strategy Director, C.

The final quote in Table 5.3 below also emphasises the very informal and soft nature of some customer engagement and satisfaction measurement. The quotes show a considerable variation in how case company respondents assess customer satisfaction and engagement ranging from internal checks to agencies doing independent checks.

134

Table 5.3: Measurement of customer satisfaction and engagement comments

“the customer service team do monitor Trustpilot Review15 […] Finance Director, A. if you talk about service level, in that regard, then it is reviewed and appraised. “We use Rant & Rave16 [….] we ask three questions, and that's Customer Director, B. how we monitor it. We ask for a little free text, and that comes automatically into all our inboxes.” “I use Google Analytics mainly […] this is where I think it's Marketing Director, A. going to go – personalization” “We look at the amount of complaints. We encourage stores to Finance Director, D. resolve complaints […] in the store […]In the event that the customer isn't satisfied […] then we have a customer care team here. We get a monthly report.” “Customer feedback is key to us...We collect it in two main Finance Director, D. ways. One is where […] you should get handed […] a little slip that says, "Look, why don't you be a mystery shopper for us and rate us from one to ten. […] then we also do mystery DVD visits, which is much more focused […] hidden cameras […]They're external, but they only work for us[…]So that's something we live or die by”

A variety of different techniques are used from email; text messages; feedback forms; to secret DVD recordings. In summary, the metrics employed in this instance were identified as:  Customer complaints;  Guest shops;  Mystery shopper;  Rant & Rave;  Trustpilot review  Net promoter score;  Google analytics; and  CLV modelling. A clear theme is the apparent obsession with trying to understand the customer and satisfy them for as long as possible.

5.6. Sales

5.6.1. Statutory reporting

Sales was the second ranked of the FAs drawn from the case data. There is a statutory requirement to report sales defined as ‘Revenue’ by accounting standards (IASB, 2010) and disclosed in the Consolidated Income Statement of companies (Deloitte, 2012). There are many rules for income recognition and reporting, and relevant regulators are still trying to improve the accounting and disclosure procedures where sales are concerned (see Chapter 3). Almost

15 Commercially available agency operated process for on-line businesses providing a rating score out of 5 stars. 16 Commercially available agency operated customer feedback process using 3 questions via mobile phone SMS.

135

every interviewee considered statutory reporting as a compliance exercise and expressed a desire to keep disclosure on sales to a minimum. Some even questioned the value of the statutory information as noted below. “nobody will see the stats [ …] they’re unimportant to us […] the only people interested are the press when we file […] none of them actually understand them, because we never report our concession sales […] that’s 55% of our turnover. So, nobody actually really understands anything about us” Managing Director, B.

This said, the measurement of sales was identified as important by some case interviewees: “we look at all of our sales, which I think is unusual compared to other retailers, VAT inc. […] Yeah, that's old-fashioned. Daily takings are what you put in the till. The shop colleagues are motivated by sales” Finance Director, D.

“retail sales, sales growth […]Top one, obviously, is retail sales […] Sales has always been very important.” Finance Director, C.

Interviewees also identified that retailers use a variety of metrics to chart sales and these are explored below.

5.6.2. Internal reporting of sales Most retailers have systems and processes in place to gather information internally on sales, both offline and online, usually from their EPoS systems or via other summaries of transactions. In this respect, there would appear to be a uniquely retail industry focus on a weekly reporting process of cash takings, sales or trading figures. Some of these terms are used interchangeably by the retailers although most acknowledge that for statutory reporting purposes ‘revenue’ excludes VAT. Comparing a week’s sales with the previous week, the same week in the previous year, and with other internally set targets or budgets, is embedded in most retailers’ reporting and performance monitoring. As one interviewee explained: “What we do say is that, provided sales beat last year, so, the current week against the comparative week of last year - and that ought to be our main decision as to whether we've had a good week or not” Operations Director, D.

Case company A went further in setting daily targets for their sales. “Daily revenue versus budget and target […] primarily as a key performance indicator […] is moving substantially outwith a percentage variance. That is telling, not just myself, but the other managers in the business that I'm going to be vibrating and they should watch out” Chairman,

“we do have targets - we have the daily income targets, so each day we assign a target, which is basically a prior year lifted, depending on the split from our model.” Finance Director,

“The biggest thing looking at probably will be the sales per day, something you live and die with and if it's not there [as in not achieving it] then it can ruin your entire day” Managing Director.

Thus far, internal reporting of sales has been discussed from the perspective of temporal variation, in terms of weekly or daily reporting. However, case interviewees also emphasised that internal sales reporting can be segmented in terms of a variety of factors, such as

136

departments, brands and products and even stock keeping units (SKUs). The following two quotes illustrate the variety of measurement approaches taken in this respect: “Within our weekly trading meetings, we have turnover and we have it by brand, by geography and by our product category, and actually by SKU” Finance Director, A.

“We will have an overall commentary, and then we'll go into each business. We'll go into the W business, including the split by High Street and supermarkets. We'll go into the X business, again, split by the various categories in there. Y, which is franchised, Z et cetera”. Finance Director, D.

5.6.3. Like-for-Like Sales

LfL sales are used by every retailer interviewed to monitor performance. It is also a metric that is widely published by the retailers in their trading statements and ARA. The key comments are below and in Table 5.4. Table 5.4: Comments on like-for-like sales measurement

“the like-for-like comparison is incredibly important to us Finance Director, A. […] that's why I ranked it as number one.” “Like-for-like sales, is always a top agenda item - How Operations Director, D. is the business doing? Our guide is simply that, "Did we beat last week's figures?" “Like-for-like sales. is discussed at […] trading meeting, Marketing Director, A. management meeting and board meetings, we talk about that and improvement or lack of” “like for likes - whilst a favoured metric, forensic metric Chairman, A. […] is still very useful” “like for like sales […] Continuous sales, always looking Finance Director, D. at that. Make sure we're performing […] especially given the number of stores we're adding. And the fact that we add new services.”

The LfL sales metric is clearly important for the retailers studied, but some were also keen to emphasise that it is not something they regard as definitive: “Like for like sales is important. But it is not top of the list. The biggest reason why it's not is that in this industry there's too many [mobile phone handset model launches by the manufacturers] things every year [LfL] so is not the be all and end all”. Risk Assurance Director, E.

“It's the same mind that like-for-like sales are good, but it's not the be all and end all for me. It's, have I made money year on year?” Managing Director, B.

5.6.4. Summary

Case study analysis reveals that sales are clearly an important metric and for retailers, internal sales reporting and LfL sales appear to dominate management attention. This said, it is clear that different retailers take slightly different perspectives on the sales metrics, particularly where internal reporting of sales is concerned, with reporting variation along the lines of temporality and segmentation being key.

137

5.7. People

5.7.1. Introduction

After customer and sales, people came up in conversations with case interviewees as a dominant FA. The quote below to some extent reflected recognition that good and poor performance was contingent on the quality and ability of the people, and specifically employees: “we're there to serve customers and we employ a lot of people. We're only successful through our people […] They're our most valuable commodity and they serve one of our most valuable commodities which is the customer [...] Two things that you don't actually stick on the balance sheet” Finance Director, B.

Putting a value on people or employees and accounting for them, in the retail performance measurement literature, is noticeably absent; although as identified in Chapter 3, one retail failure prediction model factored in issues of retail employee headcount although it should have been expressed as a full time equivalent (FTE) figure, which is typically done in the UK retailer ARA. Also, staff costs are usually detailed in retailers’ accounts. There is extensive disclosure of directors’ remuneration for many retailers and a statutory requirement to give the remuneration of the highest paid employee.

5.7.2. People engagement

Two key methods for monitoring employee engagement were used by the retailers. The first of these involves surveys and on-line discussion forums. Table 5.5 captures some of the case comments regarding the monitoring engagement in this manner. In addition, it was also clear that case retailers took staff turnover figures as an indicator of how engaged and committed their staff were, high turnover levels were seen as potentially problematic. Table 5.5: Measuring people engagement via surveys and forums

“One of our key measures is people engagement […] Strategy Director, C. Every year we do a survey on all our people […] to create a score for engagement” “why the Happy Index is so important is because of our Finance Director, D. upside down management. The most important people are the colleagues that serve the customers” “We have intranet sites […] designed for people to Finance Director, C. instantly raise issues we call ‘yammer site’ […] It's a discussion forum […] 'your shout' which is an internal forum for issues.”

As two interviewees noted: “Staff turnover […] We need continuity of people. We spend a lot of time training people and therefore making sure we get that continuity […]. When I joined we had in excess of 50% labour turnover. Now we have low 20s […] it allows us to train and manage and develop, but it also allows us to manage our bottom 10% in under-performers” Managing Director, B.

138

“Labour Turnover [...], across the stores it's around 30% [...] the main concern is the store turnover […] Five and a half thousand people in the business, of which around 5,000 of those are in-store.” Risk Assurance Director, E.

5.8. Margin

5.8.1. Introduction

The conversations about margin as an FA varied extensively between case retailers, although they all expressed a focus on profitability to some degree. Margin is a FA that appeared to hold the most confusion in terms of thinking how it might be measured; emphasising a need for clarity in terms of both its definition and subsequent calculation. Indeed, in many ways such clarity appeared to mark a key distinction between how failed retailers amongst the cases had measured margin, and how successful retailers continued to do so. It is therefore worth analysing margin as a possible performance metric from the perspective of failed retailers and successful retailers.

5.8.2. Failed retailers and margin measurement

Comments noted below, from a ‘failed’ retailer constituting one of the case studies, suggest confusion and a lack of understanding of margin management: “the concept of markdown - which was lost - nobody understood that […] intake margin to exit margin, nobody understood all of that. I don't call those financial measures, I call those commercial measures” Finance Director, F.

“There was no understanding of the component parts of margin, as well […] I couldn't tell you how much markdown, what original selling prices were […] So I couldn't tell you intake margin to net margin - waterfall, as you call it - which is exactly I agree, because it was just an unknown concept.” Finance Director, F.

From these two interview excerpts, it seems clear that this failed retailer did not have the necessary components of margin measurement correctly identified or visible to them. As a result of this absence in the appropriate identification and flow of information relating to margin, the company was unable to establish if it had made a profit. With another more recently failed retailer from the case studies there did appear to be greater visibility of appropriate information but not in a joined up way to give accurate margin metrics as the quote below indicates: “At the macro level we are into supplier funding, network promotions and […] offline discount which adds straight to the bottom line […] we do have all the supplier funding and the rebates, it's actually calculating an accurate exit margin […] that's a huge challenge. Now, having said that, thinking back to my ASDA days, what I did know that ASDA did do, will actually allocate the rebates and the retros right back down to SKU level, so you actually knew at that level exactly what your profitability was by SKU. ” Risk Assurance Director, E.

An acute awareness of what might be required to calculate margin effectively is given in the quote above by reference to previous employment history. Yet unfortunately this does not seem to translate into the accurate calculation of margin for this particular case company.

139

5.8.3. Successful retailers and margin measurement

The other case study participants representing successful retailers appear to have a better understanding and use of margin terms and margin performance management. Thus, as one such interviewee noted: “Daily margin and brand mix participation - The daily margin I watch with a hawk- like vision because I'm monitoring the targeted intake margin from the open to buy process, versus the net outtake margin [...] The output margin that we look at thereafter is the difference between negating all overhead cost in the business, to finite output where we're looking at pure profit on a margin basis”. Chairman, A.

Here then is an example of a retailer that is focused on a true ‘exit margin’ calculation and measurement to get to the ‘net net’ profit - i.e. profit after all costs are defrayed. Other comments regarding margin measurement from successful case study interviewees included: “We will talk about margin rates […] So, unbought margin rates, in particular. What they're doing. The overall mix-effect of what mix of our unbought versus concessions business there is […] it's not really the intake margin. It's the exit margin really […] we used to call it ‘net achievement margin’ after discount […] mark downs […] So, we always keep an eye on that. It's probably a big one.” Finance Director, B.

“Yes. Branch contribution is vitally important […] because our Area Managers, that's the area that they're expected to be guided by, branch contribution, rather than in the net profit.” Operations Director, D.

In the second of these two excerpts, the retailer provides services such as shoe repairs and key cutting, it is worth noting that ‘branch contribution’ (i.e. margin by outlet) is the favoured metric, rather than a margin on physical product, as in the first of the two interview excerpts, which represents the case of a more conventional department store type retailer. Additional comments from case interviewees emphasising the importance of measuring margin for successful retailers are detailed below in Table 5.6.

Table 5.6: Margin measurement comments in successful retailers

“I also oversee a daily margin analysis […] and then Finance Director, A. obviously we have a trading meeting where we'll look at that with the team on a weekly basis, and that's current week versus prior week, or month-to-date versus prior month and also prior year” “For our 1,200 or so shops, we have branch trading accounts Finance Director, D. […] branch contribution. This year against last year.” “The material margin […] is the margin on the actual Managing Director, products. Keeping that at a level where there is 50% or 40% A. on the actual products we sell, that can be eroded by discounting, by clearance products […] so we need to keep an eye on that.” “the margins on the concession business tend to be smaller. Customer Director, If we get a lot of growth, or our growth is coming in terms of B. results from concessions, it will drive the business in a completely different margin than if it's on both.”

140

5.8.4. Summary

The importance of understanding the margin calculations and in particular the exit margin has been highlighted by the data and specific comments in the interviews from both successful retailers and those that have failed.

5.9. Cash

The adage ‘cash is king’ is commonly known, as is the reason for most company failure being that they run out of cash (as the quote below acknowledges), which in reality is more of a symptom of poor decisions rather than the underlying cause. It is not surprising, therefore, that ‘cash’ emerged from the case interviews as an important FA in the measurement of retail performance. As one interviewee put it: “very good businesses fail through poor cash management […] management of cash is important to me as appraising the business results so we have a daily cash flow forecast which we monitor and anything cash related comes through me, and so we really have an incredibly tight hold on cash, and rightly so”. Finance Director, A.

Ways of measuring cash featured amongst the favoured retail performance metrics mentioned, particularly amongst the Finance Directors who often claimed to monitor the cash position of their retail business on a daily basis. One such interviewee explained it thus: “Looking at the cash balance, this year versus last year, it gives me an instant picture, as to which way the company’s going […] The chairman, and the owner of the business, has used that for many a year, and I've now learnt how beneficial it is […] Every day we have an email that’s sent down, that says, "This is the balance this year. This is how it was last year," and we do that by business as well [...] Just from the bank accounts. It's a guide” Finance Director, D.

Eight (non-finance) directors interviewed shared this same view about the importance of cash management as a performance measure. There is a very strong emphasis on the ‘daily cash balance’ as a key metric. Nevertheless, one interviewee identified that the focus on cash and cash flow as something worthy of measuring has intensified post the 2008 recessionary crisis: Very few retailers - up until fairly recently, I think and even ourselves, included - looked at cash flow […] up until the 2008 crisis, people got caught out, because cash flow isn't what you looked at, especially if you were a PLC. It's crazy, really, when you think about it […] It's the most important thing in my book.” Finance Director, B.

In line with this, another interviewee commented that “it takes a crisis to get a focus sometimes”. A potential problem identified with cash as a metric for retail performance was that it is not very predictive. Specifically, it would seem a retail business can go from being cash rich to cash poor in a very short space of time, through poor management of various business functions and activities. One case interviewee, detailing the collapse of the failed retail business for which they had worked, identified how easily this can happen and the volatility of cash flow in retailing:

141

“Now, cash was a concept that nobody really got […] in the organization […] there were so many things that were just consuming cash […] Online was consuming cash, […] our bad stock management was consuming cash. We had too many people. We had too many stores […] At this point […] it became too late […] less and less cash going into the business full-stop […] 'Oh my goodness, we are going to run out of money' [...] the irony was that actually the year before we went bust, I think [F Retail] made profit and generated cash.” Finance Director, F.

5.10. Debt

The approach to debt varies from retailer to retailer, as examined below, and it appears to be influenced primarily by ownership structure. For example, a private equity owned retailer, such as case E, is highly leveraged, and as a result were comfortable with relatively high debt levels within the business: “because of the nature of the set up with the private equity house […] there is a lot of debt involved, and private equity want this business to generate cash.” Risk Assurance Director, E.

By contrast, owner-managed businesses such as cases A and D have little debt, and for them therefore, the attitude towards debt was more cautious. As interviewees from both case A and D explained: “I suppose historically […] we've shied away from debt anyway, and it's just kind of the ethos that we've built the business on; as business people that's the way we've always operated.” Marketing Director and Co-founder, A.

“We don't like debt, and we like to make sure we understand where the cash is. That's really why I put it as number two. It's more about cash than it is debt.” Finance Director, D.

Debt is also closely related to the cash and cash flow metrics via the ability of a retailer to service their business’s interest burden. Unmanageable debt, or debt burden that could not be adequately serviced was one key reason for the failure of retailer case E, whereas for case F failure was primarily due to cash being starved from the business. As an interviewee for case F explained: “there was a bond that came to the end of its term […]They paid down the bond […] So a lot of the financing fell away and they had to re-finance […] And I remember saying […]"You're going to have to re-finance to a higher number than that, because I don't think you're giving yourselves enough head room", and either they wouldn't or they couldn't […] We then […] put an asset-backed structure in place, raised finance based on stock […] Never ever do it. It's the worst thing you can do […] The reporting requirements, it was just shocking - the whole thing […] the credit insurance issue became, therefore, increasingly important and increasingly more difficult as time went on. When things really broke, it became very apparent to me and it was a real awakening […] credit insurers started pulling out” Finance Director, F.

The entire retail sector was affected in 2008 when the credit insurers pulled out of financing the retail market. F was unable to get stock and unable to pay its suppliers or raise further finance and went into administration.

142

In summary, using levels of debt in a business as a measure of performance is clearly important, but it is obviously affected by a retailer’s perception of what level is acceptable and what is not, and this, as shown above, varies enormously. Moreover, for those businesses which carry little debt within their business model, the use of debt as a performance metric may be fairly irrelevant. For example, and interviewee from case B was keen to identify that: “I don't like running businesses with lots of debt […] Having lots of that is something that you're trying to avoid. I'm very conservative in my approach.” Finance Director.

The fact that such prudent attitudes to borrowing existed in the business effectively meant that debt was not an issue, and as a result measuring it was pretty meaningless in this instance. This explains the contention of another interviewee from the same case that: “So debt analysis is absolutely unimportant to us”. Managing Director.

5.11. Property

One of the ways retailers have historically grown their business is with the addition of stores or outlets (Dawson, 2004), and as a result it is not surprising that ‘property’ was identified in the case studies as a key FA for measuring retail performance. An important consideration here is how the property assets of stores relate to retail sales and profits. In this respect, one metric frequently referred to by interviewees is ‘sales density’ typically measured by ‘sales per ft2’ as noted below: “so, if you're looking how efficiently you're using your space it's literally, "does that brand give us a contribution?” […] we'll be looking at contribution density […] if you want a snapshot for how well we're doing […]The sales density will give you […] a good indication.” Finance Director, B.

In some cases, particularly where newly opened stores are concerned, the ‘sales per ft2’ metric was being calculated and examined by case study retailers on a daily basis to chart performance: “For our 1,200 or so shops, we have branch trading accounts […] branch contribution […] we measure new shop performance […] where we open a new store, we will look at the sales every day […] we look at their profitability every week. We have a weekly profit flash for those”. Finance Director, D.

Another key issue with the ‘sales per ft2’ or ‘sales density’ metric is that it can vary significantly across a retailer’s store portfolio, and this obviously relates to issues of store location in terms of remoteness vs. centrality and associated potential footfall. As one case interviewee explained: “the store portfolio and […] location, very important [...] 720 stores, some of which are in remote locations […] In some you can range from a hundred square foot store in the Trafford Centre, which has a massively disproportionate sales per square foot […] So it's such a difference in the nature of stores.” Risk Assurance Director, E.

Whilst the sales per ft2 metric is primarily focused on sales figures in relation to space, it emphasises an underlying concern around the costs and liabilities of property, and how well

143

these were geared against sales performance. One such property cost or liability, emphasised by some interviewees, related to property rental levels: “Clearly the rentals […] made individual store P&Ls suddenly change […] if you were doing an internal rent, the cash is still within the group. In this case, it was a real rent, there was no cash, and so what that meant is, we couldn't invest in the store chain […] and not only that, but the rents that were put in place, were called Horizon Rent deals and it was just killing us. I think for 800 locations there are over 600 landlords - very difficult to negotiate, and it was a massive problem, and the store estate just was winding down and down and down in terms of quality” Finance Director, F.

Another property cost and liability, identified by one consumer insight expert who had conducted a review of the failed retailer’s store estate commenting on the case study companies, was the locked-in nature of long retail leases on store premises: “I think [Retailer F] fundamentally had enormous coverage across the UK and a very large commitment to bricks-and-mortar retail. And really an awful lot of inertia in that portfolio really thanks to very long leases […] And obviously, its financial position with the level of debts and the onerous leases that it was left with following the [Group Top Co] structuring many years before, meant that it couldn't withstand that decline in performance for very much longer […] its balance sheet was weak, and it needed relatively little to tip it over the edge.” Consumer Insight Expert, F.

A key issue with these kinds of property costs is that they eat into the profits made on sales per ft2. This means that smaller stores in less attractive locations with lower property costs may actually be more profitable, even though their sales per ft2 are often much lower: “From the point of view of [Retailer F] and its locations - and don't forget there were over 800 […] We examined every single store […] the smaller the store, the more profitable it was […] In fact, the bottom 500 stores, in other words, I'm talking about the lowest revenue 500 as opposed to the top 300, actually made net profits of 12% which in any retail business will be regarded as very healthy” Consumer Insight Expert, F.

Interviewees also stressed that the calculation of the sales per ft2 metric only ever really includes the floor area of in-store selling space, with back office and warehousing areas removed from any calculation. However, this is problematic if previous floor area devoted to selling space is simply decommissioned from use because it is deemed no longer profitable, or indeed if previous warehousing space is reallocated as selling space. Both scenarios demonstrate how easily manipulated the sales per ft2 metric is. One interviewee highlighted this problem, both in terms of their own organisation ‘recovering’ selling space from previous back office areas within their properties, and in terms of other retailers decommissioning selling space in the opposite manner. “The only metrics that I focus on is every six months I have a pounds per square foot of every inch of the building […] We've just moved back offices […] I would guess over the last 10 years we've recovered 100,000 square feet of floor space. Next year we will recover 40,000 square feet. If you […] go to Birmingham and look at the major large House of Fraser’s, at least two floors are now empty, because they can't fill it […] They can't profitably fill this square footage […] Everything we do is done on profit pound per square foot.” Managing Director, B.

For case study retailers with a predominantly online offer the calculation of the sales per ft2 metric becomes especially problematic as they have little or no actual selling space. For this

144

reason, one case interviewee working for such an organisation suggested that bringing other property space into the sales per ft2 calculation, such as warehousing space, might be a way forward in the future. “Sales per square foot - not applicable in this business, although you may question whether you would or wouldn't measure the sales per square foot coming out of the warehouse in a forensic view going forward.” Chairman, A.

Such variations and fluctuations in how selling area is drawn into the calculation of the sales per ft2 metric may suggest it’s use for monitoring retail performance should be embraced with caution, both over time or indeed between different retailers. Nevertheless, for the case study retailers, it is evident that sales per ft2 remains a critical measure of performance, despite any inherent flaws or inconsistencies within its method of calculation.

5.12. Stock control

Stock control, as a FA, manifests in many ways amongst the retail case studies, and appears more important to those retailers that buy products to stock and re-sell (case study companies A, B, C, and F see section 4.5.4.3) compared to those that offer a service or rental element (case study companies D and E see section 4.5.4.3) in this instance shoe repairs or mobile phone contracts. The importance of stock control and the metrics preferred are given, below and in Table 5.11, in the quotes from the retail interviews which emphasise product sell through17, stock returns18, availability19, markdowns20 and stock loss21 as important metrics. “Product sell- through […] is how you drive the profitability of the business. If I am hitting low 60's, I'm not making any money […] On average we exit every season with a 70% sell through […] my weekly meeting […] goes through results for the period, then where we are on the sell- throughs” Managing Director, B.

“Sell-through is important [...] Customer returns” Finance Director, C.

“a marked down rate, because if your sale is growing very strongly then your actual cash mark down might go up naturally. Your mark down rate should always be controlled, and it's tied in obviously to sell through […] simple as that.” Finance Director, B.

“7% returns percentage - We think we're low in terms of returns […] It's a measure of how effective we are at becoming a tactile online platform as oppose to a bricks and mortar retailer”. Chairman, A.

17 Product sell through is a metric that monitors product quantity, typically in units, sold as a percentage of quantity purchased. 18 Stock returns represent, typically in physical units, quantity of products returned by customers to the retailer 19 Availability is typically measured as products available for sale as a percentage of products stocked, by stock keeping unit (SKU), category and / or all stock. 20 Markdowns refer to the reduction in price of products and are calculated both in physical quantity and price reduction value terms 21 Stock loss typically has two components, known loss represented by recorded markdowns or waste which is accounted for and therefore known, and unknown loss which is often referred to as shrinkage.

145

“Stock accuracy is stock loss […] just making sure you are not losing everything or those getting stolen. Being as accurate as possible to your stock file, so you know what your availability is and your offer to the customer is […] we'll define as shrinkage as much as anything. Waste will be included in that and then loss is just known loss […] Actually, this organisation has probably the best metric […] It's half a percent […] I've never ever come across it […] That means processes and systems are rigorous, I'd stand by […] So, the focus actually is there in the employees and that's why it's very important. If 0.5% went to 2%, that's millions of pounds straight off the bottom line.” Finance Director, B.

For one retailer that has a dependency on brands for resale, brand mix participation is discussed as noted below. “The brand mix participation is to ensure that the mix of those new acquired brands is in the same plan […] At ten brands the influence of the value of the stock was really not considered.” Chairman, A.

“Brand numbers - so that's quantity of brands we hold on in the business at any one time […] We try and get in a lot of brands in order to sell- through” Managing Director, A.

The more traditional monitoring of stock levels and stock turn also feature in the interviewee comments as noted below. “Stock turn - it's how quickly or how effectively we turn that stock through and the mix between wholesale, the full price sale, and the discounted level”. Managing Director, A.

“Stock turn percentage, of course every retailer has got that one on there. Again, depending on where you are in the business that goes higher or lower” Strategy Director, C.

“we look at stock levels regularly […] obviously stock holding, stock days and because of a lot of our stock is carried forward, there's less of a risk around the stock holding” Finance Director, A.

Linking together the entire retail process involving stock as a key component is neatly summed up by one of the interviewees as follows: “You can then almost link cash-flow, stock-turn and sell-through […] you go to your sale and you sell it, cut it out as close to zero as possible, and you go to the next season with little or no old season's stock […] If your focus is on full price sales […] then that affects how much you buy. How you market it. How you manage your stock […] your finance structure, everything about how you run the store in terms of marketing.” Finance Director, B.

However, retailer F that had failed, seemingly had considerable performance management issues when it came to stock management and these are noted in the comments below. “It wasn't a typical fashion business where stock is coming and going. A lot of stock items were there all year […] Whether they held too much stock is a different matter because, again, the system just wouldn't have been good enough to get stock levels down to the levels that you would expect these days. They had these huge warehouses - huge, full of stuff, and stock rooms in each store taking up a lot of space.” Financial Controller, F.

146

“[Big F] Filling them with stock, we had stock coming out of our ears, because it was buyer-led […] I couldn't tell you how much markdown, what original selling prices were […] So the concept of markdown - which was lost - nobody understood that concept […] So those things – sell-throughs - just never talked about or when they were talked about ultimately in the end, we talked about 100% sell-through. It was just stupid […] Full price sell- through […] no one gets 100% full price sell- through […] commercially, just poor […] Stock information was very poor.” Finance Director, F.

To summarise, managing stock is seen as vitally important and the successful retailers concentrate on particular favoured retail performance metrics such as stock levels, availability, sell through, stock accuracy, markdown with the failed retailers struggling with the information and management culture relating to stock management metrics.

5.13. Buying and products

The FA here is the buying process and the products and services for retail reflecting views, as noted below, on how and what to buy with a few references to specific metrics used. Brands, brand positioning and merchandising with authority in a category are noted as important to get customer attention as the quotes below suggest. The category and product authority is emphasised by a case company and a retail expert: “To me there is the real issue of authority. If you go down into our watch room and you look at IWC [a luxury watch brand]. In any other store, you would be lucky to find one Portuguese. You will find ours in rose gold, in gold, in white gold, in platinum and in steel […] And people are now coming to us, because of this authority in category.” Managing Director, B.

“Fundamentally, the variety store offering of [F Retail] was very focused on a particular customer base. And unfortunately, if you look across all of the categories that [F Retail] sold…there was usually a specialist or several specialists that were actually out competing […] so [F Retail] was increasingly a second or even a third choice for many of those categories, and I think the only exception to that in fairness would be Pick and Mix […] Yes, music and books are obvious examples where both in terms of price, and [F Retail] really couldn't succeed if it was being beaten on price. And also homewares and many of the stores, because they were of a particular footprint, would actually still try and cram all of these different categories in. And in many of the stores, the authority within each of the categories was poor.” Consumer Insight Expert, F.

Product category again is a concern and in particular the mix of brands and the seasonality of the ranges on offer illustrating the complexity in the buying process and the challenge of getting the right products at the right time, as noted below: “we have changed our entire focus as a company, we've gone from more countryside to fashion […] so you got to make sure you got the right brand mix, the right product mix […] that's something that we measure, pretty much daily, really […] we've been focused on the winter product market […] We're changing that for next summer, we'll have brands in, bikini brands, swimwear brands, more life style brands so we'll have plenty of products for the summer […] It's all very well having a brand but then you've got to buy the right products from that brand, and the right mix of products.” Managing Director, A.

147

The complexity of the product types and need for categorisation is illustrated by the quote below where the retailer breaks down the way they group products and ranges for ease of management. “We are divided divisionally by categories, so we have toys, fashion accessories, and within fashion we measure different men's wear, then women's wear. Then we have furniture, technology, home, children's wear. We have a lot of divisions, like 11. Restaurants’ is a different one. We also measure activity based on sub- businesses within the business, and that's another cut that we look at every week…All those metrics we review and we look at as a management team on a weekly basis.” Customer Director, B.

The way the buyers operated in the failed retailer, retailer F, was discussed as a contributing factor to many of the challenges faced by the retail company as noted below and specifically relating to this FA. “It was a buyer-led culture […] it had always been number one for music […] I think that it soon became clear that you actually could get your CD for 9.99. So why would you spend 11.99 […] Then that became, I think, the beginning of the demise of what [F Retail] was, because it was very strong in music […] Seasonality - the big issues, the big seasons, were Easter and Christmas, Christmas much bigger than Easter. Easter was all Easter eggs […] I recall, even with Easter we struggled to make a profit in that month. The profit- making month was leading up to Christmas. Even then it's difficult to understand why [F Retail’s] sales would double in one month. What were we selling? Well, probably music, lots of music stuff, and maybe gifts from the clothing range or the housewares. Chocolates, I suppose, boxes of chocolates, and they used to try and work hard to get special deals with the manufacturers because of the size of [F Retail]. But the size was only because of the number of stores” Financial Controller, F.

“I think it was exactly that. So let's buy 'x' thousand units and look, 'I've done a great deal so let's put it in the store'. Well, hold on, work out whether you can sell it first and then work a deal around that. It was the other way round, it was buyer-led rather than consumer-led.” Finance Director, F.

One retailer specifically concentrated effort on ‘open to buy’22 as a key metric for buyer control and retail performance management, this clearly being absent in retailer F based on the comments above. “brands that were in the targets for acquisition to the seasonality of the acquisition by the business be open to buy” Chairman A.

In summary, open to buy, authority in category, products and brands are prevalent in the data and discussions.

5.14. Business model

A number of different aspects of the comments, as noted below, have been grouped under business model as a FA. Business model refers to the way a firm conducts its business (Zott et al., 2011, Zott and Amit, 2013). This has three core components: the strategy; the organisational structure to execute the strategy; and the processes and systems to execute the

22 Open to buy represents the gap between products already purchased against a target, typically the amount in currency value terms. Hence limits are set on buying amounts to control cash and stock purchase commitments.

148

strategy. For example Retailer E had the need to drive EBIT and cash to service a debt burden owed to private equity that leaves little scope for adjustment of terms where the retailer is dependent on two or three suppliers for their entire business proposition. So the organisation structure (i.e. private equity ownership) imposed a requirement to drive EBIT and generate cash (overarching strategy) leaving little flexibility in its processes and systems and is illustrated by the quote below. “EBIT is important here rather than profitability because of the nature of the set up with the private equity house here. There is a lot of debt involved and private equity want this business to generate cash as opposed to making any profit.” Risk Assurance Director, E.

A criticism of the private equity backed retail business structure is also made using an analogy of another retailer as noted below by one of the interviewees. “I have worked for PLCs and private companies [experience of different business models]. I think rent is a massive factor in the performance and non-performance of retail companies - especially chains [structure of retail companies]. Take a New Look [uses an analogy to make the point about structure and strategy], for example and where they might have 800 stores. They grow - looking fantastic growth, borrow open new stores grow, borrow open new stores great, we can present this story of fantastic growth, sell it on. Make a quick buck and run off to the hills. Well, it didn't quite work out like that [referring to Permira and Apax private equity funders’ inability to IPO] because of lots of things. Partly it’s because people looked at the business and said, "your growth has been coming from your new businesses and actually if you look at your existing business take out your new investments all the time they're actually going nowhere in fact they're probably going backwards and events are going to kill you and you've got 800 stores when actually you only could do with 150 or less. So, I think the private-equity model [refers to a specific ownership structure that impacts strategy and execution] in retail is a dangerous one. You know, it can go very wrong” Finance Director, B.

Another business model example is having the cash stripped out of the business, originally by Group Top Co (organisational structure) via rental burdens and then to feed a sister company (organisational structure and strategy) whose business model was cash hungry and flawed i.e. the bigger it grew the more cash was required to fund the working capital cycle see the quote below. “So [F Group] was split into three businesses: retail - which were the stores, Entertainment or [FUK] - which was the distribution company of DVD, CDs, books, et cetera, and 2Entertain and [F Group] […] All the focus was on [FUK]. Good margin model […] But it was a dying product, that's the problem. Because its CD's and DVD's […] it is an interesting model […] that you had to pay royalties immediately, you had delays in your customers paying you. So the faster you could grow, the greater your debt […] They were growing very, very quickly, but therefore consuming huge amounts of cash […] At this point, then it became too late, because the point around the cash had been getting eaten up in [FUK], just became exasperated as there was less and less cash going into the business full-stop […] But interestingly enough, just to be clear, [FUK] brought all these down. That's the irony, not the other way around. Yet, because it was called The [F Group], that's how it got reported […] but [FUK] actually brought [F Retail] down” Finance Director, F.

These ‘strategic rifts’ seem to have led to operational dictate that could be considered as the final reason for failure as illustrated above.

149

Metrics for business model are harder to get to but could be considered under the broad category of ‘culture (implementation of processes and systems) and organisation structure’ i.e. the two key components for delivering the strategy. So the next quote shows a change in strategy, organisation structure and operations to refine the business model. “This is the realization that we're an integrated online retailer [realisation about the business model]. And from the business concept here [consequent need to adjust operations], we sold stores, catalogue stores and the material parts of other businesses. A lot of the middle, small JVs were really closed out […] So we sold off the distractions at that point […] we really restructured ourselves to become an integrated online retailer.” Finance Director, C.

The quote below shows a change to the strategy and hence business model by going on-line and executing this poorly exacerbating a culture of “retail sins” (processes and systems to execute the strategy). “They went down the route of furniture at one point online. It was at one point, the fastest growing online site in Europe. The more it grew, the more we lost, because a lot of the costs were directly variable, and logistics was killing us. They set up an online site to be able to deliver a packet of Maltesers to the Outer Hebrides, and made it available and people ordered that. So those things were just sins - retail sins” Finance Director, F.

A change to strategy may not always be problematic but a considered decision that allows focus and direction as noted below. “Our future success is driven by the fact that we are absolutely clear on our purpose […] our ambition [strategy is clear], and we're absolutely clear on the places we will and importantly won't focus to get there […] we took choices [structure and execution decisions] - we chose not to go international […] not many online retailers have made that choice.” [choice on the shape of the business model] Strategy Director, C.

The quote below actually refers to ‘business model’ as a ‘fundamental challenge’ and hence is the voice of the participant. It also references the organisation structure and type of business historically i.e. ‘high up in the FTSE at one point’. Then the participants goes on to explain when he believes ‘the start of the end’ was i.e. when Group Top Co demerged. “The fundamental challenge with [F Retail] was actually-- probably, it was never a phenomenally good business model full-stop for an awful long time. Although it was high up in the FTSE at one point in all those things, but I think the start of the end was when [Group Top Co] demerged” Finance Director, F.

The quotes, individually explained above, clearly show references to substantial changes or requirements of the business model as defined at the beginning of this section and the ‘business model’ was the context of the conversation. A cold read of the quotes out of context may lead reviewers to think about alternative reference points based on their definition of ‘business model’ or their limited knowledge and experience of retail and the data presented. Multiple views may be possible depending on the depth of understanding and closeness to the data. Therein lies the challenge of interpretation. Cold reading and external interpretation of published information may lead to a superficial understanding as discussed in Chapter 3.

150

Further explanations in square brackets are given for the quotes presented below where the interviewee reflects on the business models of other retailers after recent conversations with peers in the retail sector. “I spoke to […] our ex-finance director who runs Game [using an analogy to make a point about strategy and ineffective business model] He said, "What do you think?" and I said, "I really would be staggered, if institutional shareholders are so stupid to have watched the business go through a business cycle. Let it go into receivership, watch the business cycle come to a head, recognize there is no strategic future in the category [specific reference to business model – has no strategic future] and still invest in an IPO [new ownership structure].They have got three or four banks that will support it […] but it's not even an economic cycle. They've had Grand Theft Auto, which was a massive release. They've had two new Consoles launch which always grow a category, and that's a two to three year cycle. Next year there'll be no new Consoles, there'll be no mega game, so they'll go down again.” Managing Director, B.

“It's an interesting sort of change, because I think that the traditional department stores will struggle [reference to the business model of department stores], and that's why Debenhams are struggling […] House of Fraser is also trying to flog itself so that is the interesting bit, but that's the PE House wanting to get out, and the management wanting to cash in […], John [reference to the then CEO of House of Fraser] has been trying to get out of that for the last three years. I think he is just trying to find a sucker who will take it on […] and I told him because I said it to him last time when we talked over Christmas, he asked how we did and I said “we’re fine”. He said, “we had good growth”, yeah, but “you told me you did it all on the internet, which means you made 5% less margin on it”. “Yes”, he said, “but nobody really understands that” Managing Director, B.

In Summary, a varied view on business models that work (or do not) for retail companies is given. A preference prevails for private company organisation ownership structure but not private equity financed as it involves large amounts of debt. There is also a debate around the strategy and execution processes i.e. size of real estate and rentals that can affect the viability of the business model. Similarly, on-line retailing strategies impacting the retail model were discussed.

5.15. Risk and regulation

The awareness of retail risk and regulatory compliance varies considerably amongst retailers, even within each case study by board member. One thing they unanimously agree on is the irrelevance of the company beta value in their understanding of retailer risk and how they perceive and manage risk. “I won't spend any time addressing my ignorance on […] the company beta value” Chairman, A.

“the company beta value, I couldn't even tell you what that was.” Operations Director, D.

“the company beta value, that really is an unknown quantity […] It's not really understood by the Board” Risk Assurance Director, E.

151

The retail interviewees displayed broad operational understanding of risk, regulations and compliance which is more attuned to the regulatory pronouncements such as the Turnbull report (ICAEW, 1999) rather than academic perspectives of financial structure risk (Modigliani and Miller, 1958, Sharpe, 1964) proposed by the company beta calculation, and this is evidenced by the quotes below. “if you take customer risk […]You get, regulatory risk […] That goes down the new FCA [Financial Conduct Authority] environment […] You also get electricals’ regulations […] Are we doing all the right things in terms of our client database […] payment protections, access to data and complying with the Data Protection Act?, are we treating our customers fairly [TCF]?” Finance Director, C.

“We keep a very detailed risk register. The risk register drives all of our internal audit processes and our continued improvements.” Managing Director, B.

“We do go through all of these registers […] it is a well-run process that will inform decisions around how we prioritize our IT spend, for example […] Health and Safety around food, working practices, all of those kinds of things, those risk measures” Finance Director, B.

“What the Board does cover is a more general risk process which is facilitated by yours truly […] when you do come into a business - a very, very reactive business like E, it's very dynamic […] a day to day mitigation of risk definitely with less theory.” Risk Assurance Director, E.

Specific risk categories were also mentioned such as banking regulations and monopolies and competition rules as noted below. “we have a bank […] that's a whole other set of risk measures attached to it […] you've got the whole FSA [Financial Services Authority] rules for the bank that you have to deal with” Finance Director, B.

“we didn’t have Monopolies and Merger Commission [“MMC”] approval to merge the businesses in 2004 […].if you go back to 1999, [C] at that point tried to merge with Freemans and the MMC basically said No […] It’s against public interest.” Finance Director, C.

For the relatively new retailer, retailer A, the process of risk management appears to be rudimentary as noted below. “The risk measures, I would have to say, in business at this juncture, are financially influenced more than not, are intuitive more than not”. Chairman.

In summary, a common theme around risk and regulation is that the interviewees do not recognise beta as a measure of risk and that risk management practice and metrics in the case study companies are diverse with no tangible or commonly used ‘hard’ metric that can be picked out of the interview comments although compliance with regulations was acknowledged and considered pertinent.

5.16. Profit

Somewhat like sales and margin discussed earlier, the term ‘profit’ has been used as a metric but it does seem to have different emphasis amongst the interviewees with one favoured metric

152

being ‘net profit’ and another being Earnings before Interest, Tax, Depreciation and Amortisation (“EBITDA”). “The money that the business makes […] is the difference between what your core costs are and the money that you're making on top [i.e. Net Profit]” Marketing Director, A.

“Key metrics around […] EBITDA, PBT [Profit before Tax] operating profit, profit by division […] absolutely key” Finance Director, C.

“In the first half of the year - because our year end was ended January […] we used to have to work hard to make a profit.” Financial Controller, F.

“Then, beneath cash is EBITDA and PBT” Strategy Director, C.

“effectively from a business finance perspective, it's always your bottom line profitability that is your ultimate appraisal of the business” Finance Director, A.

In summary, the retailers have an understanding and use of the metrics for profit although the focus and emphasis between case studies and metrics favoured differ from Net Profit to Profit before Tax to EBITDA. However, the common theme is a focus on profit as an area of management attention making this a pertinent FA.

5.17. Return on investment

A divergence of views emerge about return on investment (ROI) depending on whether or not it is about specific project returns or returns to shareholders. When considering returns to shareholders, there is a polarisation of view with one retailer considering it paramount and another considering it unimportant. “Return on capital employed […] we do discuss that at both management and at board meetings.” Chairman, A.

“We don't look on return on shareholders' equity, because that's relatively unimportant to us […] What we look at is our return on any investment that we make in the business.” Managing Director, B.

Managing project returns using return on investment, payback period and internal rates of return as metrics is a key concern and actively used. “the main element there is return on investment for a new store opening and number of weeks payback.” Risk Assurance Director, E.

“Return on investment […] we review it monthly. Monthly we look at all the capital investments and all the capital projects, and if they are a green, amber or red.” Customer Director, B.

“we do have internal models, which try to discriminate between different types of expenditure against our internal rates of return” Finance Director, C.

“you don't want to be sending traffic to a site that potentially you've paid for via PPC [pay per click], that will immediately bounce […] I suppose it's about getting the right ROI on that particular channel.” Managing Director, A.

153

“when you try and make an exact science […] you could catch a cold […] we always convince ourselves I've got two year payback, probably […] 50% was going to go in that direction anyway because of […] other factors” Finance Director, B.

The last quote also illustrates a pragmatic approach to understanding claims made about the projects impact on the retailer’s performance. In summary, payback, ROCE and ROI are key metrics emerging from the interviews as metrics used when considering this FA.

5.18. Brands

If a retailer is engaged in the business of selling branded items then the measurement becomes vitally important. Even if the retailer does not sell brands, knowing what it stands for i.e. its brand values is also seen as important and was conveyed by the participants.

Brands in the context of recognisable consumer products are mentioned below and how the different directors view them from within retailer A, who are active in managing the performance of the brands. “we've had for years […] Barbour and Hunter […] So the idea is to add in brands around those, elevate our mix of brands upwards towards luxury / premium […] Yves St. Laurent or Chanel or Gucci, and maybe Mulberry or Burberry […] We've gone from 12 brands last September to 58 now. We'll have 80 by December” Managing Director,

We try and get in a lot of brands in order to sell through, but it's about refining that mix and refining the brand number.” Finance Director,

“we're measuring […] what the business did with the brands […] in the same period last year, versus the new brands this year, with the category displacement that those new brands have against the old ones”. Chairman.

Contrast the above comments with those of retailer F that failed and seemed unable to exploit the brands noted below. “If I think back, Ladybird as a brand was amazing. Chad Valley as a brand was amazing […] They couldn't quite exploit” Financial Controller, F.

The corporate brand value was also considered important and two of the interviewees discussed this with one having a clear ‘mantra’ and the other describing the process of engaging with the customers to understand what they think about the retailer. “we've got what we term as our brand values which is: the British, the luxury, innovative, service, and sensation. Those are our core brand values” Managing Director, B.

“Let's ask them what they really think about the […] brand both […] rationally and emotionally […] we now have a business strategy […] defining four or five really specific rational things and two or three really specific emotional areas.” Strategy Director, C.

154

In summary, the case study participants make comments about brands that are acquired for resale as well as what the retailer brand represents, so brand values and brands for resale feature within the salient FAs.

5.19. Digital

The reference to digital as a focus area is made in many different terms such as on-line and mobile mainly referring to the way customers shop as noted below by reference to website activity and customer acquisition through pay per click search engine process. “If we look at a website, at [B].com, that's a different channel.” Customer Director, B.

“Organic metrics is […] measuring […] how our growth comes on the website. The inorganic, in PPC [pay per click] and the search engine optimization” Chairman, A.

Another case company C, referred to the vision of the shareholders and how they wanted to digitise the traditional hope shopping business and then refers to the increasing migration of customers towards using mobile technology. “the shareholders […] were […] very interested in the online world and […] knew that the home shopping business was actually the online shopping model of the future” Finance Director, C.

“Probably by two or three years ago […] we’re […].pushing down the Laptop and PC [channel] and now what we’re seeing is Mobile and Tablets. Probably 25% of our Sales are now driven by Mobiles and Tablets.” Finance Director, C.

The comments suggest some on-line metrics as well as digital being an important channel for some of the retailers.

5.20. Competition

Competition is one focus area all the retailers mentioned and frequently it was named competitors indicating an acute awareness of the competitor set, as noted below. “we've got Net-a-Porter, we've got My Wardrobe, we've got Mr. Porter. So we're looking constantly at what the competition is doing. We also look cross sector, so Wiggle have a fantastic website” Managing Director, A.

“We're head-to-head with Carphone and we're semi head-to-head with the [mobile telecom] network stores.” Risk Assurance Director, E.

“we compete with a number of other northwest businesses, N.Brown's and probably Express Gifts to move away from catalogue business.” Finance Director, C

“there was a Yorkshire store called Wilkinson's, whilst [F Retail] collapsed when it did, it would be even harder for it to survive these days with the likes of Poundland and Primark.” Financial Controller, F.

Competition is therefore noted as a key FA.

155

5.21. Supply chain

Supply chain is a broad term and referred to by the interviewees, as noted below, in terms of specific metrics like availability, service levels, unit handling cost and warehouse operations. Specific mention of availability and service levels were emphasised by one interviewee as noted below. “demand [availability] is absolutely gross, what customers are looking at […] whether or not it’s serviced [service levels]. Demand is a key metric within the business” Finance Director, C.

Another case company considered the free shipping as paramount in their operational success.

“When I did free UK shipping that was pretty much unheard of, but the free worldwide shipping thing, I don't think anyone did it at the time”. Managing Director, A.

“What [A] have done is […] free delivery […] so that when the person goes on the website, what they see is what they pay”. Finance Director, A.

Other comments focused on logistics companies that provided the delivery service and the unit handling cost as noted below. “We used Royal Mail and DPD. We've tried pretty much every courier under the sun […] DPD offer […] a fifteen minute window. So it pretty much tells you exactly when the parcel's going to arrive […] By no means are they the cheapest.” Managing Director, A.

“I'd say Yodel is […] a prime service provider to us” Finance Director, C.

In addition, some expressed views about the cost and efficiency of warehouse operations in the supply chain discussions as illustrated below. “We do very well in Australia [...] We've got a full plane going across there pretty much every week, full of stock. We even have a local warehouse” Managing Director, A.

“we need to understand the efficiency and effectiveness of our warehousing operation” Finance Director, A.

“there was a big distribution centre, all manually-intensive distribution centre. That was one of two, the other one was in Swindon” Financial Controller, F.

Supply chain, including the logistics and warehouse operations are mentioned as well as implied availability and service level measures under this FA.

5.22. Market share

For some of the retailers, the market share metric was discussed but considered less relevant in their sectors whereas others stated it as hugely important as noted below. “Hugely important […] if our market share goes down […] we know that's because people are not getting what they want […] and they've gone to Carphone Warehouse.” Risk Assurance Director, E.

156

“every week we have a share report […] we have a few different suppliers, but Kantor is one of them.” Strategy Director, C.

“Market share, we talk about it in the board meetings, but there's no actual metric that we use to measure it” Marketing Director, A.

For some interviewees, market share was acknowledged as a metric but in practical terms did not rate highly for their personal attention with one highlighting the difficulty of measurement on a global scale. “Market share, do we compare market share, relative market share? Not particularly, to be honest” Finance Director, C.

“Market share is not that important to us” Managing Director, B.

“We're on a global landscape. So […] to appropriately state our market share is almost impossible” Finance Director, A.

In summary, where market share information or metrics are available they are sourced and discussed in broad terms. So it is a critical FA for some but not others.

5.23. Growth

Growth has been referred to in different ways by the retailers and appears to be an implicit and accepted part of the retailers’ purpose and hence retail performance metrics as noted below. “we're all here to create sustainable profit for growth” Strategy Director, C.

“we're measuring the two and sometimes three various areas of how our growth comes on the website […]You need to look at prior and prior years to look at the growth of the business” Chairman, A.

“Income growth is very important […] FS [Financial Services] book gross yields, so how much interest are we making off our financial services book.” Finance Director, C.

“growth rate. How are you changing year-on-year and how that can be portrayed in percentage terms […] CAGR [Compound Annual Growth Rate]” Marketing Director, A.

In summary, metrics like sales growth, profit growth and even interest growth are mentioned by the interviewees making growth a salient FA.

5.24. Summary of focus areas and metrics

The first order analysis of interview comments that have been categorised into 20 FAs and these are summarised below in two tables, Table 5.7 and Table 5.8, giving the FA and related metrics used by the retailer case studies. They are presented in this way in order to make sense (Gioia and Chittipeddi, 1991) of what retailers think about and discuss in the boardroom and what metrics they actually use. There has been a deliberate use of extensive quotes to give voice to the participants in the analysis and reporting of the findings. This techniques also serves to validate the themes (FAs) reported. What has been identified in this first order analysis is discussed by way of a second order analysis (Langley, 1999, Gioia et al., 2013) in the subsequent chapter by also comparing to academic literature and theoretical models.

157

Table 5.7: First Ten focus areas and related metrics

Focus Area Metrics

1 Customer Customer numbers (footfall); Customer numbers (clicks); Customer conversion i.e. average weekly transactions; Net Promoter Score; Customer satisfaction index; Customer complaints; ‘Rant & Rave’; Mystery shopper i.e. guest shops or DVD audits; Customer feedback card; Trust Pilot Review; Google Analytics; Loyalty programme; CLV model 2 Sales Total Sales i.e. Turnover or Income; Income by geography, branch (store), brand, department, category, segment, SKU; Sales by year, month (four weekly months for some retailers), week, day; Sales including VAT, excluding VAT; LfL Sales. Sales v prior period. 3 People Staff numbers (FTE); Staff Turnover percentage; Staff Retention percentage; Staff engagement e.g. Happy Index, Feedback and discussion forum / portal i.e. ‘yammer site’; Training completions 4 Margin Gross profit margin; Operating profit margin; Net profit margin; Buying gross margin; Commercial gross margin; Buyers net margin; Contribution by branch; Margin components, waste, markdown, stock loss, shrinkage; Intake margin; Exit margin; Margin by, day, brand, product, own v concession 5 Cash Daily cash balance; Daily cash flow; Cash performance i.e. cash generated and cash drivers; Cash by, period v prior period v prior year v forecast; 6 Debt Borrowings i.e. debt owed; Bad debt write-off; Debt analysis i.e. leverage or gearing ratio; Bond; Borrowing headroom; Interest cover; Credit Insurance 7 Property Sales density i.e. sales or profit, £ per ft2 ; Brand contribution; Store contribution; Number of stores; New store, daily sales, contribution & profits; Store size; Store rents; Long leases; Sale and buy backs; Store network coverage 8 Stock control Owned stock holding by number by value; Concession stock holding; Stock by, department, brand, category, SKU; Stock turnover i.e. number of days or percentage; Stock sell-through; Aged Stock by, season, price, discount; returns percentage; Mark down rate; Carry forward lines; Stock accuracy; Known stock loss; Unkown stock loss (shrinkage); Stock availability i.e. out of stock 9 Buying and Store offer; Category authority; Product attribute; Promotions; Brand products mix; Product mix; Seasonal product; Open to buy; Supplier income i.e. rebates; supplier relationships; 10 Business model Private v Public; Group structure by division; Ownership by equity house, sovereign wealth fund, family; public listing; Culture and control; Stores and On-line; New business by segments by geography

158

Table 5.8: Second Ten focus areas and related metrics

Focus Area Metrics

11 Risk and regulation Regulatory risk such as MMC scrutiny, FCA compliance, EU directives e.g. weee; Data Protection Act; TCF (treating customers fairly); Risk registers; Dynamic risk management process; Health & Safety; Note: Company beta not used. 12 Profit Net profit; EBITDA (earnings before interest, tax, depreciation and amortisation); Profit before Tax 13 Return on Investment ROCE (return on capital employed); ROI (return on investment) by channel, by capital project spend type, by store; Payback period; IRR (internal rate of return) by green, amber, red 14 Brands Brands for resale numbers; Brand mix; Corporate brand values

15 Digital Sales by channel, on-line, mobile, tablet, new v repeat v organic v PPC (pay per click) v SEO (search engine optimisation); Google analytics dashboard 16 Competition Named competitors e.g. Net –a Porter, Carphone Warehouse, Express Gifts 17 Supply chain Availability; Service level, Distribution centre cost, Unit handling cost; Unit delivery cost; Delivery time windows

18 Market share Market Share and market share by segment

19 Cost base Pay to sales ratio or wage percentage; Overhead costs; Operations cost growth; Staff numbers 20 Growth Sales growth by segment; CAGR (compound annual growth rate); Profit growth

159

Chapter 6: SECOND ORDER ANALYSIS

6.1. Introduction

This chapter builds on the first order analysis presented in Chapter 5 by taking the 20 FAs as the first dimension and filtering 20 retail performance metrics, one for each FA thereby creating the second, the vital few, of five dimensions and then introducing three further dimensions of:  six retail performance metrics that are consistently persistent in use;  the adaptive resilience of retail performance metrics as retailers’ change; and  the journey of retailers in becoming trust intelligent in their reporting of retail performance metrics. These are outlined below and discussed in more detail in their various sections in this chapter.

As a reminder of the research gap and as noted in the quote below, retail specific financial management has still had little academic scrutiny even though the point was raised over a decade ago. “In view of the scale of the retail industry, surprisingly little research attention has been devoted to retail financial management in general.” (McGoldrick, 2002, p210)

So this research starts the discussion of retail financial management from the ‘top down’ i.e. retail boardrooms. With a particular emphasis on trying to understand how, at boardroom level, retail companies manage their performance through the use of retail performance metrics. This analysis also concentrates attention on non-financial retail performance metrics as illuminators of actual performance compared to what retail companies publish.

This chapter sets out the observations and reflections on the results with a particular emphasis on the second order analysis (Langley, 1999, Abdallah and Langley, 2014). The first order analysis, presented in Chapter 5 identified 20 FAs for the analysis of case studies, within which there was a abundance of metrics in use. This represented the first important indication of what the boards of the case study retailers actually consider when running their businesses and consequently also becomes the first dimension.

Another dimension that is linked to the one above is to sift the metrics into clusters (such as good, better and best) actually used to get a better understanding of the quality and resilience of the metrics themselves. This practical and pragmatics process, i.e. the ‘sifting matrix’ discussed in section 5.3 of Chapter 5, allows for getting to one metric out of many, see Tables 5.7 and 5.8) for each focus area, that if disclosed could provide a better understanding of a retailer’s performance.

160

A further dimension was to set aside the financial measures that are required for mandatory reporting and concentrate on the non-financial retail performance metrics, particularly those that are identified as being consistently used and referenced over time, within and across the case studies. Six of these were preponderant and are discussed.

When viewing the data and particularly interview comments as ‘past, present and future’ statements, two dimensions of a journey over time came to the fore. This is examined later with some direct quotes and important viewpoints under the applied terminology of a ‘journey’. The first viewpoint is best illustrated with a vignette of one of the case studies (Retailer C), demonstrating how the metrics and internal performance reporting has changed over time. The other four ‘successful’ retail case studies are also mapped on their journeys in this respect, illuminating the extent of the change they have undergone in terms of retail performance measurement. This provides some confirmatory evidence to the notion of adaptive resilience as applied to retailers and their use of metrics. The second viewpoint relates to disclosure choices and the journey to becoming ‘trust intelligent’ when providing retail metric information. Again this is examined and illustrated through interrogation of all the data using the explanatory matrix. In summary this chapter covers the following five dimensions.

 The categories of metrics that aggregate to 20 FAs for retail performance, voiced by retail board members interviewed as being where they apply their attention.  Twenty specific retail performance metrics for illuminating retailer performance noted as the vital few.  From the vital few, six retail performance metrics for disclosure which, it is argued, can enhance understanding of retailer performance, whereby this six form a core set that has been noted as consistently persistent and therefore represent constants over time.  The adaptive resilience of retail performance metrics as retailers’ change.  The notion that retailers are on a journey to becoming ‘trust intelligent’ in their voluntary performance reporting of retail marketing metrics.

6.2. Twenty Focus Areas

The retailers have already given voice to their FAs, together with their preferred metrics, see Tables 5.7 and 5.8. Although one of the objectives of the thesis was to search for commonality, when considering differences between directors’ functional responsibilities these are discernible from the quotes in Chapter 5, although very little came to the fore to be considered sufficiently salient. As noted in Chapter 5 (section 5.9), Finance Directors were more vocal on cash as a FA. Apart from this, all directors were focused on all the FAs as a part of the boardroom conversation. This is not surprising as all board members will usually look at the same set of management accounts and other periodic reports (Kaplan and Atkinson, 2015). Again

161

unsurprisingly, the only tangible difference between online only retailers and those with stores was the metric related to sales density (i.e. sales per ft2). Common sense would suggest that this is not a novel understanding emanating from the data, given that online retailers do not have a store based property portfolio, and therefore is considered insufficiently salient to merit any further research or thought. The differences between company ownership and structure are best illustrated by the journey to becoming trust intelligent and this is presented in section 6.6. Delving into a more detailed functional analysis e.g. use of metrics by marketing, finance, logistics etc., clustering them into strategic, tactical or operational, or categorising in terms of their use by stakeholders such as investors, creditors, employees etc. is not done here because the purpose of the thesis is to understand the boardroom conversation and not functional operations. The first order analysis in Chapter 5 and this second order analysis have shown this type of research pursuit unlikely to produce salient differences based on the data collected. This may be because the research was conducted at boardroom level, and not at the level of individual departments or strategic business units that functional differences are less apparent. What did emerge strongly was the commonality rather than differences, of the FAs and also the salient dimensions noted below. Furthermore, there is already academic and other research that conveys metric categorisation by user or function or purpose in: retail; strategic; accounting; and marketing, literature (McGoldrick, 2002, Porter, 1993, Dyson, 2010, Farris et al., 2010).

6.3. The Twenty 20

6.3.1. Introduction

The findings in Chapter 5 presented 20 FAs and related metrics, given in Tables 5.7 and 5.8, used by the case study participants. All of the metrics were used at board room level in varying numbers and combinations within the case companies, and as the guidance from Ambler (2003) noted below suggests, an attempt has been made to get the long list down to a short list. “Keep your metrics down to the few which can be applied in every company. Don’t complicate it with long lists […] the metrics message will not get across unless it is simple” (Ambler, 2003, p118).

As discussed in Chapter 4, good research is also mindful of the potential impact and use of research findings (Saunders et al., 2012). So in considering which retail metrics would present an understanding of performance that could be published externally, 20 retail performance metrics aligned to the FAs are considered.

Reflecting on one of the objectives of the thesis during the abstraction process was to be mindful of retail performance metrics retailers’ use yet do not publish. By not publishing they may be clouding the picture of published performance. Hence, 20 retail performance metrics have been selected with one per FA given that each FA was salient, whereby publication of these retail performance metrics may illuminate the underlying performance reported. In addition reference is made to the ‘sifting matrix’ working in to the axis from the furthest corner

162

where the highest levels of quality and resilience are noted thereby first selecting the ‘best’ metrics followed by the ‘better’ and then the ‘good’ to apply against each FA i.e. working logically down the quality resilience line.

6.3.2. The vital few

After applying the logic of the sifting matrix and with detailed examination of the case study data and interviews, the 20 metrics aligned to the 20 FAs are given in Table 6.1 below and were illustrated in Figure 5.4 using the focus quadrant and the attributes for quality and resilience noted in section 5.3.

Table 6.1: UK retailer metrics the vital few

Focus Area Metric Customer Average weekly transaction numbers Sales LfL sales People Staff retention percentage Margin Operating profit margin % Cash Cash generated Debt Gearing ratio 2 Property Sales ft Stock control Stock turnover % Buying and products Supplier income Business model New business segment growth Risk & regulation Regulatory investigations Profit Net profit % Return on investment ROCE Brands Corporate brand values Digital Average weekly basket size £ by channel Competition Peer group composition Supply chain Availability % Market share Market share % Cost base Sales : payroll % Growth CAGR

Overall, they are consistent with the marketing metrics presented by Farris et al (2010), see Table 6.2, who noted 119 different metrics from their research from surveying 194 senior marketing managers across industries in the USA including retail and hence it is not surprising that there is a reasonable overlap. However, their top 20 do not rank in the same order as the 20 FAs from this study and the specific metrics mentioned do vary as they cite different metrics aimed at measuring similar performance themes, or FAs as referenced in this thesis. In coming to their 119 listing, they had initially resisted ranking the performance metrics as noted below. “From the beginning of our work on this book, we have fielded requests from colleagues, editors and others to provide a short list of the “key” or “top ten” marketing metrics. The intuition behind this request is that readers (managers and students) ought to be able to focus their attention on the “most important” metrics. Until now we have revisited this request” (Farris et al, 2010 p10).

163

They cite four reasons to be wary of their ranking table (Farris et al, 2010, p10-11):  Rankings are company dependent and will vary from company to company;  Metrics tend to come in matched sets that are industry and company dependent;  Not all companies have the resources or access to all metric information; and  A metric’s ranking may change as management become familiar with it.

They are essentially giving a ‘health warning’ on using a ranked list out of context and relevance to the individual business circumstances. The Top 20 listed by Farris et al (2010) are presented below in Table 6.2 and compared with the vital few in Table 6.1 i.e. the results from this thesis research in Table 6.3.

Table 6.2: The top 20 marketing metrics

Rank Metric 1 Net profit 2 Margin % 3 Return on Investment 4 Customer satisfaction 5 Target revenues 6 Sales Total 7 Target volumes 8 Return on sales 9 Loyalty 10 Annual Growth % 11 Dollar Market share 12 Customers 13 Unit margin 14 Retention rate 15 Sales potential forecast 16 Unit market share 17 Brand awareness 18 Variable and Fixed costs 19 Willingness to recommend 20 Volume projections

From Table 1.2 page 13, (Farris et al., 2010)

164

Table 6.3: The vital few compared with the top 20 marketing metrics

Relative Focus Area Metric Farris et al (2010) Metric Rank ranking 1 Customer Average weekly transaction Customers; loyalty; customer 12; 4; 9; numbers satisfaction; retention rate; 14;19 willingness to recommend 2 Sales LfL sales Sales Total 6 3 People Staff retention percentage 4 Margin Operating profit margin % Margin %; unit margin 2; 13 5 Cash Cash generated 6 Debt Gearing ratio 7 Property Sales ft2 8 Stock control Stock turnover % Inventories 36 9 Buying & products Supplier income 10 Business model New business segment growth 11 Risk & regulation Regulatory investigations 12 Profit Net profit % Net Profit 1 13 Return on investment ROCE Return on investment 3 14 Brands Corporate brand values Brand awareness 17 15 Digital Average weekly basket size £ 16 Competition Peer group composition 17 Supply chain Availability % Out of stock 65 18 Market share Market share % Dollar Market share; unit 11; 13 market share 19 Cost base Sales : payroll % Variable & Fixed costs 18 20 Growth CAGR CAGR 51

There are notable differences for instance the vital few for retailers has customer as the first and most important FA with the average weekly customer transaction count number as the metric, whereas the Farris et al (2010) listing above references customers at 4, 9, 12, 14 and 19 using different metrics and their top three metrics are actually financial measures not marketing measures. The top three being financial may suggest an acceptance amongst marketing managers of the dominance of financial information in measuring and reporting performance i.e. a de facto standard discussed in Chapter 3.

Table 6.3 also shows FAs for this case study retail boards and their related metrics for: People; Cash; Debt; Property; Business Model; Risk & Regulation; Digital; and Competition which are noticeably absent from the marketing metrics list of Farris et al (2010). Although, Farris et al (2010, p11) state the seniority of their survey respondents as “100 [out of 195] held the title of Vice President/Director/Manager or “Head” of marketing”, the marketing focus may imply little influence or interest at a retail board room level given the gaps noted above suggesting that the marketing perspective maybe somewhat divorced from those who need to take a holistic view when directing a retail company. This divergence is more likely to be functional myopia and or hierarchical differences in responsibilities within organisations rather than major differences being due to industry given that Farris et al (2010) had retailers within their cross section of companies surveyed. For example marketing managers may have limited interest in the debt burden of the company whereas the board of directors’ should be keenly interested and aware of this. However, the cross industry nature of their survey compared to a retail only focus of this thesis may also have some contributing factor in the differences. Having said that, there are a

165

number of ‘web’ metrics referenced by Farris et al (2010) that could apply to the Digital FA although they do not specify the one identified in this thesis in Table 6.3.

As mentioned in the previous paragraph, some of the differences may be ascribed to the fact that this thesis was concentrated on the UK retail industry so making direct comparison and the generalisability of the findings beyond retailing and the UK limited. However, the vital few list, in Table 6.1, could start the debate about creating the suite of key performance indicators for the retail industry sector, which currently does not exist that the regulators are keen to see (Beattie et al., 2004, ICAEW, 2014, IIRC, 2015).

If the metrics were presented in one place as a summary table, it would be easy to spot the movement between channels and the changing business models. As shown in Table 6.4, showing the changing business models and the balance between physical (stores) and digital (e.g. on-line, phone, mobile etc.) with comparative totals, illuminates trading performance because the switch from physical to online can be readily discerned as it is disclosed.

This disaggregation for different channels would enable inter-channel performance comparison. Taking the first noted metric of customer transaction numbers as an example, current year physical may show a downturn to prior year but digital may show an increase. The relationship between physical and digital can also be determined with this disclosure example report. In addition, total customer numbers are also shown with a three year history so there is visibility of any trend upwards or downwards.

Given that business models are changing with greater adoption of digital channels, making these channel comparisons become important and this metric information would provide greater transparency for all interested stakeholders.

6.4. Consistently persistent use of metrics

In the interviews, particularly where the case study companies had been trading for more than 25 years (i.e. all cases apart from Retailer A), references were made about the past, present and future development of their performance metrics (discussed in the next section). One aspect that did emerge was persistent reference to specific metrics, Table 6.5. This implied a degree of consistency of use or favour attached to these metrics over a long period of time. Having a persistent set of metrics may appear inconsistent with the notion of adaptive resilience which is discussed in the next section. However, the persistent set of metrics was a sufficiently salient feature and merits mention. It would appear that where retail performance metrics are concerned there is always some element of metric constancy and fixity to counteract metric dynamism i.e. some metrics do not change much whereas others change a lot, it is just the proportions of this that might change depending on the retailer type and the level of retail change (see section 6.5).

166

Table 6.4: Inter-channel retail performance metrics example report

Current Year Prior Year Prior Year -1 Metric Physical Digital Total Physical Digital Total Physical Digital Total 1 Customers - Average weekly transaction numbers 2 LfL sales 3 Staff retention % 4 Operating profit margin % 5 Cash generated 6 Gearing ratio 7 Sales ft2 8 Stock turnover % 9 Supplier income 10 New business segment growth 11 Number of regulatory investigations 12 Net profit % 13 ROCE 14 Corporate brand values 15 Average weekly basket size £ by channel 16 Peer group composition 17 Availability % 18 Market share % 19 Sales : payroll % 20 CAGR

167

Table 6.5: The minimum retail ‘core’ disclosure standard

Metrics Current Year Prior Year Prior Year -1 Physical Digital Total Physical Digital Total Physical Digital Total 1 Customer numbers1 2 Customer spend2 3 LfL Sales3 4 Sales density4 5 Wage percentage5 6 Staff retention percentage6

Notes: all retail performance metrics require agreement on definition & calculation bases. 1. Customer numbers - average weekly transaction count 2. Customer spend - average weekly £ basket size 3. Like for Like Sales - same stores 4. Sales density - Sale per ft2 5. Wage percentage - sales to payroll ratio 6. Staff retention percentage - inverse of staff turnover %

168

Putting aside the financial metrics for mandatory reporting, as this has to be done, Table 6.5 shows certain non-financial retail performance metrics that could be considered as having the most longevity and were derived from the case study data. Reflecting on Table 6.5 and what has been published in the public domain in retailer ARAs, these six metrics emerge as ‘constants’ i.e. have maintained their relevance over time as referenced by the retailer case studies. These are referred to as the ‘core’ of the vital few, given in Table 6.5, and could represent the minimum retail industry standard for non-financial performance metrics disclosure.

One metric, sales ft2, is not that relevant to digital retail businesses i.e. on-line only retailers, although Retailer A, did suggest that they may be minded to apply this metric to their warehouse and despatch operations as discussed in section 5.11, Chapter 5 where they could take a broader view and include warehouse space. Then again, these digital retail businesses have not been around that long e.g. 25 years for a set of constant metrics to emerge for that retail business model.

6.5. Adapting metrics

6.5.1. Introduction

The interview participants have referred to improving their management reporting and in particular the performance metrics they use. These comments have ranged from generic statements to specific examples as noted below. The quote below discusses the timelines of developing the measurement of the case company’s performance and highlights the journey process from past to present to future. “where I see this business is not necessarily where I see it in six, 12, 18 months’ time. The way that we'll measure the business will be very different than how it is today and how it's been measured in the past”. Finance Director, A.

The Chairman, of case company A, notes below the significant change the original company founders and retail board members are going through. The retail performance metrics including the disciplines of reporting and reviewing them is often linked to internal re-organisations and system changes. “They are going from a company that didn't have board meetings, that didn't have management meetings, that had completely unstructured entrepreneurial drivers, to a discipline [daily, weekly and monthly management reports and meetings] So it's going through […] changes in the management structure, a classic environment change.”

In a similar vein, the interviewee comments on new information technology systems, where this change frequently result in improved retail performance metrics, being implemented and how this will give a better view of retail performance metrics around the FA of customer. “so we have changed completely this year […] The new system gives them all the customer information to understand not only how much worth the customer is today for us, but actually a 360 view of the customer.” Customer Director, B.

169

Certain retail performance metrics and retail performance reporting can be championed by senior people within retail companies and sometimes there can also be tension or resistance to this change and the Finance Director of case company D alludes to this in the quote below. “Dashboards […] We don't really do. One of our Non-Execs has asked us […] because he's now Chairing, […] he's relatively new into the business. He asked for a dashboard to prepare for that. He finds it very helpful; I find it useless”

Building on this point of senior individual influence, the quote below shows how the energy and vision of one person can change retail metrics being measured and also alter the focus to concentrate on important developments such as new store performance. “For new stores, we look at their profitability every week. We have a weekly profit flash for those. We do an overall weekly group profit flash. That's something I introduced.” Finance Director, D.

In summary, there is a lot of commentary about change in the past, currently in progress or planned for the future for performance reporting and metrics.

6.5.2. Changes in performance metrics exemplars

This section considers how the four ‘successful’ retailers (A/B/C/D), where internal management documents have been made available, have changed over the last ten years and how they have adapted their performance metrics. The cases are presented in order of greatest change in the sequence of Retailer: C; A; D; and B.

6.5.2.1. Retailer C

To get a fuller picture of how dynamic change can be, it is helpful to focus on one retailer at a time and look at their changes in performance reporting over a ten year period, whilst accepting that many of the case study retailers had their own comparable levels of dynamism and turbulence in performance reporting as indicated above (section 6.5.1), this first presentation is the best illustration given the amount of change in the retailer’s business. Noted below are detailed comments, illustrated with examples, from the various time periods mentioned. The Finance Director describes the development of management performance reporting and metrics under the following time periods: 1. Pre- 2003, PLC; 2. 2003 – 2005, Acquisition; 3. 2005 – 2008, Integration; 4. 2009 – 2013, Development; and 5. 2014 +, Current. The background to each of these five periods is briefly explained next.

The pre-2003 period refers to two substantial home shopping businesses operating as independent public companies based in Liverpool and Manchester and hence called the ‘PLC’

170

period. Then there is the acquisition period from 2003-2005 where the private investor ‘brothers’ acquired first one then the other with a view to merging them. The 2005-2008 period is termed integration as the two companies were brought together integrating head office, distribution, systems etc. The next period, 2009-2013, is termed development to reflect the changed nature of the then single entity and the development of its strategy. The current period is the present day. This breakdown of the timeline reflects the major change periods of the retailer but also the changes in the way performance reporting was done.

Each time period reports have profit and loss accounts, balance sheets and cash flow statements in varying formats but the key performance metric focus, arguably changes as the retailer transitions through the phases noted above.

PRE-2003 PLC PERIOD

The key performance metrics drawn on in this period are shown in graphical format in Figure 6.1 and consist of:  Sales: (by Period and Year to Date) by division with Actual versus Budget versus Prior Year;  Operating Profit: (by Period and Year to Date) by division with Actual versus Budget versus Prior Year;  Interest: (by Period and Year to Date) by division with Actual versus Budget versus Prior Year;  Exceptionals: (by Period and Year to Date) by division with Actual versus Budget versus Prior Year;  Profit Before Tax [PBT]: (by Period and Year to Date) by division with Actual versus Budget versus Prior Year;  FTEs: (by Period and Year to Date) by division with Actual versus Budget versus Prior Year;  ROSF[Return on Sales Financed]: (by Period and Year to Date) by division with Actual versus Budget versus Prior Year; and  Cashbook Borrowings by month versus budget.

This is supported by an analysis of Trading Profit, shown in (redacted to preserve confidentiality) Figure 6.2, in the traditional columnar format. The focus of the board of Retailer C at this time was on traditional financial measures, as the quote summarises, and this would be expected of a PLC that is required to report mandatory financial information. The internal reporting of performance metrics at this time, as evident in Figures 6.1 and 6.2, mirrors this external reporting requirement. “the sort of thing that these guys would focus on, obviously sales and sales by division, operating profit, profit by division were absolutely key. Breaking out: interest; exceptions; PBT; full-time employees and cash book. If you look at the trading P&L, that gives you a fair view. A nice breakdown of where money's made. The results are pretty clear. But at that time it was a PLC highly reliant on the agency model”. Finance Director, C.

171

The Finance Director notes above the key areas of focus considering it as: “a fair view […] of where money’s made”.

Figure 6.1: Retailer C key retail performance metrics 1998-99

172

Figure 6.2: Retailer C key retail performance metrics group summary

173

2003-2005 ACQUISITION PERIOD

This period relates to the acquisition of the first PLC and a further acquisition with a view to merging them. The key performance metrics reported as being used in this period are shown in Figures 6.3 and 6.4 and consist of a traditional management account columnar format. The Finance Director arrived when the first acquisition had been made and comments on what he saw and considers to be poor analysis with no key performance indicators.

“So when we arrived, I would say there was not particularly a strong set of management accounts. Pretty poor analysis and I think the issues of the day weren't really being represented. No KPIs, nothing really to focus on, other than sales and the rest of it…started to get a little bit better, as we started to work on it.”

No clear metrics are employed, just financial information and numbers supported by some narrative on variances.

So this period could be deemed a step back in the quality of reporting. It would seem as if the reporting reverted to traditional accounting reports, having been acquired and taken private, it no longer had the need to produce PLC standard reporting resulting in a deterioration of management information. As the quote below shows, the business itself was much the same as the pre-acquisition stage i.e. agency (where agents consult customers and take their orders), catalogue stores, stores, delivery business, property and so on yet the reporting became a mass of numbers with a lack of clarity on the key performance indicators. This lack of clarity means the real issues that need management attention are not surfaced. “A very similar shape in terms of agency, [catalogue] stores, stores, business [delivery], property, similar profile of business. If anything, I think the clarity of the KPIs was much worse. There were horrendous schedules, a lot of horrendous schedules like this [Figure 6.4], which the management team would find very difficult to focus on the key issues.” Finance Director, C

174

Figure 6.3: Retailer C key group summary sales 2002-03

175

Figure 6.4: Retailer C key group profit & loss account 2002-03

176

2005-2008 INTEGRATION PERIOD

This period consisted of merging the retail stores and catalogue businesses that had been acquired after obtaining MMC approval. There is therefore a recognition that all metrics cannot be universally applied to all parts of the business as there are disparate operations e.g. stores, catalogue home shopping, delivery, and financial services. “we didn’t have Monopolies and Merger Commission approval to merge the businesses in 2004, that only came in 2004/5. So we weren’t really able to get on with things [including changing the reporting] until 2006 to 8.” Finance Director, C.

The reporting for this period shown in Figures 6.5 and 6.6 suggest the shape of the performance reporting has changed, after the merger and the subsequent consolidation of functions such as Finance, to include some new key metrics, noted below and presented in Table 6.5 for comparison between timeline periods, split into those relevant to the retail side of the business and those aligned to financial services side:  Retail: o On-line Sales % of total sales; o Merchandise margin % (i.e. product gross margin); o Number of customers (‘000); o Spend per customer; o Cost to income ratio %.  Financial Services: o EBITDA o FTEs o Average Interest rate on net borrowings% by securitised and other borrowing; o Customer balances; o Securitisation efficiency%; o Securitisation rate (inc. costs)%; o Net margin/ customer balances %; o Bad Debt/ customer balances %; o Customer penetration – insured balance ration %; o Cost to income ration % The clear development of relevant metrics for various business activities indicates a growing awareness that performance cannot be assessed in the same way across an organisation as the Finance Director explained below. “So we're beginning to focus on, what I call, the core issues. And […] the individual businesses have their own analysis […] and you see the business is beginning to focus in on the core business areas […] we went through the integration, we realized that the financial services business and home shopping business, were joined overhead, and you couldn't look at it as two different divisions […] and those are very big realizations […] So a very different shape of a business in those days”

Here the quote explains the new clarity of reporting, having merged businesses, there are now two divisions; Retail and Financial Services, with each having their own key metrics.

177

Figure 6.5: Retailer C management accounts home shopping 2005

178

Figure 6.6 Retailer C management accounts financial services 2005

179

The integration period enabled the consolidation of various functions such as Finance as well as IT systems and a clearer divisional structure and management teams. There was also a move to one new location for all staff from both merged businesses, which may have helped with the internal communication of information relating to performance measurement.

2009-2013 DEVELOPMENT PERIOD

This period can be seen as a strategic development period in which the metrics are becoming more focused as the future strategy is mapped and rooted within the online brands and business operations. By this period, Retailer C, is now combining the retail and financial services operations for performance reporting having recognised that the business is interdependent (i.e. there are two interlinked facets to the business – one is providing products to customers and the other is providing financing credit to enable the purchase of those products). The more strategically related metrics are finalised as a ‘Performance Dashboard’ in 2010-11 and are shown in Figure 6.7.

The retailer performance metrics areas are clearly identified in the first shaded column of Figure 6.7, in a form of a ‘balanced score card’ (Kaplan and Norton, 1996) covering five areas of focus, that also align with some of the 20 FAs identified in this thesis (i.e. Sales, Customer, People, Growth, Profit and Cash), each with a series of specific and detailed retail performance metrics and mentioned in the quote below:  Key Performance Indicators (e.g. retail sales, customer satisfaction);  EBITDA drivers (metrics that drive profits measured as EBITDA);  Customer metrics;  Team targets; and  Cashflow.

The key performance indicators show individual metrics related to the other four areas of focus so represent the ‘core’ retail performance metrics. So, in Figure 6.7, EBITDA is shown as the top metric and the related EBITDA drivers are noted in the next section of Table 6.7. In the quote below, the Finance Director, gives comment on the ‘Performance Dashboard’ presented in Figure 6.7 explaining the metrics composition and rationale for two areas of focus i.e. customer metrics and team targets. “Then, from a customer marketing type message, we have last season's trading rates, credit recruits, season to dates, that's average order value online, average order value offline, so how much the customer is ordering from us on average, average order frequency, so help them if they're coming back. Demand per trader, so how much was each traded and how much was each basket, how many active accounts there were….From a credit perspective, this is really around the credit customers, how many were true credit customers, opposed to straight active accounts. Gross merchandise margins, operation cost growth.” Finance Director, C

180

Figure 6.7: Retailer C performance dashboard 2010-11

2014+ CURRENT PERIOD

This on-line home shopping business currently operates as a ‘group’ structure. Since the appointment of the new CEO in 2012 the group has been reorganised as one operating entity with one executive board (covering the various online brands) and the financial services business. A strategy director has put the strategy and consistent reporting into place and a performance dashboard with clear metrics to engage and drive the entire organisation in 2010- 11. So internally, compared to previous periods of development, they have greater clarity and since 2010-11 consistency on what metrics are important to them. So the retail performance metrics may have settled, as there is no difference (except the customer satisfaction index has been replaced by net promoter score which does the same thing measuring customer engagement) to the performance dashboard now compared to 2010-11, the retailer continues to refine the internal operations (people, systems, processes) to align with the strategy.

SUMMARY

The exemplar presented as Retailer C clearly shows a change from the original PLC in 1999 to a business in transition to its current state. To make it simpler for readers unfamiliar with this retailer, Table 6.6 is given below, which summarises the key differences and similarities in the retail performance metrics over the time periods discussed in the sections above so these comments are not repeated again. It should be noted that what can be termed traditional monthly management accounts, which include profit and loss accounts, balance sheets and cash flow statements similar to the ARA, are produced as a matter of routine and are therefore always extant as financial measures of the retailer’s performance. The financial measures, therefore have not changed from period to period whereas the non-financial metrics have been honed to reflect the strategy of the retailer as an online home shopping business.

Putting the performance dashboard at the front of the management reporting pack allows the board to concentrate on the key drivers of the business by focusing on the retail performance metrics, most of which are non-financial. The metrics described in the vignette above have changed as shown in Table 6.6 below. The changing metrics, from traditional financial statements to, for example, EBITDA and average order value online, to support the changing business over time, lends credibility to the notion of adaptive resilience in the development of performance metrics for successful retailers.

182

Table 6.6: Retailer C adapting retail performance metrics over time

Pre- 2003- 2005- 2009- 2014+ 2003 2005 2008 2013 Monthly Profit & Loss Accounts √ √ √ √ √ Monthly Balance Sheets √ √ √ √ √ Monthly Cash Flow Statements √ √ √ √ √ Sales by division v budget v prior year √ Operating profit by division v budget v prior year √ Interest by division v budget v prior year √ Exceptionals by division v budget prior year √ Profit before tax by division v budget v prior year √ FTEs by division v budget v prior year √ Return on Sales Financed by division v budget v prior year √ Cashbook Borrowings by month v budget √ Trading Summary of Sales by period, year to date v budget √ Detailed Profit & Loss accounts period actual, year to date √ v budget and variance Retail: On-line sales % of total sales √ Retail: Merchandise margin % √ √ √ Retail: Number of customers (000) √ Retail: Spend per customer √ Retail: Cost to income ratio √ FS: EBITDA √ √ √ FS: FTEs √ FS: Ave interest rate on net borrowing % √ FS: customer balances √ FS: securitisation efficiency % √ FS: securitisation rate % √ FS: net margin / customer balances % √ FS: bad debt / customer balances % √ √ √ FS: customer penetration – insured balance ratio % √ FS cost to income ratio % √ EBITDA (£m) √ √ Retail Sales (£m) √ √ Retail Sales Growth √ √ Credit Mix of New Recruits √ √ Net Cash Flow (£m) √ √ Customer Satisfaction Index √ √ Online Mix % √ √ Natural Search & Direct to URL √ √ Service Level 1 % √ √ Operations Costs (£m) √ √ FS Gross Income (£m) √ √ FS Gross Income Growth % √ √ FS Gross Yield % √ √ Last Season Trade Rate √ √ Credit Recruits (‘000) √ √ Average Order Value - online √ √ Average Order Value - offline √ √ Average Order Frequency – season to date √ √ Demand Per Trader (£) √ √ Credit customers (‘000) √ √ Gross Merchandise Margin (£m) √ √ Operations Cost Growth √ √ Debtor book (£m) √ √ Debtor Book Growth % √ √ Stock Forward Cover (weeks) √ √ Leverage % √ √ Cash Headroom (£m) √ √ Underlying EBITDA √ Net Promoter Score √

183

6.5.2.2. Retailer A

DEVELOPMENT PERIOD

This is an online only retailer and it was formed in 2006 by a husband and wife team (the founders). Previously, Mrs Founder designed and marketed her own range of highly technical and innovative equestrian products and with Mr Founder, who sold them through a predecessor company, set up in 1998. Having gained an insight into textile manufacture and technology, the founders saw the opportunity to diversify into clothing in order to extend the product range. Sales started predominantly through a presence of the business at prestigious British country shows. This was coupled with mail order and telephone sales, which soon developed to online sales. The online clothing sales began to perform well and recognising this shift in buying habits, the founders decided to specialise and focus on this core sales channel.

In the first year it had sales of £0.5m (Financial Year (“FY”) 2007) primarily through country shows and on-line was launched with the marketing effort being predominantly ‘Google Adwords’ selling via eBay. In the following year FY2008, sales reached £1.2m with the company moving to larger premises, employing a dedicated dispatch manager and implementing a Search Engine Optimisation (“SEO”) strategy. In FY2009, sales reached £2m and the company added a head of purchasing, office manager and warehouse manager to the team whilst relocating to larger premises. Sales and marketing activity was enhanced with activity on Twitter and Amazon.com. In FY2010, sales reached £3.4m with the addition of accessory brands to extend the product range. The warehouse was extended to accommodate the range extension and an integrated sales, stock control and dispatch system was developed. In addition, a Facebook page was added to the promotion and communication toolkit. In FY2011, sales were £5.7m with further extension of the clothing range to include children’s wear. In the warehouse a unique barcode system was implemented for picking, packing and dispatch to increase order capacity management. A sales showroom (shop) was developed on site for direct sales and a studio was set up for product development shots for the website. The first catalogue was published and a blog launched. Within five years the business had reached sales of £6.9m (FY2012) with the addition of more brands such as Loake, Zatchels and British Duffle extending the product range. The management team was increased with an e-commerce specialist and system developer. The website was rebranded and re-launched using ‘Magento’ (a website publishing tool) with new features such as upgraded catwalk videos, ‘live’ stock updates and online customer chat facilities.

The growth story information above and the numbers in the table below are not available in the public domain. To get an incisive insight, a researcher needs to get inside the organisation. The company only publishes abbreviated accounts for minimalist compliance with company reporting rules. This means there is no sales, profit or loss information just some balance sheet data filed at Companies House. There was no formal reporting or management metrics. The

184

financial information was compiled by a part-time accountant who undertook periodic statutory reporting duties e.g. VAT returns, payroll administration and year-end accounts.

CURRENT PERIOD

Since raising finance for business expansion, Retailer A has had to undergo substantial change in operational disciplines including the reporting of key metrics. So it has gone from informal family management to a formal structure of recognised metrics for performance measurement embedded within a recording and reporting structure with daily, weekly and monthly performance review meetings. All this change was implemented within a very short period of time. The key metrics being monitored now are noted below in the first column of Figure 6.8 (with the budget and forecast figures).

Figure 6.8: Retailer A retail performance metrics and targets 2013-2017

Budget Forecast Forecast Forecast Forecast 2013 2014 2015 2016 2017 Income Statement Summary £'000 £'000 £'000 £'000 £'000 Current Brands 8,929 10,163 10,742 11,157 11,528 New Brands 360 2,401 5,257 9,100 12,394 Income 'Clicks' 9,289 12,564 15,999 20,257 23,922 Income 'other' 238 238 238 238 238 Total Income 9,527 12,802 16,237 20,495 24,160 Product cost 4,554 5,937 7,343 9,051 10,548 Intake Margin 4,973 6,865 8,894 11,444 13,612 Clicks & Distribution costs 1,917 2,576 3,275 4,134 4,945 Gross Margin 3,056 4,289 5,619 7,310 8,667 Overheads 1,771 2,159 2,340 2,630 2,849 EBITDA 1,285 2,130 3,279 4,680 5,818 Exceptional Items 23 32 - - - Depreciation 41 58 66 81 98 EBIT 1,267 2,104 3,213 4,599 5,720 Product margin % 52% 54% 55% 56% 56% Gross Margin % 32% 34% 35% 36% 36% EBITDA % 14% 17% 20% 23% 24%

SUMMARY

From a retail performance metrics perspective, this retailer has gone through less changes than Retailer C as it has less history, essentially only one change from no reporting (except for statutory report filing) to formal reporting. This formal reporting is summarised in Figure 6.8 above showing the high level financial measures as well as the non-mandatory measures. It should be noted that as a small business Retailer A does not need to deliver Income Statement

185

information to Companies House and therefore all of Figure 6.8 could be considered non- mandatory. As presented in Chapter 5 section 5.6, via the quotes, they now have daily, weekly and monthly discussions around key metrics. This case study evidence supports the notion of retailers adapting metrics as their business changes.

6.5.2.3. Retailer D

BACKGROUND

This retail service business was established in 1869, and although a PLC is wholly owned by a family trust. It started as one shoe repair shop in Manchester and now is a multi-service business operating 1,102 stores of different fascia and shop sizes. Sales by product category now in rank order are from:  Photo Shop services;  Key cutting services;  Shoe repair services;  Watch repair services;  Photo merchandise;  Dry cleaning services;  Shoe care products;  Engraving services;  Photo ID services;  Locksmith services;  Other merchandise;  Pubs;  Keys direct;  Internet; and  Engraving merchandise.

METRICS AND REPORTING

The internal management reporting regarding performance has not changed over the last ten years (and has been extant for many years prior to this) and can best be described as traditional retail, i.e. it relies on manually created weekly cash summaries returned from branches. An example of a weekly cash summary sheet is presented in Figure 6.9. These types of returns were preponderant as operating procedures amongst most UK retailers prior to the introduction of EPoS systems. Retailer D has not changed to an EPoS system and prefers to keep to branch-centred control on a manual basis. This branch-centred approach being part of their ‘upside-down’ management culture. It may also be a way to minimise investment in technology given the number of branches (in excess of 1,000) where automating via EPoS is not going to reduce headcount and save costs since this is a service business.

186

Figure 6.9: Retailer D, retail performance metrics in branch reports 2014

The process for gathering the performance information from branches is given in the quote below. “What happens practically is, our Area Managers will ring our branches on a Friday. We're not driven by technology so we have to write down the sales - so the Area Managers will ring their branches and get an overview of how the week went […] That's collated, and then communicated to the Managing Director […] That's called the Friday flash, so we know exactly where we were up or down on that particular week [comparing to the same week in the prior year]” Operations Director, D.

The reporting is therefore branch league tables by sales which are then aggregated by area and region and nationally. The line items numbered 1-20, on Figure 6.9, may differ between store

187

service type, e.g. shoe repair compared to photo shop, but the focus on daily sales and branch contribution is relentless. This branch control and performance focus and branch bonus system keeps branch management and staff incentivised to exceed targets.

Apart from adding new services and products, the retailer has not undergone any real change in management, ownership or strategy and hence the performance reporting has not changed fundamentally with the exception of introducing ‘new store opening’ reports to monitor a new store’s early progress noted by the Finance Director in section 6.5.1 The performance metrics that keep boardroom attention are:  Cash balance  Weekly sales  Wage percentage (sales to pay ratio for branches)  Bonus earned (primary method to keep branch staff incentivised)  Branch contribution (net branch controllable profit)  Happy Index (annual staff engagement survey)  Colleagues of concern (list of staff not completing their training or performance is not to the standards set)  Net trading profit

SUMMARY

The Happy Index represents a score of employee engagement aggregated from annual upward feedback returns. The colleagues of concern are those employees that are not suited to the business or have not completed their training in the requisite amount of time. The ‘paternal’ management style, top down and the published ‘upside down’ management culture the retailer engenders is different to other retail organisations as it treats branch management as the most important part of the retail organisation and by so doing creates a performance culture centred on the retail performance metrics noted above.

There has not been any substantive change in this retailer’s strategy or operations over the last ten years and consequently no change in the retail performance metrics being monitored.

6.5.2.4. Retailer B

BACKGROUND

This retailer has been established since 1849. It is an iconic retailer, brand and tourist destination based in London. The flagship department store operates out of a listed building over seven floors and is the single largest retail outlet in the UK with about five acres (about 1m ft2) of selling space purveying a wide variety of merchandise with stock keeping units in excess of two million. Its motto is ‘Omnia Omnibus Ubique’ translated as Everything to Everyone

188

Everywhere. Having listed on the London Stock Exchange as one of the first 30 companies in 1889 and former holder of Royal Warrants it now has retail outlets in airports and overseas as well as operating the following business divisions:  International (Retail stores and concessions overseas);  Air (VIP Helicopter and management services);  Estates (Residential sales, lettings and property management services);  Bank (Personal Banking services); and  Aviation (Aircraft handling services at Stansted and Luton Airports). It is currently owned by a sovereign wealth fund that acquired this retailer in May 2010. The current Managing Director has been in post for about ten years and was in situ with the previous private owner.

METRICS AND REPORTING

The current internal management reporting has been in place for many years and specifically since the Managing Director joined this retailer. There is a 103 page monthly management reporting pack. The key metrics reported are:

 Net sales by location group (store, direct, airports, wholesale) and then further breakdown by the 19 departments;  EBITDA by location group;  Summary P&L showing: o Gross Sales: . Own Bought net sales (sales of products the retailer buys) . Concession net sales (sales of products made by concessions) o Net Sales: . Own Bought gross profit . Concession margin . Net margin adjustments o Net Margin: . Selling payroll . Personal shopping payroll . Buying payroll . Management & selling support payroll o Direct payroll . Direct expenses . Marketing & reward card costs o Business Unit (BU) Contribution . Indirect costs o EBITDA

189

 Key ratios: o Own Bought (OB) sales mix o OB Gross Margin % OB Net Sales o Gross Profit % OB Net Sales o Concession Margin % Concession Net Sales o Total Net Margin % Total Net Sales o Selling Payroll % OB Sales o Direct payroll % OB Sales o Total BU Contribution % Total Net Sales o Indirect Expenses % Total Net Sales o EBITDA % Total Net Sales There is a view that the management pack could be improved and simplified by the Finance Director. However the Managing Director takes a forceful view that the detail is necessary to keep the management teams focused and believing that the Managing Director reads all the detail. “We have a very thick pack every month, That's retail. I have to say and I'm really sad because I think a lot of people have moved away from that […] every year we get this, 'Oh, we need to reduce the level of reporting. We need to' "No," I said, […] 'But you may not read all of this.' I said, "Actually, it doesn't matter. He thinks I'm reading it." Managing Director, B.

The Managing Director actually has his favoured metrics which are:  Profit ft2 (different from Sales ft2);  Sell through;  Guest shop;  Staff turnover;  Customer complaints; and  Brand values These favoured retail performance metrics have been described and explained in Chapter 5.

SUMMARY

Given the nature of the retailer and the negligible operational change undergone in the last ten years, as well as the tenure of the same Managing Director, it is not surprising that the metrics and reporting have not changed. The traditional approach of detailed management reporting packs, which could be viewed as stale, seems to be a proven formula for this retailer. In addition, being an iconic and unique retailer affords a degree of protection from economic downturns. As noted earlier in Chapter 5, the management continue to drive performance by traditional retail techniques of ‘authority’ in category, section 5.13 and reclaiming back office space for selling space, section 5.11 and constantly improving customer engagement systems and processes, section 5.5.

190

6.5.3. Summary

Overall the successful retailers that have gone through a transition have suggested that their performance reporting and management accounts have changed as illustrated by the case study Retailers A and C above. The retail performance metrics seem to be directly linked to other aspects of change such as boardroom churn, changing ownership, changing focus of the business, rapid expansion, a shift in channel focus etc. Where there has not been any substantive change the metrics seem to have stayed the same as shown by Retailers B and D.

The exemplar presented as Retailer C, above, clearly shows a change from the original PLC in 1999 to a business in transition to its current state. This retailer’s journey in relation to metrics use and disclosure is discussed in the next section. From the above discussion it becomes evident that there is always dynamism in terms of the fact that the metrics used by this given organisation are always changing. However, the degree of dynamism fluctuates depending on how the business changes in terms of business model and related operations. As such the use of metrics does not always follow a consistent or logical and unidirectional path of incremental improvement and refinement, indeed metric types and their usage can arguably regress as well as progress with the passage of time to become less sophisticated - this is illustrated later. The link between performance metrics and business models and strategy in prior research literature (Collins and Porras, 1996, Kaplan and Norton, 2001, Kaplan and Norton, 2004) has been made but some of the directional changes noted above have been implicit and this thesis makes them explicitly.

Although two (retailers A and C) of the case study retailers confirm to some extent that successful retailers adapt their metrics. Two stand out as exceptions (retailers B and D) and could be deemed to confirm adaptation in that they have not changed so do not need to change their retail performance metrics. This of course could be argued in the opposite way in that successful retailers do not necessarily change their retail performance metrics. It is not possible to conjecture the inverse relationship i.e. that failed and failing retailers do not adapt their performance metrics. The main reason being that there is insufficient evidence available. Both Retailer E and F being in varying degrees of Administration and Insolvency mean that internal management information is unavailable for public or academic scrutiny.

What is known from the case study data, is that the two failed retailers, failed due to what has been designated, by this thesis, the ‘business model’ in the FAs where the structure and strategy determined or controlled performance capability. Retailer F, failed as it had the cash stripped from its retail operations, twice (once through rental burdens and then to feed a sister (FUK) company) by the Group Head Office, and this cash (daily cash balance noted as a key metric in Chapter 5) was applied elsewhere in the group and not invested in its retail operations. Retailer E was tasked with generating cash by its private equity owner and was also in fierce

191

competition with its only four suppliers. When the supply contracts were withdrawn the company failed.

Although a possible conjecture may be that if these failed retailers, had appropriate metrics, these metrics were insufficiently useful to produce ‘early warnings’ or ‘red flags’ or if they did produce warnings, these may have been ignored by the ownership structures as explained in the preceding paragraph. This may have implications for future ‘failure’ research to consider business model dependencies.

6.6. Being trust intelligent

6.6.1. Introduction

Implicit in gaining trust in an environment of disclosing information such as performance metrics, to all stakeholders, is the element of transparency (ICAS, 2010, ICAEW, 2013, IIRC, 2015) in corporate reporting. “Organizations use a range of mechanisms to enhance credibility and trust, of which assurance is one, and it is clear from the debate that because Integrated Reporting (IR) is relatively new and still evolving, assurance on IR will need to evolve alongside the practice of reporting itself.”.(IIRC, 2015)

Integrated Reporting (IR) is a new global reporting framework produced by the International Integrated Reporting Council (IIRC) being recommended for adoption by many of the national accounting bodies as ‘best practice’. The long term aim of the IIRC is to promote IR which combines all sorts of information including financial and non-financial so that it provides all stakeholders with a clearer understanding of how an organisation creates value. This value creation focus is seen as the next stage in corporate reporting by the IIRC.

It is suggested that retailers use many (hundreds) metrics to manage performance to varying levels of operational detail (Walters, 1977, Walters and White, 1987, McGoldrick, 2002, Fernie and Sparks, 2014), but which of these should they actually share outside the board room and put into the public domain is a challenge to determine. This journey of awareness (attitude) and disclosure practice is considered below.

6.6.2. A framework for understanding metric disclosure practice

Increasing the transparency of metric disclosure practice aspect to build trust can be viewed through the capability or competence of the retailer in making quality disclosures, and this can arguably be assessed through a series of steps ranging from highly competent to incompetent. Such competence can then be aligned with the level of awareness or conscious use of performance metrics to impart performance information.

192

There may also be situations were retailers are unaware they are using something as a metric, in which case it is more difficult to disclose consciously something about which they are only unconsciously using. For instance, supplier income which includes items such as ‘retro-credits’ based on product turnover. Retro-credits are essentially product discounts applied retrospectively by the supplier to invoice price on the achievement of target volume sales. Negotiating supplier terms including retro-credits has become an intrinsic part of a retail buyer’s role such that it is now custom and practice. Thereby becoming either conscious or automatic and hence an unconscious part of daily routine.

When considering the disclosure of non-financial metrics, the attitude and practice of disclosure can be categorised into a four step framework which maps the journey to becoming ‘trust intelligent’. The framework, created and illustrated in Figure 6.10, provides four process steps related to capability development (Humphrey, 1989, Curtis et al., 2009), and this has been aligned to awareness stages or states of conscious behaviour (Geller, 2002). The framework proposed shares some similarities to the capability maturity model, originally used in the design of software development, which shows five stages and four process steps (Humphrey, 1989). This has been adapted to improving the competence in the disclosure of non-financial performance metrics, on the basis that disclosure will lead to greater transparency which will in turn improve the understanding of trading performance.

Stage one represents limited disclosure of non-financial performance metrics in terms of disclosing performance information. How the case studies map against these stages is illustrated later on. Stage two represents a better knowledge of non-financial performance metrics but an inconsistent and restricted use of these to disclose performance information. Hence capability has improved but practice is restricted. Stage three represents greater knowledge and hence capability and the use of the non-financial metrics is selective in the disclosure process. The final stage is the greatest knowledge demonstrated by open and automatic disclosure of non-financial metrics. This is overlaid with the levels of awareness and learning represented by conscious and unconscious behaviour from psychology modelling (Geller, 2002).

Consequently the quality of disclosure and resulting transparency has been put into the four stages, for ease of description, in the framework representing a journey noted below:  Unconsciously incompetent;  Consciously incompetent;  Consciously competent; and  Unconsciously competent.

The framework created in Figure 6.10 is illustrated below and using a matrix presentation in Figure 6.11. The mapping of the case studies is done in section 6.8 but for now, each stage is described below.

193

Figure 6.10: Being trust intelligent through retail metric disclosure

Quality of disclosure practice

Open

Selective

Restricted

Transparent

Hubristic

Limited

Minimalist

Unaware

Unconsciously Consciously Consciously Unconsciously Trust Incompetent Competent Competent Incompetent Intelligence

The journey to becoming Trust Intelligent

Unconsciously incompetent describes automatic (in the psychological processing sense) behaviour and is considered an at-risk habit and hence the quality of disclosure is limited. Consciously incompetent is a recognition (either internally or they have been made aware) by the retailer that the disclosures and hence transparency is restricted and far from ‘best practice’. Consciously competent is the stage where (either internally or with external assistance) efforts are made to improve transparency and disclosure quality but it is still overtly controlled and essentially selective. Unconsciously competent represents the stage where the quality of disclosure is transparent and this is natural and an automatic habit within the retailer and there is no desire to keep secrets about performance, representing an open approach to information sharing.

194

Figure 6.11: The conceptual framework journey as a matrix

Conscious

Minimalist Hubristic

Incompetence ‘Ringfence’ Competence

Unaware Transparent

start stop

Unconscious

6.6.3. Reading the conceptual framework

The framework has been conceptualised to show a direction of travel, through four steps, that eventually leads to a transparent way of disclosing non-financial performance, as a journey to becoming trust intelligent. The first point to make is that although four steps are shown for ease of description, in practice there may not be such a rigid framework. With retailers potentially straddling the steps and indeed moving back and forth depending on the circumstances see section 6.8. A key point is that the logic of the framework shown in Figure 6.10, means that a retailer can only straddle two adjoining steps such that being ‘unaware and ‘transparent’ at the same time is not possible. This is shown by the forked axis for unconscious and the ‘ringfence doughnut’ in the middle of Figure 6.11 to indicate the centre cannot be straddled i.e. a retailer cannot be in all four stages at once. The same case study participants A, B, C and D can be used to illustrate the framework and evidence the stages as these represent different stages in the case study companies’ performance measurement disclosure journey. It should be noted that in the illustrations, the historic view traces back ten years only (except retailer A: 2006-2013 only) and these case companies may have been at different stages in periods prior to the illustrations given.

195

The vertical axis represents the type or quality of disclosure practice. Four steps of quality improvement at disclosure are noted: Limited; Restricted; Selective; and Open. Along the horizontal axis which represents increasing trust intelligence, the improvement in competence is demonstrated through moving from step one being unaware (hence unconsciously incompetent), where very little is disclosed and is termed ‘limited’, to step two being minimalist hence conscious of the disclosure but incompetent in relation to transparency and building trust such that retail performance metrics are withheld and hence ‘restricted’ in disclosure. Step three is described as hubristic where the retailer is competent but makes disclosures ‘selectively’ to provide a controlled view of performance. The fourth and final step is transparent representing an ‘open’ approach to disclosure that is automatic and inherent in the retail business. Not all retailers may necessarily wish to be at step four, preferring to keep certain metrics secret thereby believing they have competitive advantage. Although regulators and academics may desire transparency as it allows for a better understanding of actual performance and an increase in the trust of the performance reported. The four stages of development have been ascribed labels to help with understanding the approach of the retailer to making performance disclosure and the quality of the information presented through the metrics reported each stage is described in more detail below.

6.6.4. The four stages of the journey

6.6.4.1. Unaware

In the unaware stage it is proposed that the retailer is classified into a group where there is a lack of knowledge about retailer non-financial performance metrics and what is appropriate to disclose. At this stage the retailer is essentially still learning about the metrics and what is relevant over and above the mandatory financial information. “They are going from a company that didn't have board meetings, that didn't have management meetings, that had completely unstructured entrepreneurial drivers, to a discipline [of daily, weekly, and monthly meetings discussing performance reports and metrics] […] a classic environment change.” Chairman, A.

So there is inexperience (a lack of capability or competence) and a lack of awareness about retail performance metrics with zero disclosure of non-financial performance metrics.

6.6.4.2. Minimalist

In the minimalist stage it is proposed that the retailer recognises (is aware of) the performance metrics and may also be aware of best practice but chooses to only disclose what is necessary to meet mandatory requirements and present nothing more i.e. adopting a restricted approach. This has been the stance taken by Retailers A, B and C. “at the moment we're on the cusp of still being classed as a small company,[…] So as far as published information is concerned, we're still with abbreviated accounts […] in fact, the information that's in the public domain is very limited, and deliberately so.” Finance Director, A.

196

“I have worked for PLCs and private companies, and regulatory reporting in private companies does not drive any decision making whatsoever […] Private companies don't have that pressure and can post their results nearly a year after they've finished the financial year. Therefore there's no real focus on them and they can be as private as they want” Finance Director, B.

As the quotes above show there is clear awareness but also a lack of appetite in disclosing anything more than is needed to keep the retailers compliant with the mandatory regulations. Both quotes indicate a level of inertia in making better disclosures, an unconscious acceptance of the rules and compliance with them. If the rules force better disclosure then they may comply but are unwilling to go further voluntarily.

6.6.4.3. Hubristic

At the hubristic stage the retailer knows about the performance metrics and is making conscious effort to provide information but in a selective way. For example, in one of the case studies, Retailer D, the focus of performance metrics was more for the benefit of employees in the annual report and hence the numbers and the narrative were controlled. This numbers and narrative management is best illustrated by reference to Figures 6.12, 6.13 and 6.14 which show six out of 14 pages from an Annual Report produced in A5 colour format. These pages have grayscale and redaction applied for inclusion in the thesis. The pages shown are:  Chairman’s Report;  Financial Performance;  Big numbers;  Training Report;  Benefit Statement; and  Shops and Sales. The Chairman’s report covers the years activity and provides a positive spin even on shoe repairs which have a 6% drop on sales. “The 6% drop in shoe repairs is mainly due to a lack of rain, rather than any fall in the number of customers” Chairman, Annual Report 2013 (Figure 6.13).

The Financial Performance report is a simplified Profit and Loss Account showing sales by activity; major costs summarised and investment. The ‘Big Numbers’ page highlights activity in quantitative pictogram format shown in Figure 6.13. The Training report as expected from a retail service company concentrates on the colleagues and training delivered. The Benefits statement emphasises benefits to colleagues and the public as well as tax paid and also in Figure 6.14, there is a summary of shops and sales by category. The case study data indicates that the modus operandi in this stage is to be economical with transparent disclosure and spin the messages in the favour of the organisation. In the case of retailer D it is essentially simplified to meet the messaging strategy to keep the store branch employees informed.

197

Figure 6.12: Retailer D annual report extract one

Figure 6.13: Retailer D annual report extract two

198

Figure 6.14: Retailer D annual report extract three

This style of reporting shown in Figures 6.12, 6.13 and 6.14 above has been termed hubristic as it aims to convey a particular view of performance. More information is provided than mandated by regulatory reporting rules, so the retailer has moved from the Minimalist stage discussed earlier, driven to a large extent by being a PLC. As noted in the quote below, selective metrics, see Figure 6.13, are used to manage the readers understanding. “In the annual report […] you'll see […] we've got the financial numbers, top level, and we talk about what's been happening. This report is sent to every colleague in the business. The external people see it, but it's not produced with them in mind […] what’s filed at Companies House is much more regulatory compliance. If I did enough to spend 50 grand a year doing an audit, I won't. Biggest waste of money every year. Absolute waste of money and Deloitte’s do it and I've told them that, they never find anything. We only have one owner, we don't have any debt - there's not really any risk there. But we’re a PLC […] we have to do it because of that.” Finance Director, D.

The above quote also implies an element of minimal compliance. A reluctance to report anything other than the minimum required by mandatory reporting yet being a PLC forces additional reporting requirements. There is therefore a possibility that some retailers could actually bridge the two steps and be part minimalist and part hubristic. As noted earlier, the four step journey has been conceptualised to illustrate key points on the journey to becoming trust intelligent and it is recognised that the steps imply rigidity and a unidirectional trajectory but this may not be the case in practice as illustrated in Figures 6.15 to 6.18.

199

6.6.4.4. Transparent

It is suggested that ‘transparent’ is the last stage, and a logical extension of the journey improving awareness and capability. In this stage, the disclosure of non-financial metrics to support the performance information should be clear and consistent thereby automatically providing greater clarity. This clarity engendering trust in the numbers and narrative. None of the case studies has reached this open quality practice and hence transparent stage. As a conceptual framework, this represents the final position of unconscious competence depicted in the model in Figure 6.11 even though none of the case studies have reached this position and the quote below implies that neither have any other retailers. As mentioned earlier some retailers have no desire to progress beyond minimalist. As noted above, even public companies have a preference to control the view they present and tend to be at the ‘Hubristic’ stage. This opaque ‘hubristic’ approach used by the large UK retailers is of concern to regulators as noted in the quote below. “Transparency of information, in the form of clarity and consistency in results and reports, is increasingly in demand both from regulators and markets […] The key performance indicators used in the retail sector remain problematic.” (ICAEW, 2014, p3)

The quote above is referring to the extensive use of a single metric LfL sales, which the regulators acknowledge is inadequate for making comparisons across the retailers since different retailers calculate the metric using different definitions and assumptions.

6.7. An alternative approach to narrative analysis

As referenced earlier (Beattie, 2002, Beattie et al., 2004, Hussainey et al., 2003, Schleicher et al., 2007) using statistical models and regression analysis in particular has been the preponderant method for the analysis of narrative information disclosure in corporate reporting. This is an empirical deductive approach (and a positivist philosophical stance) to capital market based research where a link is hypothesised between the accounting information published and, for example, share price movement.

The challenge with most of these quantitative model types of assessment are that they may or may not produce correlations, between the data counted in the narrative statement with the market price movement, yet they rarely deliver a causal model i.e. proof that the data disclosure actually caused the share price movement. Whereas in a discrete event such as the Tesco accounting scandal disclosure in 2014 (Sparks, 2014) it is clearly evident that this caused a share price movement. So the disclosure of a non-financial performance metric, (i.e. one of 20 proposed in this thesis), evidences a share price movement and can be considered a direct causal link. This researcher suggests that the fundamental issue with these quantitative models is that they are being built on subjective attributes which render them inherently weak when applied to narrative statement analysis of a qualitative nature.

200

In one of the papers for example (Hussainey et al., 2003), that uses text based searches via a computer model, in the conclusion it states: “We believe that the process of analysing narrative disclosure has scope for further refinement. At the moment our methodology equates disclosure quality with the amount of information provided. This is done by counting the number of text units with certain themes like profits, turnover or costs.” (Hussainey et al, 2003, p294).

In a more recent paper that uses text based searches the researchers (Schleicher et al., 2007) acknowledge the problem of the methodology but continue with the design as noted in the quote below. “simply adding up the number of statements in a mechanical way is likely to lead to a noisy measure of disclosure quality […] Unfortunately, avoiding this type of noise would require a completely different research design, one that is based on labour- intensive reading.” (Schleicher et al, 2007, p158).

The purpose of the narrative statements is that they should be read so this method of discourse analysis whereby count frequency is equated to quality is clearly somewhat limited and could be described as inordinately speculative and positivistic in nature (Glaser and Strauss, 1967) which does not fit well with the fact that narrative disclosure is a communication of performance and quality, open to an often subjective selection and judgement of metrics by the sender, and their subjective interpretation by the receiver. The framework outlined in Figure 6.10 above, although conceptual and it could benefit from further independent validation, does provide a different and qualitative way to conceptualise narrative in corporate reporting of a retailer’s performance.

6.8. Mapping the case studies

Taking each of the case studies and mapping their current position on the journey shows:  Retailer A: has moved from the unaware stage to minimalist (Figure 6.15);  Retailer B: has been at the minimalist stage for some time (Figure 6.16);  Retailer C: for the journey, see Figure 6.17, is now at the minimalist stage;  Retailer D: has been at the hubristic stage for some time (Figure 6.18);  Retailer E: was at the hubristic stage before falling into administration; and  Retailer F: was at the hubristic stage before becoming insolvent.

As a reminder, this is a mapping of their disclosure of retail performance metrics in the public domain. Retailer A is a relatively new company so it’s trajectory on the journey framework can be seen from when it first established in 2006. Whereas the other retailers B, C and D are mapped from the current position and traced back in a ten year history for this mapping exercise. There are insufficient records to trace back to when all the case study retailers were first set up (i.e. 1849; 1890 and 1869 respectively) and to follow their entire historic trajectory. Retailers E and F have failed and their internal management records are not accessible. This mapping is presented in a matrix format to provide clarity on the ends of each axis: incompetence to competence and unconscious to conscious.

201

Figure 6.15: Retailer A, journey to becoming trust intelligent.

Conscious

Minimalist Hubristic

2013 Incompetence Competence

Start-up 2006

Unaware Transparent

Unconscious

Figure 6.16: Retailer B, journey to becoming trust intelligent

Conscious

Minimalist Hubristic

2005-2015

Incompetence Competence

Unaware Transparent

Unconscious

202

Figure 6.17: Retailer C, journey to becoming trust intelligent

Conscious

Minimalist Hubristic Current

2009-2013

PLC-2003 2005-2008 Incompetence Competence

2003-2005

Unaware Transparent

Unconscious

Figure 6.18: Retailer D, journey to becoming trust intelligent

Conscious

Minimalist Hubristic

2005-2015

Incompetence Competence

Unaware Transparent

Unconscious

203

The journey for Retailer C reflects its business development stages in Figure 6.17 above and shows more movement aligned with the amount of change undergone. Although there may be a desired pathway showing increasing capability, it is clear that the trajectory or direction of travel is not one way, circular or straightforward and as shown by retailers B and D may be static.

What has been proposed is a new conceptual framework for viewing non-financial (retail metrics) in corporate reporting based on the case study data. It has been created from a blend of unrelated previous models in the literature (Humphrey, 1989, Geller, 2002) and the new framework is designed to understand the journey retailers might take to becoming trust intelligent i.e. gaining trust through transparent disclosure of performance metrics. This is a different approach to the inconclusive quantitative models that are extant and replete in the literature.

6.9. Summary

The discussion of the results has presented five salient dimensions or viewpoints arising from the case studies. The first viewpoint discussed was presented in the thematic analysis in Chapter 5 which distilled 20 FAs that are considered sufficiently important for board room attention. Given the quantity of metrics available to retailers, a tool was created, also presented in Chapter 5 with a recognition that there is a need for a method to sift the metrics into clusters such as good, better and best. Using attributes for quality and resilience a qualitative framework was proposed. Accepting that both quality and resilience are subjective terms reducing the method to different scientific verification such as quantitative testing is unlikely to deliver an improvement in analysis. However, the use of expert focus groups or Delphi panels may deliver a more consistent set of clusters than currently done by one researcher.

An attempt has been made to divine one metric for each of the FAs using the sifting method with a particular emphasis on external disclosure. This has been done with a viewpoint that should these metrics be presented as a suite of performance indicators by channel (physical and digital) with comparatives then it may give a better understanding of the retailers’ underlying performance. Within this 20 twenty ‘vital few’ suite of metrics, a consistently persistent ‘core’ of six metrics was avidly discussed with the researcher and mentioned and used by the case study participants over a long period of time. These ‘six of the best’ are presented as a starting point for the debate on setting a standard for retail metrics disclosure.

The case studies revealed that retailers are on a journey, which is illustrated in Figures 6.15 to 6.18 with Retailer C as an exemplary vignette, and its journey has been shown as not straight forward. Two features of the journey came to the fore. First that they adapt their metrics as the business changes lending credibility to the notion of adaptive resilience in relation to retail metrics used. The second was a desired trajectory on the journey to becoming trust intelligent.

204

A framework is proposed for analysing where a retailer sits in the four steps to understanding their non-financial performance disclosures.

There is ample scope for this research to be replicated amongst a different set of retailers in the UK and overseas to validate the research results and test the sifting method and journey framework presented.

There is also a need for consistency in the definitions and calculation methods for some of the metrics, particularly non-financial metrics such as LfL sales and Sales per ft2. This would enable better comparability of retailer performance which at present is based on inappropriate comparisons as analysts and commentators are ignorant of the fact that these metrics when presented by retailers are calculated differently. In spite of a demand by regulators for more integrated reporting the regulators have not stepped forward to create consistency in the non- financial metric reporting. There is therefore a risk that any integrated reporting may become more inconsistent and create more confusion than clarity.

205

Chapter 7: CONCLUSION

7.1. Reflections on the thesis

7.1.1. Introduction

This chapter summarises the thesis by considering the aim and objectives of the research and how these have been met. This is followed by a reflection on the literature review in Chapters 2 and 3. In addition, performance frameworks, performance reporting, marketing metrics and retail performance metrics literature research is reviewed for relevance in supporting the thesis objectives. The methodology and methods used is considered in relation to the key findings revealed which are recapped. This chapter ends with concluding remarks on the value and contribution of this thesis, and some thoughts on further research.

7.1.2. Aim and objectives met

The aim of this thesis was to work towards an understanding of UK retailer performance. It has been shown that the de facto standard of ARA and trading statements made by retail companies are insufficient to provide a comprehensive understanding of a retailer’s performance and hence better ways to assist with understanding retailer performance have been explored through addressing three objectives, as noted below:

 To identify what retail performance metrics are used by retail boards to manage their performance. In so doing: o To identify any commonalities amongst the performance metrics used by retail board directors; and o To determine whether or not retail performance metrics change over time.  To identify what retailers claim about their performance in the public domain.  To explore any disconnect between the two objectives above i.e. the connectivity between the performance metrics retail boards’ use and those they publicly report.

Each of these objectives is discussed in turn below.

7.1.2.1. Objective 1: Identify retail performance metrics

A key part of the thesis was the search for retail performance metrics actually used and considered important. Much academic literature consists of retail metrics at an operational and detailed level. Most of this is from case study research where there is little insight into what retail boards consider important and actually use.

206

The retail performance metrics uncovered by the thesis research show a abundance of metrics currently in use by the boards of case study companies. These are summarised in Chapter 5 in Tables 5.7 and 5.8. It is also clearly shown that the boards of the retail case study companies concentrate on 20 FAs that they consider important for a holistic view of their retail company performance management. These FAs are supported by differing numbers of retail performance metrics and the retail performance metrics actually used vary from retailer to retailer. For instance in measuring customer engagement a variety of metrics are employed such as: net promoter score; rant and rave; google analytics, trustpilot review; customer feedback index; guest shop; and CLV.

The thesis also revealed that the development of the retail performance metrics lacks guidance on definitions, calculation bases and certainly there is no mandatory requirement to disclose any of these. Consequently, the retail performance metric calculations have been done differently by the case companies even for the same metric such as LfL sales and Sales ft2, making retailer performance comparison problematic.

7.1.2.2. Objective 1, sub-objective 1: search for commonality

There is evident commonality in the FAs as well as certain preferred metrics. The FAs in themselves provide substantive evidence of commonality as areas considered important for the boards’ attention. Given the abundance of metrics a sifting matrix is devised as a mechanism to get to ‘the vital few’ metrics which were preponderant i.e. 20 key retail performance metrics that map to each of the 20 FAs in Chapter 5 and section titled the Twenty 20 in Chapter 6 and presented in Table 6.4.

In addition there are a core element within this 20 presented in Table 6.4, referred to as ‘six of the best’ retail performance metrics. These six seem to have been used consistently and persistently throughout the ten year period of each of the case study companies’ histories. These are presented again below as they could form the basis of an industry standard for reporting retail performance metrics.  Average weekly transaction numbers;  Average weekly transaction spend;  LfL sales;  Sales ft2;  Wage %; and  Staff retention %. Overall the 20 FAs in themselves and the retail performance metrics clustered around the 20 FAs provide a contribution to the understanding of what retail boards see as important and employ to manage their retail company performance.

207

7.1.2.3. Objective 1, sub-objective 2: Adaptation of metrics

The notion of adapting metrics was explored (see Chapter 6). Retailer A and C of the case companies, both of which had undergone organisational change during the research period studied, showed evidence of changing performance metrics too. Retailer A, moving from an informal family business to an investment-backed formal retail board structure, with new disciplines of performance reporting. Retailer C moved from a PLC to being privately owned, and then merged with another former PLC, which had been taken private. The merged businesses were then transformed with substantial divestment of stores and an integration of both businesses (including, call centres, warehouse, distribution and logistics businesses, property, head office functions and IT infrastructure) to refocus home shopping and become online only. As a result of all this change both retailer A and retailer C adapted their retail performance metrics to reflect the new organisational business models. This evidence is illustrated by interviewee comments as well as company provided information lending support to the notion of adaptive resilience.

For the other two successful retailers no change to the strategy or business model was evident, and this was mirrored by minimal change in the retail performance metrics they use. This could be deemed to support the notion of retailer change reflecting an adaptation of metrics, i.e. if the business model and strategy is successful, there is no need to change it consequently there is no need for adaptation of retail performance metrics.

7.1.2.4. Objective 2: Identify what retail boards publicly disclose

The second objective explored the connectivity between what retail performance metrics retail boards used and what they actually reported in the public domain. This is covered in Chapter 6 under the retailers’ journey to being trust intelligent. A conceptual framework, in the form of a ‘journey matrix’, has been proposed that shows a retailers awareness and competence at disclosing non-financial information following a four-step process of Unaware, Minimalist, Hubristic, and Transparent. The journey is about disclosing retail performance metrics to provide an understanding of performance to gain stakeholders trust. Hence the description of a journey to becoming trust intelligent.

7.1.2.5. Objective 3: Exploring any disconnect

The journeys are mapped for the four retail case companies in Chapter 6. It is evident that companies do not follow a unidirectional path and also that most prefer to adopt a minimal compliance approach. This suggests a tension between the stated intent of retailers for being transparent and wanting to be trusted and the actual practice of reporting disclosure. The regulators (ICAEW, 2014, IIRC, 2015) also want more transparency but have still to provide defined guidance on retail performance metrics disclosure.

208

7.1.2.6. Summary

Overall, the three objectives noted above, have been met by the findings from this research. To some extent this was inevitable due to the research design being informed grounded theory discussed in Chapter 4, whereby it is a discovery of what is said and done rather than a testing of hypothesis. However, given that this discovery has not been done before, for UK retail board room performance metrics, then the objectives being met provide useful insight and a contribution to academic research.

7.2. Theory, knowledge and contribution

Much research has been done in the broad area of corporate performance (Neely et al., 2005, Berry et al., 2009, Mellahi and Wilkinson, 2010) and business strategy such as competitive strategy and advantage (Porter, 1980, Porter, 1985). There has also been substantial publications guiding management to consider non-financial performance metrics through for instance the use of a balance score card (Kaplan and Norton, 1992) and strategy mapping (Kaplan and Norton, 2004, Kaplan and Atkinson, 2015). All of the prior research on performance concentrates on internal management. This thesis builds on the idea of non-financial performance metrics using UK retailers as a focused example and asserts the need for public disclosure for retail performance metrics used by retail boards. The public reporting of these retail performance metrics creates transparency which in turn engenders trust in what is reported and hence creates the explicit link to good governance implicit in stewardship theory. The current de facto standard of performance reporting resides within the ARA which has been shown, using examples of Woolworths and Tesco, to be problematic in providing an understanding of a retailer’s trading performance. Similarly, the mandatory reporting rules for corporate governance, internal controls and risk are consider to be poor illuminators (Abraham and Cox, 2007): ‘Turning to internal control risk reporting, it may be that some aspects of regulation inadvertently serve to censor firm-specific variation in reporting. The approach to statements on internal control has been likened to ‘box-ticking’, and the language ‘boiler- plate’, both of which suppress known idiosyncrasies in existence across firms that users of accounts are assumed to want highlighted’ (Abraham and Cox, 2007, p244).

By the transparent disclosure of retail performance metrics, retailers can break away from ‘boiler- plate’ and demonstrate their fiduciary duties inherent in stewardship theory through good practice reporting. As noted earlier in Chapter 3, there is a desire amongst regulators for a suite of retail performance metrics (ICAEW, 2014, IIRC, 2015). The Tesco accounting scandal has shown (supplier income disclosure is non mandatory) how supplementary information can impact share price. Indeed, the recent Volkswagen ‘diesel emissions’ metric manipulation has had a dramatic impact on the Volkswagen share price. This is an industry metric that is one of many indicators of a product’s (car) performance. It is conjectured that there seems to be a failure in compliance within the FA, as identified in this thesis, of risk and regulation where

209

Volkswagen are concerned. Looking further afield, the healthcare sector, including the NHS is a good example of where non-financial performance metrics have been extensively defined, for example waiting times for diagnostic tests. This guidance is detailed by test with how to calculate the wait time. Although constant revision to the calculation bases can create tension and confusion about performance. For example, the calculation and reporting of waiting times is difficult to comprehend and has been frequently changed the most recent iteration being in 2015 (NHS-England, 2015).

Getting to a suite of retail performance metrics is possible although there is a need for the regulators to clarify; definitions; calculation bases; and periodic reporting. This thesis provides a starting point for a suite of metrics having identified the ‘vital few’ i.e. 20 retail performance metrics aligned to 20 FAs and within this, the consistently persistent ‘six of the best’ retail performance metrics.

A conceptual framework has been created to assess a retailer’s competency in transparent disclosure as they journey towards becoming trust intelligent. This framework underpins the theoretical link asserted above of retailer performance metric disclosure creating transparency and trust and hence explicitly enhancing good governance practice implicit in stewardship theory.

The purpose of the thesis was to work towards an understanding of retailer performance. First, it shows that to put a retailer’s performance into context requires an understanding of the UK retail landscape as a prerequisite. . Theories and models of retail change were examined. They were identified as being either too generic e.g. the business lifecycle (Schumpeter, 1980) or were easily falsifiable such as the wheel of retailing (McNair, 1931, McNair and May, 1978). Hybrid models of retail change (Brown, 1987, Brown, 1991, Levy et al., 2005) were also considered, but were suggested as inadequate as they assumed that all retailers will want to move towards the mass market. The concept of the mass market has been scrambled by the advent of the digital age and retailers that offer exclusive luxury do not readily fit into any of the hybrid models.

Most of the theories and models of change were conceived in the days of bricks and mortar retail and are considered ‘analogue’ whereas we now live in a digital world rendering them outdated. However to get to a holistic view, and building on the ‘macro micro’ (Hollander, 1960) notion, an alternative observational approach to understand the current UK retail landscape was taken by reviewing a 25 year history to 2015 searching for Mega trends and Paradigm shifts as the driving forces of change. Due to the interdependent and multi-relational aspects of the driving forces, unravelling the complexity is referred to as ‘deconstructing the kaleidoscope’. A six faceted viewpoint is proposed where Mega trends and Paradigm shifts are concerned, these being:

210

 Regulation;  Global influence;  Socio-demographic;  Physical dynamics of stores and supply chain;  Digital dynamics; and  CLV.

One idea that did hold some appeal was the notion of adaptive resilience (Wrigley and Dolega, 2011, Wrigley and Brookes, 2014) which had been proposed by reference to the UK high street. This has been explored in relation to retail change and the adaptation of retail performance metrics where there is some evidence from the case studies showing that retailers adapt their performance metrics as their business models change.

In summary, the case study analysis provided five overriding dimensions:  The assimilation of the abundance of retail performance metrics into 20 FAs;  Using the sifting matrix to select one retail performance metric for each FA as a means of building a picture of ‘the vital few’ metrics that retailers use;  The emergence of a core of ‘six of the best’ consistently and persistently used retail performance metrics;  Evidence of successful retailers that undergo change adapting their retail performance metrics and supporting the notion of adaptive resilience in retail performance metric use;  The conceptualisation of a four-step process whereby retailers can be mapped on a journey to becoming trust intelligent through the disclosure of non-financial performance metrics. A large amount of rich data was collected from the case companies, experts and through secondary research. With the aim and objectives front of mind during the analysis process, five salient dimensions have been noted each adding to knowledge about retail performance metrics and what retail board members favour and what they choose to disclose in the public domain. Indeed, the FAs may present a novel way of exploring a retail organisation given that this is how retail boards view their business. This is different from the traditional functional or operational view typically taken by researchers.

The multiple retailer case study findings show that retailer board members actually use an abundance of metrics to manage their retailer’s performance yet they do not disclose these metrics in their public reporting of performance. There is one retail metric, LfL sales that is extensively used and reported. However, the underlying assumptions and calculation basis vary from retailer to retailer and sometimes year to year. In fact there is no agreed or approved standard for reporting non-financial retail performance metrics. In short, as the reporting of retail performance metrics in general is non-mandatory, their use varies from retailer to retailer as does the underlying assumptions and calculations. With increasing demand for reporting of

211

industry metrics by regulators, but no guidance on their calculation, presentation or use, as there is for mandatory financial information, there is a clear and present danger that the voluntary reporting will increase complexity and further confusion when trying to understand a retailer’s trading performance.

7.3. Thoughts on further research

The thesis was conducted using four successful retail companies and two failed companies to identify metrics actually used by retail board members. The research sample could theoretically be expanded to the top 200 retail companies representing 95% of the population by turnover. Similar research could be undertaken in overseas countries to confirm or refute the findings dependent on the regulatory reporting regimes. Now that there is some body of evidence, this could be used to conduct survey style research although it is acknowledged that getting retail board members to complete surveys can be problematic.

A specific focus on digital or on-line business research could be undertaken as there was some difference in the metrics used. This difference was not sufficiently salient for this thesis as it aimed for commonality of features but it is acknowledged that with a different set of objectives other non-financial metrics related to digital retail businesses may come to the fore.

Another aspect of further research is to consider the failed retailers in more detail and evaluate the business models and retail performance metrics used. It was unclear from this research whether or not retail performance metrics were used and did not raise red flags or if the red flags were overridden by strategic (albeit ultimately incorrect) decisions. Given the legal administrative position of the retailers at the point of time of the research, it was not possible to get to internal company documentation.

The thesis shows that retail performance is better understood by a focus on retail performance metrics and with appropriate boardroom attention on the best metrics can make the difference between success and failure. This thesis identifies a suite of metrics as a basis for providing guidance on a possible industry standard. Furthermore retailers should disclose these best metrics if they are to become trust intelligent.

212

REFERENCES

AAA 1969. American Accounting Association Report of Committee on Managerial Decision Models. The Accounting Review, 44, 43-76. AAA 1974. Report of the Committee on Concepts and Standards-Internal Planning and Control. The Accounting Review, 49, 79-96. AAKER, D. A. 1991. Managing Brand Equity: Capitalizing on the Value of a Brand Name, University of California, Free Press. AAKER, D. A. 1992. The Value of Brand Equity. Journal of Business Strategy, 13, 27-32. ABDALLAH, C. & LANGLEY, A. 2014. The Double Edge of Ambiguity in Strategic Planning. Journal of Management Studies, 51, 235-264. ABRAHAM, S. & COX, P. 2007. Analysing the determinants of narrative risk information in UK FTSE 100 annual reports. The British Accounting Review, 39, 227-248. ACS 2014. The Local Shop Report. In: ACS (ed.) The Local Shop Report. www.acs.org.uk: Association of Convenience Stores. AGARWAL, V. & TAFFLER, R. 2008. Comparing the performance of market-based and accounting-based bankruptcy prediction models. Journal of Banking and Finance, 32, 1541-1551. AGARWAL, V. & TAFFLER, R. J. 2007. Twenty-five years of the Taffler z-score model: Does it really have predictive ability? Accounting and Business Research, 37, 285-300. AHRENS, T. 2008. Overcoming the subjective–objective divide in interpretive management accounting research. Accounting, Organizations and Society, 33 (2008), 292-297. ALEXANDER, N. & DOHERTY, A. M. 2009. International Retailing, Oxford, Oxford University Press ALEXANDER, N. & MYERS, H. 2000. The retail internationalisation process. International Marketing Review, 17, 334- 353. ALTMAN, E. I. 1968. Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy. Journal of Finance, 23, 589-609. ALTMAN, E. I. 1993. Corporate Financial Distress and Bankruptcy: A Complete Guide to Predicting & Avoiding Distress and Profiting from Bankruptcy, New York, John Wiley & Sons. ALTMAN, E. I., EOM, Y. H. & KIM, D. W. 1995. Failure Prediction: Evidence from Korea. Journal of International Financial Management & Accounting, 6, 230-249. ALTMAN, E. I., HALDEMAN, R. G. & NARAYANAN, P. 1977. ZETA Analysis a New Model to Identify Bankruptcy Risk of Corporations. Journal of Banking & Finance, 1, 29-54. ALTMAN, E. I. & HOTCHKISS, E. 2006. Corporate Financial Distress and Bankruptcy: Predict and Avoid Bankruptcy, Analyze and Invest in Distressed Debt, New Jersey, John Wiley & Sons Inc. AMBLER, T. 2000. Marketing Metrics. Business Strategy Review, 11, 59-66. AMBLER, T. 2003. Marketing and the Bottom Line: The marketing metrics to pump up cash flow, Harlow, UK, Pearson Education Limited. AMBLER, T., KOKKINAKI, F. & PUNTONI, S. 2004. Assessing Marketing Performance: Reasons for Metrics Selection. Journal of Marketing Management, 20, 475-498. ANTLE, R. & DEMSKI, J. S. 1988. The Controllability Principle in Responsibility Accounting. Accounting Review, 700- 718. ARGENTI, J. 1976. Corporate Planning and Corporate Collapse. Long Range Planning, 9, 12-17. ARGENTI, J. 1983. Predicting Company Failure. Accountant’s Digest 1-25. ARGYRIS, C. 1952. The Impact of Budgets on People, University of Michigan, Controllership Foundation. ARIBI, Z. A. & GAO, S. S. 2011. Narrative disclosure of corporate social responsibility in Islamic financial institutions. Managerial Auditing Journal, 27, 199-222. ARIELY, D. 2009a. The curious paradox of ‘optimism bias’. Business Week, 48. ARIELY, D. 2009b. Predictably Irrational, New York, Harper Collins. ARIKAN, A. & PEPPERS, D. 2008. Multichannel Marketing: Metrics and Methods for On and Offline Success, New Jersey, Wiley. ARINO, A. & DE LA TORRE, J. 1998. Learning from failure: Towards an evolutionary model of collaborative ventures. Organization Science, 9, 306-325. ARNOLD, S. J. 2002. Lessons learned from the world’s best retailers. International Journal of Retail & Distribution Management, 30, 562-570. ARROW, K. J. 1951. Alternative Approaches to the Theory of Choice in Risk-Taking Situations. Econometrica, 19, 404- 437. ARROW, K. J. 1984. The Economics of Agency, California, Institute for Mathematical Studies in the Social Sciences, Stanford University. ASHTON, D., BEATTIE, V., BROADBENT, J., BROOKS, C., DRAPER, P., EZZAMEL, M., GWILLIAM, D., HODGKINSON, R., HOSKIN, K. & POPE, P. 2009. British research in accounting and finance (2001–2007): the 2008 research assessment exercise. British Accounting Review, 41, 199-207. ASHTON, D., DUNMORE, P. & TIPPETT, M. 2004. Double Entry Bookkeeping and the Distributional Properties of a Firm's Financial Ratios. Journal of Business Finance & Accounting, 31, 583-606. ATRILL, P. & MCLANEY, E. J. 2006. Accounting and Finance for Non-Specialists, Harlow, UK, Pearson Education Limited. AZIZ, A., EMANUEL, D. C. & LAWSON, G. H. 1988. Bankruptcy Prediction an Investigation of Cash Flow Based Models. Journal of Management Studies, 25, 419-437. BALCAEN, S. & OOGHE, H. 2006. 35 years of Studies on Business Failure: An overview of the classic statistical methodologies and their related problems. British Accounting Review, 38, 63-93. BALL, R. 2006. International Financial Reporting Standards (IFRS): pros and cons for investors. Accounting and Business Research, 36, Supplement 1, 5-27. BARNES, P. 1987. The Analysis and Use of Financial Ratios: A Review Article. Journal of Business Finance & Accounting, 14, 449-461. BAZELEY, P. & JACKSON, K. 2013. Qualitative Data Analysis with NVivo, London, Sage Publications Limited. BBC 2011. TV's Mary Portas to head High Street shops review. BBC News On Line.

213

BCSC 2007. Future of Retail Property: How Much Space? Executive Summary. London: British Council for Shopping Centres. BDO 2015. BDO High Street Sales Tracker February Monthly Review. London: BDO LLP. BEATTIE, V. 2002. Traditions of research in financial accounting, Andover, UK, South Western Cengage Learning. BEATTIE, V., MCINNES, B. & FEARNLEY, S. A methodology for analysing and evaluating narratives in annual reports: a comprehensive descriptive profile and metrics for disclosure quality attributes. Accounting Forum, 2004. Elsevier, 205-236. BEATTIE, V. & THOMSON, S. J. 2007. Lifting the lid on the use of content analysis to investigate intellectual capital disclosures. Accounting Forum, 31, 129-163. BEAVER, W. H. 1966. Financial Ratios as Predictors of Failure. Journal of Accounting Research, 4, 71-111. BECCHETTI, L. & SIERRA, J. 2003. Bankruptcy risk and productive efficiency in manufacturing firms. Journal of Banking & Finance, 27, 2099-2120. BEINHOCKER, E., DAVIS, I. & MENDONCA, L. 2009. The 10 trends you have to watch. Harvard Business Review, 87, 55-60. BERGER, P. D. & NASR, N. I. 1998. Customer lifetime value: Marketing models and applications. Journal of Interactive Marketing, 12, 17-30. BERNSTEIN, P. L. 1996. Against the Gods: The remarkable story of risk, New York, John Wiley & Sons Inc. BERRY, A. J., COAD, A. F., HARRIS, E. P., OTLEY, D. T. & STRINGER, C. 2009. Emerging themes in management control: A review of recent literature. The British Accounting Review, 41, 2-20. BERRY, L. L. & LUSCH, R. F. 1996. Making Corporate Performance ‘SOAR’. Marketing Management, 5, 12-24. BERRY, L. L., SEIDERS, K. & GRESHAM, L. G. 1997. For love and money: the common traits of successful retailers. Organizational Dynamics, 26, 7-23. BETTS, J. & BELHOUL, D. 1987. The effectiveness of incorporating stability measures in company failure models. Journal of Business Finance & Accounting, 14, 323-334. BHARATH, S. T. & SHUMWAY, T. 2004. Forecasting Default with the KMV-Merton Model, . AFA 2006 Boston Meetings Paper. Boston, USA: Social Science Research Network. BHARGAVA, M., DUBELAAR, C. & SCOTT, T. 1998. Predicting bankruptcy in the retail sector: an examination of the validity of key measures of performance. Journal of Retailing and Consumer Services, 5, 105-117. BILDERBEEK, J. 1979. An empirical study of the predictive ability of financial ratios in the Netherlands. Zeitschrift fur Betriebswirtschaft, 49 (5), 388-407. BLODGETT, J. G., WAKEFIELD, K. L. & BARNES, J. H. 1995. The effects of customer service on consumer complaining behavior. Journal of Services Marketing, 9, 31-42. BOHLEN, B., CARLOTTI, S. & MIHAS, L. 2009. How the recession has changed US consumer behavior. In: MCKINSEY (ed.) McKinsey Quarterly. New York. BOULDING, K. E. 1991. What is evolutionary economics? Journal of Evolutionary Economics, 1, 9-17. BOWERS, R. 1944. Terminology and Form of the Income Sheet. The Accounting Review, 19, 274-279. BRAINARD, W. C. & TOBIN, J. 1968. Pitfalls in financial model building. The American Economic Review, 58, 99-122. BREALEY, R. A., MYERS, S. C. & MARCUS, A. J. 2012. Fundamentals of Corporate Finance Seventh ed. New York: McGraw-Hill/Irwin. BRENGMAN, M., GEUENS, M., WEIJTERS, B., SMITH, S. M. & SWINYARD, W. R. 2005. Segmenting Internet shoppers based on their Web-usage-related lifestyle: a cross-cultural validation. Journal of Business Research, 58, 79-88. BRENNAN, N. M. & CONROY, J. P. 2013. Executive hubris: the case of a bank CEO. Accounting, Auditing & Accountability Journal, 26, 172-195. BRENNAN, N. M. & MERKL-DAVIES, D. M. 2013. Accounting Narratives and Impression Management. The Routledge Companian to Communication in Accounting. Bangor, Wales: Bangor Business School, Centre for Impression Management in Accounting Communication. BROMLEY, R. D. F. & THOMAS, C. J. 1993. Retail change and the issues. In: BROMLEY, R. D. F. & THOMAS, C. J. (eds.) Retail change: Contemporary issues. London: Routledge. BROMWICH, M. 1986. Management accounting research: the state of the art, London, Pitman. BROWN, S. 1987. Institutional Change in Retailing: A Review and Synthesis. European Journal of Marketing, 21, 5-36. BROWN, S. 1991. Variations on a marketing enigma: the wheel of retailing theory. Journal of Marketing Management, 7, 131-155. BUBER, R., RUSO, B., GADNER, J., ATZWANGER, K. & GRUBER, S. Evolutionary store design. How water, plants, animals and sight protection affect consumer behaviour. Proceedings of the Australian and New Zealand Marketing Academy (ANZMAC) conference, 2007. 325-331. BURDEN, S. 2014. Retailer Performance Management (M&A Distress - expert interview). In: TEJI, T. (ed.) Doctoral Research Programme. Manchester Business School: University of Manchester BURRELL, G. & MORGAN, G. 1979. Two dimensions: Four paradigms. Sociological paradigms and organizational analysis, 21-37. BURT, S., MELLAHI, K., JACKSON, P. & SPARKS, L. 2002a. The failure of retail internationalisation in Marks and Spencer. European Retail Digest, 27-30. BURT, S., MELLAHI, K., JACKSON, T. & SPARKS, L. 2002b. Retail internationalization and retail failure: issues from the case of Marks and Spencer. The International Review of Retail, Distribution and Consumer Research, 12, 191-219. BURT, S. & SPARKS, L. 1994. Structural change in grocery retailing in Great Britain: a discount reorientation? International Review of Retail, Distribution and Consumer Research, 4, 195-217. BURT, S. L. & SPARKS, L. 2003. Power and competition in the UK retail grocery market. British Journal of Management, 14, 237-254. BUTTONWOOD. 2015. What's wrong with finance. Economist [Online]. Available: http://www.economist.com/blogs/buttonwood/2015/05/finance-and-economics? [Accessed 9th May 2015]. CADBURY, S. A. 1992. Report of the committee on the financial aspects of corporate governance: the code of best practice. Corporate Governance. London: Gee Professional Publishing,. CALAFELL, J. V. & SOMOZA, A. 2008. 40 Years of Models of Business Insolvency: A Proposal for its Taxonomy. Available at SSRN 1298955.

214

CAMERON, K. S., SUTTON, R. I. & WHETTEN, D. A. 1988. Issues in Organizational Decline. In: CAMERON, K. S., SUTTON, R. I. & WHETTEN, D. A. (eds.) Readings in organizational decline: Frameworks, research, and prescriptions. Cambridge, Massachusetts Ballinger Publishing Company, Harper Row. CAÑIBANO, L., GARCIA-AYUSO, M. & SÁNCHEZ, P. 2000. Accounting for intangibles: a literature review. Journal of Accounting Literature, 19, 102-130. CARROLL, A. B. 2003. The four faces of corporate citizenship. Business and Society Review, 100, 1-7. CASSELL, C. & SYMON, G. 2004. Essential guide to qualitative methods in organizational research, London, SAGE Publications. CHAMBERS, A. 2002. Tolley's Corporate Governance Handbook, London, Tolley Publishing Company Limited. CHAMBERS, A. D. & WEIGHT, C. 2008. Corporate Governance Handbook, Haywards Heath, Sussex, UK, Tottel Publishing. CHARALAMBOUS, C., CHARITOU, A. & KAOUROU, F. 2000. Comparative analysis of artificial neural network models: Application in bankruptcy prediction. Annals of Operations Research, 99, 403-425. CHARITOU, A., NEOPHYTOU, E. & CHARALAMBOUS, C. 2004. Predicting corporate failure: empirical evidence for the UK. European Accounting Review, 13, 465-497. CHARITOU, A. & TRIGEORGIS, L. 2001. Explaining bankruptcy using option theory. SSRN. CHARMAZ, K. 2003. Grounded theory. In: CHARMAZ, K. & SMITH, J. (eds.) Qualitative Psychology: A practical guide to research methods. Second ed. Thousand Oaks, California: Sage. CHARMAZ, K. 2014. Constructing Grounded Theory, Thousand Oaks, California, Sage. CHEN, W. & WELLMAN, B. 2005. Charting digital divides: Comparing socioeconomic, gender, life stage, and rural- urban internet access and use in five countries. In: DUTTON, W. H., KAHIN, B., O'CALLAGHAN, R. & WYCKOFF, A. W. (eds.) Transforming Enterprise: The Economic and Social Implications of Information Technology. Massachesetts, USA Massachusetts Institute of Technology. CHEN, Y. & XIE, J. 2008. Online consumer review: Word-of-mouth as a new element of marketing communication mix. Management Science, 54, 477-491. CHESTERS, L. 2015. Google in race for Tesco's Clubcard firm Dunnhumby as internet giant is after data analysis and customer loyalty know-how. Daily Mail, 19 June 2015. CLARKE, I. 2000. Retail power, competition and local consumer choice in the UK grocery sector. European Journal of Marketing, 34, 975-1002. CLARKE, I., BENNISON, D. & PAL, J. 1997. Towards a contemporary perspective of retail location. International Journal of Retail & Distribution Management, 25, 59-69. CLARKE, T. 1998. Research on corporate governance. Corporate Governance: An International Review, 6, 57-66. CLARKE, T. 2004. Theories of Corporate Governance: The Philosophical Foundations of Corporate Governance, New York, Routledge COCA-STEFANIAK, A. & CARROLL, S. 2014. Managing town centres during the crisis: from retail-focussed management to the experience economy and beyond. In: WRIGLEY, N. & BROOKES, E. (eds.) Evolving High Streets: Resilience & Reinvention. Southampton: ESRC. COLLINS, J. C. & PORRAS, J. I. 1996. Building your company's vision. Harvard Business Review, 74, 65-77. CONNER, D. R. 1993. Managing at the Speed of Change: How Resilient Managers Succeed and Prosper where Others Fail, New York, Randon House. COOPER, D. J. & SHERER, M. J. 1984. The value of corporate accounting reports: Arguments for a political economy of accounting. Accounting, Organizations and Society, 9, 207-232. CORK, K. 1982. Insolvency Law and Practice: Report of the Review Committee (Cork). London: HMSO. CRESWELL, J. W. 2013. Research Design: Qualitative, Quantitative, and Mixed Methods Approaches, Thousand Oaks, California, Sage publications. CRESWELL, J. W. & CLARK, V. L. P. 2011. Designing and Conducting Mixed Methods Research, Thousand Oaks, California, SAGE Publications CURTIS, B., HEFLEY, B. & MILLER, S. 2009. People Capability Maturity Model (P-CMM) Version 2.0. Boston: Software Engineering Institute. DAHMASH, F. N., DURAND, R. B. & WATSON, J. 2009. The value relevance and reliability of reported goodwill and identifiable intangible assets. The British Accounting Review, 41, 120-137. DAMBOLENA, I. G. & KHOURY, S. J. 1980. Ratio stability and corporate failure. The Journal of Finance, 35, 1017- 1026. DARWIN, C. 1859. On the Origins of species by means of Natural Selection, London, Murray. DAVIDSON, W. R., BATES, A. D. & BASS, S. J. 1976. The retail life cycle. Harvard Business Review, 54, 89-96. DAVIES, B. & WHITTRED, G. P. 1980. The Association Between Selected Corporate Attributes and Timeliness in Corporate: Reporting: Further Analysis. Abacus, 16, 48-60. DAVIS, G. F. & COBB, J. A. 2010. Resource dependence theory: Past and future. In: SCHOONHOVEN, C. B. & DOBBIN, F. (eds.) Stanford's Organiztion Theory Renaissance, 1970-2000 (Research in Sociology of Organizations). University of Michigan: Emerald Group Publishing Limited. DAVIS, J. H., SCHOORMAN, F. D. & DONALDSON, L. 1997. Toward a Stewardship Theory of Management. Academy of Management Review, 22, 20-47. DAVIS, S. M., MEYER, C. & GREGORY, J. 1998. Blur: The speed of change in the connected economy, New York, Capstone Pub. DAWSON, J. 2004. Retail change in Britain during 30 years: the strategic use of economies of scale and scope, Scotland`, Centre for the Study of Retailing in Scotland. DAWSON, J. A. 2007. Scoping and conceptualising retailer internationalisation. Journal of Economic Geography, 7, 373-397. DE JONG, A., DEJONG, D. V., MERTENS, G. & ROOSENBOOM, P. 2007. Investor relations, reputational bonding, and corporate governance: The case of Royal Ahold. Journal of accounting and public policy, 26, 328-375. DEAKIN, E. 1977. Business Failure Prediction: An Empirical Analysis In: ALTMAN, E. & SAMETZ, A. (eds.) Financial Crises: Institutions and Markets in a Fragile Environment. New York: John Wiley & Sons. DEAKIN, E. B. 1972. A discriminant analysis of predictors of business failure. Journal of Accounting Research, 167-179. DELOITTE 2006. Talking Shop: MeasuringTrading Performance. In: TEJI, T. (ed.) The Retail Review. London: Deloitte LLP.

215

DELOITTE 2009. Consumer insight into changing habits and shifting patterns. In: TEJI, T. (ed.) The Retail Review. London: Deloitte LLP. DELOITTE 2012. iGAAP 2012 Annual report disclosures for UK listed groups, London, LexisNexis DELOITTE 2014. The Changing Face of Retail. In: ACADEMY, T. D. (ed.) The Deloitte Academy Briefing. London: Deloitte. DELOITTE 2015. Global Powers of Retailing 2015. In: DELOITTE (ed.) Global Powers of Retailing. New York: Deloitte Touche Tomatsu DEMSKI, J. S. 1993. Managerial uses of accounting information, Massachusetts. USA, Kluwer Academic Publishers. DEMSKI, J. S., FELLINGHAM, J. C., IJIRI, Y. & SUNDER, S. 2002. Some thoughts on the intellectual foundations of accounting. Accounting Horizons, 16, 157-168. DENNIS, C. & MCCALL, A. 2005. The savannah hypothesis of shopping. Business Strategy Review, 16, 12-16. DEWEY, J., MOORE, A. W., BROWN, H. C., MEAD, G. H., BODE, B. H., STUART, H. W., TUFTS, J. H. & KALLEN, H. M. 1917. Creative intelligence: Essays in the pragmatic attitude, New York, Henry Holt Company. DICK, P. 2006. Discourse Analysis. In: CASSELL, C. M. & SYMON, G. E. (eds.) Essential Guide to Qualitative Methods in Organizational Research. London: Sage. DONALDSON, L. & DAVIS, J. H. 1991. Stewardship theory or agency theory: CEO governance and shareholder returns. Australian Journal of Management, 16, 49-64. DONALDSON, T. & PRESTON, L. E. 1995. The stakeholder theory of the corporation: Concepts, evidence, and implications. Academy of Management Review, 65-91. DRUCKER, P. 2001. Management Challenges for the 21st Century, London, Harper Paperbacks. DRUCKER, P. F. 1954. The practice of management, London, Pan DRUCKER, P. F. 1964. Managing for results: Economic Taks and Risk Taking Decisions Oxford, Heineman Professional Publishing. DRUCKER, P. F. 1998. The discipline of innovation. Harvard Business Review, 76, 149-157. DUBOFF, R. S. 2007. The wisdom of (expert) crowds. Harvard Business Review, 85, 28. DUTTA, S. & BILBAO-OSORIO, B. The Global information technology report 2012: Living in a hyperconnected world. World Economic Forum and INSEAD Business School, 2012. DUTTON, W. H., KAHIN, B., O'CALLAGHAN, R. & WYCKOFF, A. W. 2005. Transforming Enterprise: The Economic and Social Implications of Information Technology, Massachusetts, USA, Massachusetts Insitute of Technology. DYSON, J. R. 2010. Accounting for non-accounting students, London, Financial Times/Prentice Hall. EDGAR, J. 2013. Retailer Performance Management (Case F). In: TEJI, T. (ed.) Doctoral Research Programme Manchester Business School: University of Manchester EDMISTER, R. O. 1972. An empirical test of financial ratio analysis for small business failure prediction. Journal of Financial and Quantitative Analysis, 7, 1477-1493. EDMONDSON, A. C. 2011. Strategies For Learning From Failure. Harvard Business Review, 89, 48-55. EISENHARDT, K. M. 1989. Building theories from case study research. Academy of Management Review, 14, 532-550. EISENHARDT, K. M. & GRAEBNER, M. E. 2007. Theory building from cases: opportunities and challenges. Academy of Management Journal, 50, 25-32. ELGERS, P. T. 1980. Accounting-Based Risk Predictions: A Re-Examination. The Accounting Review, 55, 389-408. ENDERS, A. & JELASSI, T. 2000. The converging business models of Internet and bricks-and-mortar retailers. European Management Journal, 18, 542-550. ETGAR, M. 1984. The retail ecology model: a comprehensive model of retail change. Research in Marketing, 7, 41-62. EUROPEAN-COMMISSION 1997. Green paper on the convergence of the telecommunications, media and information technology sectors, and the implications for regulation: Towards an information society approach. Brussels: Office for Official Publications of the European Communities. EUSKE, K., LEBAS, M. & MCNAIR, C. 1993. Performance management in an international setting. Management Accounting Research, 4, 275-299. EVAN, W. M. & FREEMAN, R. E. 1993. A stakeholder theory of the modern corporation: Kantian capitalism. In: CHRYSSIDES, G. D. & KALER, J. H. (eds.) An Introduction to Business Ethics. London: Thomson. EVANS, J., BRIDSON, K., BYROM, J. & MEDWAY, D. 2008. Revisiting retail internationalisation: Drivers, impediments and business strategy. International Journal of Retail & Distribution Management, 36, 260-280. EVANS, J. R. & MATHUR, A. 2013. Retailing and the period leading up to the Great Recession: a model and a 25-year financial ratio analysis of US retailing. The International Review of Retail, Distribution and Consumer Research, 1-29. FAMA, E. F. & FRENCH, K. R. 1993. Common risk factors in the returns on stocks and bonds. Journal of financial economics, 33, 3-56. FAMA, E. F. & FRENCH, K. R. 2004. The capital asset pricing model: theory and evidence. The Journal of Economic Perspectives, 18, 25-46. FARRIS, P., BENDLE, N. T., PFEIFER, P. E. & REIBSTEIN, D. J. 2010. Marketing METRICS: The Definitive Guide to Measuring Marketing Performance, New Jersey, Pearson Education. FARRIS, P. W., BENDLE, N. T., PFEIFER, P. E. & REIBSTEIN, D. J. 2006. Marketing Metrics: 50+ metrics every executive should master, New Jersey, Pearson Prentice Hall. FERNIE, J. 1997. Retail change and retail logistics in the United Kingdom: past trends and future prospects. Service Industries Journal, 17, 383-396. FERNIE, J. 1998. The breaking of the fourth wave: recent out-of-town retail developments in Britain. The International Review of Retail, Distribution and Consumer Research, 8, 303-317. FERNIE, J., DOHERTY, N. F. & ELLIS-CHADWICK, F. 2010a. Internet retailing: the past, the present and the future. International Journal of Retail & Distribution Management, 38, 943-965. FERNIE, J., ELMS, J., CANNING, C., DE KERVENOAEL, R., WHYSALL, P. & HALLSWORTH, A. 2010b. 30 years of retail change: where (and how) do you shop? International Journal of Retail & Distribution Management, 38, 817-827. FERNIE, J. & SPARKS, L. 1999. Logistics And Retail Management: Insights Into Current Practice And Trends From Leading Experts, London, Kogan Page. FERNIE, J. & SPARKS, L. 2009. Logistics and Retail Management: Emerging Issues and New Challenges in the Retail Supply Chain, London, Kogan Page.

216

FERNIE, J. & SPARKS, L. 2014. Logistics and Retail Management: Emerging issues and new challenges in the retail supply chain, London, Kogan Page Publishers. FERNIE, J., SPARKS, L. & MCKINNON, A. 2010c. Retail logistics in the UK: past, present and future. International Journal of Retail & Distribution Management, 38, 894-914. FERRARA, W. L. 1964. Responsibility Accounting-A Basic Control Concept. NAA Bulletin, 46, 11-22. FIELD, A. P. 2009. Discovering Statistics Using SPSS, London, SAGE publications Ltd. FINDLAY, A. & SPARKS, L. 2012. Far from the ‘Magic of the Mall’: Retail (Change) in ‘Other Places’. Scottish Geographical Journal, 128, 24-41. FINDLAY, A. & SPARKS, L. 2014. High streets and town centre policies. In: WRIGLEY, N. & BROOKES, E. (eds.) Evolving High Streets: Resilience and Reinvention persepctives from social science. Southampton: ESRC. FOWLER, F. J. 2009. Survey research methods, Sage. FRANCIS, J. & SCHIPPER, K. 1999. Have Financial Statements Lost Their Relevance? Journal of Accounting Research, 37, 319-352. FREEMAN, R. E., HARRISON, J. S., WICKS, A. C., PARMAR, B. L. & DE COLLE, S. 2010. Stakeholder theory: The state of the art, Cambridge, Cambridge University Press. FREESTONE, O. M. & MCGOLDRICK, P. J. 2008. Motivations of the ethical consumer. Journal of Business Ethics, 79, 445-467. FROW, N., MARGINSON, D. & OGDEN, S. 2005. Encouraging strategic behaviour while maintaining management control: multi-functional project teams, budgets, and the negotiation of shared accountabilities in contemporary enterprises. Management Accounting Research, 16, 269-292. FUNK, J. L. 2007. The future of mobile shopping: The interaction between lead users and technological trajectories in the Japanese market. Technological Forecasting and Social Change, 74, 341-356. GALANXHI-JANAQI, H. & NAH, F. F. H. 2004. U-commerce: emerging trends and research issues. Industrial Management & Data Systems, 104, 744-755. GALLOPÍN, G. C. 2006. Linkages between vulnerability, resilience, and adaptive capacity. Global Environmental Change, 16, 293-303. GELLER, E. S. 2002. The challenge of increasing proenvironment behavior. In: BECHTEL, R. B. A. C., A. (ed.) Handbook of Environmental Psychology. New York: John Wiley & Sons Inc. GENTRY, J. A., NEWBOLD, P. & WHITFORD, D. T. 1985. Classifying bankrupt firms with funds flow components. Journal of Accounting Research, 23, 146-160. GHOSH, S. 2011. Rethinking Entrepreneurial Failure. In: GHOSH, P. S. (ed.) Harvard Business School. Harvard Business School: HBS. GINO, F. & PISANO, G. P. 2011. Why Leaders Don't Learn From Success. Harvard Business Review, 89, 68-44. GIOIA, D. A. & CHITTIPEDDI, K. 1991. Sensemaking and Sensegiving in Strategic Change Initiation. Strategic Management Journal, 12, 433-448. GIOIA, D. A., CORLEY, K. G. & HAMILTON, A. L. 2013. Seeking Qualitative Rigor in Inductive Research: Notes on the Gioia Methodology. Organizational Research Methods, 16, 15-31. GIST, R. R. 1968. Retailing: Concepts and Decisions, New York, Wiley GLASER, B. & STRAUSS, A. 1967. The Discovery of Grounded Theory, London, Weidenfeld and Nicholson. GLOUBOS, G. & GRAMMATIKOS, T. 1988. The success of bankruptcy prediction models in Greece. Studies in Banking and Finance, 7, 37–46. GMA 2007. Shopper Marketing: Capturing a Shopper’s Mind, Heart, and Wallet. In: GMA (ed.) Shopper Marketing. London: Grocery Marketing Association. GOOLD, M. 1986. Accounting and strategy. In: HOPWOOD, A. G., HORNGREN, C. T., ALLEN, D. & KIRWAN, M. (eds.) Research and current issues in management accounting. London: Pitman & Sons. GOOLD, M. & QUINN, J. J. 1990. Strategic Control: Milestones for Long-Term Performance, London, Hutchinson. GOUREVITCH, P. A. & SHINN, J. 2005. Political Power and Corporate Control: The new global politics of corporate governance, New Jersey, Princeton University Press. GREENBERG, B. S., SHERRY, J., LACHLAN, K., LUCAS, K. & HOLMSTROM, A. 2010. Orientations to video games among gender and age groups. Simulation & Gaming, 41, 238-259. GREWAL, D., IYER, G. R. & LEVY, M. 2004. Internet retailing: enablers, limiters and market consequences. Journal of Business Research, 57, 703-713. GRIMSEY, B. 2013. The Grimsey Review: An Alternative Future for the High Street. http://www.vanishinghighstreet.com/. GUINÉE, J. B. 2002. Handbook on life cycle assessment operational guide to the ISO standards. The international journal of life cycle assessment, 7, 311-313. GUPTA, S., HANSSENS, D., HARDIE, B., KAHN, W., KUMAR, V., LIN, N., RAVISHANKER, N. & SRIRAM, S. 2006. Modeling customer lifetime value. Journal of Service Research, 9, 139-155. GUTHERIE, J. 2014. M&S results not bad enough to cook Marc Bolland's Christmas goose. Financial Times, 9th January 2014. GUY, C. 2013. Planning for retail development: A critical view of the British experience, Abingdon, Oxford, Routledge. GUY, C. M. 1998. Controlling new retail spaces: the impress of planning policies in Western Europe. Urban studies, 35, 953. HAENLEIN, M., KAPLAN, A. M. & SCHODER, D. 2006. Valuing the real option of abandoning unprofitable customers when calculating customer lifetime value. Journal of Marketing, 70, 5-20. HAMPSON, D. P. & MCGOLDRICK, P. J. Consumers in a Turbulent Economy: Normative, Egoistic and Economic Antecedents to Store (DIS) Loyalty and Store-Brand Proneness. In: DATA-ON, C. M., ed. The Sustainable Global Marketplace, 2011 Florida. Springer, 160-160. HAMPSON, D. P. & MCGOLDRICK, P. J. 2013. A typology of adaptive shopping patterns in recession. Journal of Business Research, 66, 831-838. HANDY, C. 1995. Gods of Management: How They Work, and Why They Will Fail, London, Arrow Books Limited. HANDY, C. 2002. What’s a business for. Harvard Business Review, 80, 49-55. HANDY, C. 2015. The Second Curve: Thoughts on Reinventing Society, London, Random House. HARFORD, T. 2011. Adapt: Why Success Always Starts with Failure, London, Little Brown Book Group. HARRISON, R., SHAW, D. & NEWHOLM, T. 2005. The Ethical Consumer, London, Sage Publications Limited.

217

HART, C. 1999. The retail accordion and assortment strategies: an exploratory study. The International Review of Retail, Distribution and Consumer Research, 9, 111-126. HATCH, M. J. & CUNLIFFE, A. L. 2013. Organization Theory: Modern, Symbolic and Postmodern Perspectives, Oxford, Oxford University Press. HE, Y. & KAMATH, R. 2006. Business failure prediction in retail industry: an empirical evaluation of generic bankruptcy prediction models. Academy of Accounting and Financial Studies Journal, 10, 97. HELGESON, J. G., KLUGE, E. A., MAGER, J. & TAYLOR, C. 1984. Trends in consumer behavior literature: a content analysis. Journal of Consumer Research, 449-454. HENNIG-THURAU, T., WALSH, G. & WALSH, G. 2003. Electronic word-of-mouth: motives for and consequences of reading customer articulations on the internet. International Journal of Electronic Commerce, 8, 51-74. HINSHAW, M. & KASANOFF, B. 2012. Smart Customers and Stupid Companies: Why Only Intelligent Companies Will Thrive, and How to Be One of Them, New York, Business Strategy Press. HMSO 2009. Companies Act 2006. In: HMSO (ed.). London: HMSO. HOBBS, T. 2013. Discounter sector set for rampant growth, IGD predicts. The Grocer, 12 Spetember 2013. HODGSON, G. M. 1998. On the evolution of Thorstein Veblen's evolutionary economics. Cambridge Journal of Economics, 22, 415-431. HOLLANDER, S. C. 1960. The wheel of retailing. The Journal of Marketing, 25, 37-42. HOLLANDER, S. C. 1966. Notes on the Retail Accordion. Journal of Retailing, 42, 29. HONEYCOMBE, G. 1984. Selfridges: Seventy-Five Years: The Story of the Store 1909-1984, London, Park Lane Press. HOPPER, T., NORTHCOTT, D. & SCAPENS, R. W. 2007. Issues in Management Accounting, London, Pearson Education. HOPWOOD, A. G. 1974. Accounting and Human Behaviour, London, Prentice-Hall HOPWOOD, A. G. 2007. Whither accounting research? The Accounting Review, 82, 1365-1374. HORRIGAN, J. O. 1968. A Short History of Financial Ratio Analysis. The Accounting Review, 43, 284-294. HOUTHAKKER, H. S. 1957. An International Comparison of Household Expenditure Patterns, Commemorating the Centenary of Engel's Law. Econometrica, 25, 532-551. HOWER, R. M. 1943. History of Macy's of New York, 1858-1919, Cambridge, MA, USA, Harvard University Press. HUANG, B., KECHADI, M. T. & BUCKLEY, B. 2012. Customer churn prediction in telecommunications. Expert Systems with Applications, 39, 1414-1425. HUGHES, B. 2015. Buddha: Genius of the Ancient World. In: COWLING, R. (ed.) Genius of the Ancient World. Open University BBC4. HUMPHREY, W. S. 1989. Managing the Software Process New Jersey, Addison-Wesley HUSSAINEY, K., SCHLEICHER, T. & WALKER, M. 2003. Undertaking large-scale disclosure studies when AIMR-FAF ratings are not available: the case of prices leading earnings. Accounting and Business Research, 33, 275- 294. IASB 2010. International Financial Reporting Standards In: BOARD, I. A. S. (ed.). London: IASB. ICAEW 1975. The Corporate Report: A Discussion Paper Published for Comment. In: ICAEW (ed.). London: Accounting Standards Steering Committee of ICAEW. ICAEW 1999. Internal Control: Guidance for Directors on the Combined Code, London, Institute of Chartered Accountants in England & Wales. ICAEW 2013. Audit Insights Retail. In: AUDIT & ASSURANCE FACULTY, I. (ed.) Audit Insights. London: Institute of Chartered Accountants in England & Wales. ICAEW 2014. Audit Insights: Retail 2014. In: AUDIT & ASSURANCE FACULTY, I. (ed.) Audit Insights. London: Institute of Chartered Accountants in England & Wales. ICAS 2010. Making Corporate Reports Relevant. In: ICAS (ed.) Insight. Edinburgh: Institute of Chartered Accountants of Scotland. IIRC. 2015. Assurance on Integrated Reporting: Overview of feedback and call to action. Available: http://integratedreporting.org/resource/assurance/ [Accessed 12 July 2015]. IJIRI, Y. 1965. Axioms and Structures of Conventional Accounting Measurement. The Accounting Review, 40, 36-53. IJIRI, Y. & JAEDICKE, R. K. 1966. Reliability and Objectivity of Accounting Measurements. The Accounting Review, 41, 474-483. IOANNIDIS, C., PEEL, D. A. & PEEL, M. J. 2003. The Time Series Properties of Financial Ratios: Lev Revisited. Journal of Business Finance & Accounting, 30, 699-714. JACKSON, D. 1989. Determining a customer’s lifetime value. Direct Marketing, 51, 60-62. JAIN, D. & SINGH, S. S. 2002. Customer lifetime value research in marketing: A review and future directions. Journal of Interactive Marketing, 16, 34-46. JANSEN, B. J., ZHANG, M., SOBEL, K. & CHOWDURY, A. 2009. Twitter power: Tweets as electronic word of mouth. Journal of the American society for information science and technology, 60, 2169-2188. JENKINS, G. Mutlichannel High Street: Winning the Retail Battle in 2015. In: LEWIS, M., ed. The Retail Debate 3rd February 2015 2015 Trinity One, Leeds. Squire Patton Bogs. JENSEN, M. C. 2002. Value maximization, stakeholder theory, and the corporate objective function. Business Ethics Quarterly, 235-256. JENSEN, M. C. & MECKLING, W. H. 1976. Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3, 305-360. JOHNSON, A. W. 1943. Form, Function, and Interpretation of the Profit and Loss Statement. The Accounting Review, 18, 340-347. JOHNSON, H. 1992. Relevance Regained: from top-down control to bottom-up empowerment, New York, The Free Press. JOHNSON, H. T. & KAPLAN, R. S. 1987. Relevance Lost: the rise and fall of management accounting, Boston, MA, Harvard Business School Press. JONES, D. 2005. Next to Me: Luck, Leadership and Living with Parkinson's, London, Nicholas Brealey. JOPSON, B. 2006. Accountancy firms map out new world: radical overhaul of how companies report performance. Financial Times, 8th November 2006. JOYE, Y., POELS, K. & WILLEMS, K. 2011. “Evolutionary Store Atmospherics”–Designing with Evolution in Mind. In: SAAD, G. (ed.) Evolutionary Psychology in the Business Sciences. London: Springer. JSAINSBURY 2014. Annual Report and Financial Statements. London.

218

KALAKOTA, R. & WHINSTON, A. B. 1997. Electronic Commerce: A Manager's Guide, New Jersey, Addison Wesley. KAMATH, R. 2006. Business Failure Prediction in Retail Industry: An Empirical Evaluation of Generic Bankrutcy Prediciton Models. Academy of Accounting and Financial Studies Journal, 10. KAPLAN, A. M. & HAENLEIN, M. 2010. Users of the world, unite! The challenges and opportunities of Social Media. Business horizons, 53, 59-68. KAPLAN, R. & NORTON, D. 1996. Linking the balanced scorecard to strategy. California management review, 39. KAPLAN, R. & NORTON, D. 2000. Having trouble with your strategy?: Then map it, Boston, MA, Harvard Business School Pub. Corp. KAPLAN, R. & NORTON, D. 2001. Transforming the Balanced Scorecard from Performance Measurement to Strategic Management: Part I. Accounting Horizons, 15. KAPLAN, R. & NORTON, D. 2004. Strategy Maps: Converting Intangible Assets into Tangible Outcomes, Boston, MA, Harvard Business Press. KAPLAN, R. S. 1984. The evolution of management accounting. The Accounting Review, LIX, 390-418. KAPLAN, R. S. 1994. Management accounting (1984-1994): development of new practice and theory. Management Accounting Research, 5, 247-260. KAPLAN, R. S. & ATKINSON, A. A. 2015. Advanced Management Accounting, Nodia, India, Pearson International Learning. KAPLAN, R. S. & NORTON, D. P. 1992. The Balanced Scorecard: Measures that drive performance. Harvard Business Review, 70, 71-79. KEASEY, K., MCGUINNESS, P. & SHORT, H. 1990. Multilogit approach to predicting corporate failure--Further analysis and the issue of signal consistency. Omega, 18, 85-94. KEASEY, K. & WATSON, R. 1987. Non Financial Symptoms and the Prediction of Small Company Failure: A Test of Argenti's Hypotheses. Journal of Business Finance & Accounting, 14, 335-354. KEASEY, K. & WATSON, R. 1991. Financial Distress Prediction Models: A Review of Their Usefulness. British Journal of Management, 2, 89-102. KEEN, P. G. & MACKINTOSH, R. 2009. The Freedom Economy: Gaining the Mcommerce Edge in the Era of the Wireless Internet, New York, McGraw-Hill Professional. KELVIN, W. T. L. 1883. Electrical Units of Measurement, Institution Lecture. KENDALL, J. 1999. Axial coding and the grounded theory controversy. Western Journal of Nursing Research, 21, 743- 757. KEYNES, J. M. 1937. The general theory of employment. The Quarterly Journal of Economics, 51, 209. KEYNES, J. M. & MOGGRIDGE, D. 1971. The Collected Writings of John Maynard Keynes, London, Macmillan. KEYNOTE 2015. Discount Retailing. In: KEYNOTE (ed.) Market Report 2015. London: Key Note Ltd. KIETZMANN, J. H., HERMKENS, K., MCCARTHY, I. P. & SILVESTRE, B. S. 2011. Social media? Get serious! Understanding the functional building blocks of social media. Business Horizons, 54, 241-251. KING, N. 2006. Template Analysis In: CASSELL, C. M. & SYMON, G. (eds.) Qualitative methods and analysis in organisational research. London: Sage Publications. KOLMOGOROV, A. N. 1950. Foundations of the Theory of Probability, Oxford, Chelsea Publishing Co. KOOLS, S., MCCARTHY, M., DURHAM, R. & AND ROBRECHT, L. 1996. Dimensional Analysis: Broadening the conception of grounded theory. Qualitative Health Research, 6, 312-330. KUHN, T. S. 1962. The Structure of Scientific Revolutions, Chicago, University of Chicago Press. KUHN, T. S. 1977. The essential tension: selected studies in scientific tradition and change, Chicago, University Of Chicago Press. KUHN, T. S. The Road since Structure. Proceedings of the Biennial Meeting of the Philosophy of Science Association, 1990. The University of Chicago Press on behalf of the Philosophy of Science Association, 3-13. LA VERE, S. & KLEINER, B. H. 1997. Practices of Excellent Companies in the Retail Industry. Managing Service Quality, 7, 34-38. LAITINEN, E. K. 1991. Financial ratios and different failure processes. Journal of Business Finance & Accounting, 18, 649-673. LAMBERT, R. A. 2009. Agency Theory and Management Accounting. In: CHAPMAN, C. S., HOPWOOD, A. G. & SHIELDS, M. D. (eds.) Handbooks of Management Accounting Research. Oxford: Elsevier. LANDIER, A. 2006. Entrepreneurship and the Stigma of Failure. Unpublished manuscript. LANGLEY, A. 1999. Strategies for theorizing from process data. Academy of Management review, 24, 691-710. LANGLEY, A. 2007. Process thinking in strategic organization. Strategic Organization, 5, 271. LANGLEY, A. & ABDALLAH, C. 2011. Templates and turns in qualitative studies of strategy and management. Research methodology in strategy and management, 6, 201-235. LANGLEY, A., SMALLMAN, C., TSOUKAS, H. & VAN DE VEN, A. H. 2013. Process studies of change in organization and management: unveiling temporality, activity, and flow. Academy of Management Journal, 56, 1-13. LANZETTA, J. T. & KANAREFF, V. T. 1962. Information cost, amount of payoff, and level of aspiration as determinants of information seeking in decision making. Behavioral Science, 7, 459-473. LEAHY, T. 2012. Management in 10 Words, London, Random House. LEBAS, M. & EUSKE, K. 2002. A conceptual and operational delineation of performance. In: NEELY, A. (ed.) Business Performance Measurement: Theory and Practice. Cambridge: Cambridge University Press. LEE, H. L. & WHANG, S. 2001. Winning the last mile of e-commerce. MIT Sloan Management Review, 42, 54-62. LEE, K. C., HAN, I. & KWON, Y. 1996. Hybrid neural network models for bankruptcy predictions. Decision Support Systems, 18, 63-72. LEIGHTON, A. 2008. On Leadership: Practical Wisdom from the People Who Know, London, Random House Business Books. LEROUX, M. 2010. Terry Leahy. Sunday Times Business Section 13th June 2010, p.5. LEV, B. 1969. Industry Averages as Targets for Financial Ratios. Journal of Accounting Research, 7, 290-299. LEV, B. & SUNDER, S. 1979. Methodological issues in the use of financial ratios. Journal of Accounting and Economics, 1, 187-210. LEVY, M., GREWAL, D., PETERSON, R. A. & CONNOLLY, B. 2005. The concept of the “Big Middle”. Journal of Retailing, 81, 83-88. LIBBY, R. 1975. Accounting Ratios and the Prediction of Failure: Some Behavioral Evidence. Journal of Accounting Research, 13, 150-161.

219

LIN, C. M., OWENS, W. L. & OWERS, J. E. 2010. The Association Between–Market Risk Disclosure Reporting And Firm Risk: The Impact Of SEC FRR No. 48. Journal of Applied Business Research 26. LIN, F. & MCCLEAN, S. 2001. A data mining approach to the prediction of corporate failure. Knowledge-based systems, 14, 189-195. LIPSEY, R. G. & LANCASTER, K. 1956. The general theory of second best. The review of economic studies, 24, 11-32. LOWE, M. S. 2000. Britain's regional shopping centres: new urban forms? Urban Studies, 37, 261-274. LU, H. P. & SU, P. Y. J. 2009. Factors affecting purchase intention on mobile shopping web sites. Internet Research, 19, 442-458. LUSSIER, R. 1995. A Nonfinancial Business Success versus Failure Prediction Model for Young Firms. Journal of Small Business Management, 33. LYNCH, R. 2014. Tesco to face criminal probe after £263 million hole found in profits. Independent on-line. MACE, M. L. 1979. Directors: Myth and reality-ten years later. Rutgers Law Review, 32, 293. MACE, M. L. 2009. The president and the board of directors. The Value Creating Board, 3, 69. MAGRETTA, J. 2002. Why business models matter. Harvard Business Review, 80, 86-93. MAIGNAN, I. & FERRELL, O. C. 2001. Corporate citizenship as a marketing instrument-Concepts, evidence and research directions. European Journal of Marketing, 35, 457-484. MAIGNAN, I., FERRELL, O. C. & HULT, G. T. M. 1999. Corporate citizenship: cultural antecedents and business benefits. Journal of the Academy of Marketing Science, 27, 455-469. MANNING, L. 2014. Retailer Performance Measurement (President R3). In: TEJI, T. (ed.) Doctoral Research Programme. Manchester Business School: University of Manchester. MARAIS, M. L., PATELL, J. M. & WOLFSON, M. A. 1984. The Experimental Design of Classification Models: An Application of Recursive Partitioning and Bootstrapping to Commercial Bank Loan Classifications. Journal of Accounting Research, 22, 87-114. MARKOWITZ, H. 1952. Portfolio selection*. The journal of finance, 7, 77-91. MARKOWITZ, H. 2012. Portfolio Selection*. The Journal of Finance, 7, 77-91. MARKOWITZ, H. M. 1970. Portfolio Selection: Efficient Diversification of Investments, New Haven, Yale University Press. MARTIKAINEN, T., PERTTUNEN, J., YLI-OLLI, P. & GUNASEKARAN, A. 1995. Financial ratio distribution irregularities: Implications for ratio classification. European Journal of Operational Research, 80, 34-44. MCGOLDRICK, P. J. 2002. Retail Marketing, London, McGraw-Hill. MCGOLDRICK, P. J. & THOMPSON, M. G. 1991. Regional Shopping Centres: Out-of-town Versus In-town, Aldershot, Avebury. MCGRATH, R. G. 1999. Falling forward: Real options reasoning and entrepreneurial failure. Academy of Management Review, 24, 13-30. MCGRATH, R. G. 2011. Failing By Design. Harvard Business Review, 89, 76-83. MCGURR, P. & DEVANEY, S. 1998a. Predicting Business Failure of Retail Firms: An Analysis Using Mixed Industry Models. Journal of Business Research, 43, 169-176. MCGURR, P. & DEVANEY, S. 1998b. A Retail Failure Prediction Model. The International Review of Retail, Distribution and Consumer Research, 8, 259-276. MCKINSEY 2011. Big data: The next frontier for innovation, competition, and productivity. In: INSTITUTE, M. G. (ed.). mckinsey.com/mgi: MGI. MCLAUGHLIN, J. A. 1927. Amendment of the Bankruptcy Act. Harvard Law Review, 40, 341-391. MCNAIR, M. P. 1931. Trends in large scale retailing. Harvard Business Review, 10, 30-39. MCNAIR, M. P. & MAY, E. G. 1978. Next Revolution of Retailing Wheel. Harvard Business Review, 56, 81-91. MELLAHI, K. & WILKINSON, A. 2004. Organizational failure: a critique of recent research and a proposed integrative framework. International Journal of Management Reviews, 5, 21-41. MELLAHI, K. & WILKINSON, A. 2010. Managing and coping with organizational failure: Introduction to the special issue. Group & Organization Management, 35, 531-541. MELOCHE, M., DI BENEDETTO, C., YUDELSON, J. & KING, R. 1988. A framework for the analysis of the growth and development of retail institutions. King, RL (red.), Retailing: Its Present and Future, Charleston, IL: American Collegiate Retailing Association, 6-11. MERKL-DAVIES, D. M. & BRENNAN, N. M. 2011. A conceptual framework of impression management: new insights from psychology, sociology and critical perspectives. Accounting and Business Research, 41, 415-437. MERKL-DAVIES, D. M., BRENNAN, N. M. & VOURVACHIS, P. 2012. Text Analysis Methodologies in Corporate Narrative Reporting Research. 23rd CSESR International Colloquim St Andrews. MICHA, B. 1984. Analysis of business failures in France. Journal of Banking & Finance, 8, 281-291. MILLER, H. 1996. The Mulitple Dimensions of Information Quality. Information Systems Management, 13, 79-82. MINTEL 2005. UK Retail Rankings 2005 In: MINTEL (ed.) UK Retail Rankings. 2003 ed. London. MINTEL 2011. UK Retail Rankings 2011 In: MINTEL (ed.) UK Retail Rankings. 2011 ed. London. MIYAZAKI, A. D. & FERNANDEZ, A. 2001. Consumer perceptions of privacy and security risks for online shopping. Journal of Consumer Affairs, 35, 27-44. MODIGLIANI, F. & MILLER, M. H. 1958. The cost of capital, corporation finance and the theory of investment. The American Economic Review, 48, 261-297. MONKS, R. A. G. & MINOW, N. 2003. Corporate Governance London, Blackwell Publishing. MONKS, R. A. G. & MINOW, N. 2012. Corporate Governance, London, Wiley. MOORE, C. 2014. Smart beta comes into its own after smart marketing. Financial Times, 16 November 2014. MOORE, C. M., FERNIE, J. & BURT, S. 2000. Brands without boundaries-the internationalisation of the designer retailer's brand. European Journal of Marketing, 34, 919-937. MORAN, L. R. & MCCULLY, C. P. 2001. Trends in Consumer Spending, 1959–2000. Survey of Current Business, 81, 15. MORRIS, R. 1997. Early Warning Indicators of Corporate Failure: A critical review of previous research and further empirical evidence, Aldershot, Hampshire, Ashgate MORRIS, R. C. 1998. Forecasting Bankruptcy: How Useful Are Failure Prediction Models? Management Accounting 76, 22-25. MOSCHIS, G. P. 2007. Life course perspectives on consumer behavior. Journal of the Academy of Marketing Science, 35, 295-307.

220

MOSES, D. & LIAO, S. S. 1987. On developing models for failure prediction. Journal of Commercial Bank Lending, 69, 27–38. MOSSMAN, C. E., BELL, G. G., SWARTZ, L. M. & TURTLE, H. 1998. An empirical comparison of bankruptcy models. Financial Review, 33, 35-54. MUTH, M. & DONALDSON, L. 2002. Stewardship theory and board structure: a contingency approach. Corporate Governance: An International Review, 6, 5-28. MYERS, H. & LUMBERS, M. 2008. Understanding older shoppers: a phenomenological investigation. Journal of Consumer Marketing, 25, 294-301. NAISBITT, J. & ABURDENE, P. 1990. Megatrends 2000: Ten New Directions for the 1990s, , New York, NY, William and Morrow Co Inc. NEELY, A. 2002. Business Performance Measurement: Theory and Practice, Cambridge, Cambridge University Press. NEELY, A., GREGORY, M. & PLATTS, K. 2005. Performance measurement system design: A literature review and research agenda. International Journal of Operations & Production Management, 25, 1228-1263. NESLIN, S. A., GUPTA, S., KAMAKURA, W., LU, J. & MASON, C. H. 2006. Defection detection: Measuring and understanding the predictive accuracy of customer churn models. Journal of Marketing Research, 43, 204- 211. NEWHOLM, T. & SHAW, D. 2007. Studying the ethical consumer: A review of research. Journal of Consumer Behaviour, 6, 253-270. NHS-ENGLAND 2015. Diagnostic waiting times and activity. In: OPERATIONS, A. (ed.). Leeds: National Health Service. O'MEARA, B. E. 1822. Napoleon in Exile; Or, A Voice From St. Helena, London, BiblioBazaar. ODELUGA, K. 2014. Tesco share collapse accelerates after accounting error. CityIndex. ODOM, M. & SHARDA, R. A neural network model for bankruptcy prediction. IJCNN International Conference on Neural Networks, 1990 Poland. 163-168. OFCOM 2015. The Communications Market Report 2015. In: OFCOM (ed.) CMR. London: Communications Market Institute. OHLSON, J. 1980. Financial ratios and the probabilistic prediction of bankruptcy. Journal of Accounting Research, 18, 109-131. OKOLI, C. & PAWLOWSKI, S. D. 2004. The Delphi method as a research tool: an example, design considerations and applications. Information & Management, 42, 15-29. ONS 2012. Quarterly National Accounts Q3 2012. In: STATISTICS, O. F. N. (ed.). National Archives, Kew, London: HMSO. ONS 2014a. Business Demography 2013: births deaths and survival of UK enterprises. In: STATISTICS, O. F. N. (ed.). London: ONS. ONS 2014b. Family Spending, 2014 Edition. In: STATISITCS, O. F. N. (ed.) 2014 ed. London: HMSO. ONS 2015a. Economic Review March 2015. In: ADVISOR, O. O. T. C. E. (ed.). London: Office of National Statistics. ONS 2015b. Retail Sales, December 2014 London: Office for National Statisitcs. OOGHE, H. & VERBAERE, E. 1985. Predicting business failure on the basis of accounting data: The Belgian experience. The International Journal of Accounting, 9, 19-44. OTLEY, D. 1999. Performance management: a framework for management control systems research. Management Accounting Research, 10, 363-382. OTLEY, D. 2002. British research in accounting and finance (1996-2000): the 2001 Research Assessment Exercise. The British Accounting Review, 34, 387-417. OTLEY, D., BROADBENT, J. & BERRY, A. 1995. Research in management control: an overview of its development. British Journal of Management, 6, S31-S44. OTLEY, D. T. 1977. Behavioural Aspects of Budgeting. London: Institute of Chartered Accountants in England and Wales. OTLEY, D. T. 1978. Budget use and managerial performance. Journal of Accounting Research, 122-149. OTLEY, D. T. & BERRY, A. J. 1980. Control, organisation and accounting. Accounting, Organizations and Society, 5, 231-244. PAL, J., MEDWAY, D. & BYROM, J. 2006. Analysing retail failure from an historical perspective: A case study of A. Goldberg & Sons plc. The Service Industries Journal, 26, 513-535. PAL, J., MEDWAY, D. & BYROM, J. 2011. Deconstructing the notion of blame in corporate failure. Journal of Business Research, 64, 1043-1051. PARKER, C., NTOUNIS, N., QUIN, S. & GRIME, I. 2014. High Street Research Agenda: Identifying High Street Research Priorities. Journal of Place Management and Development, 7, 176-184. PEEL, M. J. & PEEL, D. 1988. A multilogit approach to predicting corporate failure--Some evidence for the UK corporate sector. Omega, 16, 309-318. PENROSE, E. 1959. The Theory of the Growth of the Firm, Oxford, Oxford University Press. PESTON, R. 2007. Qatari firm tables Sainsbury's bid. BBC News Channel, 18 July 2007. PETERS, T. & AUSTIN, N. K. 1985. A Passion for Excellence: the Leadership Difference, New York, Random House. PETERS, T. J. & WATERMAN, R. 1982. In Search of Excellence. , New York, Harper & Row. PETTIGREW, A. & MCNULTY, T. 1995. Power and influence in and around the boardroom. Human Relations, 48, 845- 873. PITTA, D. & GURAU, C. 2012. A life-stage analysis of consumer loyalty profile: comparing Generation X and Millennial consumers. Journal of Consumer Marketing, 29, 103-113. PLATT, H. D. & PLATT, M. B. 1991. A note on the use of industry-relative ratios in bankruptcy prediction. Journal of Banking & Finance, 15, 1183-1194. PLATT, H. D., PLATT, M. B. & PEDERSEN, J. G. 1994. Bankruptcy discrimination with real variables. Journal of Business Finance & Accounting, 21, 491-510. PLUMMER, J. T. 1974. The concept and application of life style segmentation. the Journal of Marketing, 33-37. PORTAS, M. 2011. The Portas Review: An independent review into the future of our high streets. London: Department of Business, Innovation and Skills PORTER, M. 1979. The structure within industries and companies' performance. The Review of Economics and Statistics, 61, 214-227. PORTER, M. 1985. Competitive Advantage: Creating and sustaining superior performance, New York, Free Press

221

PORTER, M. 1993. Competitive strategy. Measuring Business Excellence, 1, 12-17. PORTER, M. E. 1974. Consumer Behavior, Retailer Power and Market Performance in Consumer Goods Industries. The Review of Economics and Statistics, 56, 419-436. PORTER, M. E. 1980. Competitive Strategy: Techniques for Analyzing Industry and Competitors, New York, Free Press, MacMillan. PORTER, M. E. & MILLAR, V. E. 1985. How information gives you competitive advantage. Harvard Business Review, Reprint Service. PORTER, M. E. & SOLVELL, O. 1999. The Role of Geography in the Process of Innovation and the Sustainable Competitive Advantage of Firms. The Dynamic Firm, 1, 440-469. POUND, J. 1993. The Rise of the Political Model of Corporate Governance and Corporate Control. New York University Law Review, 68, 1003. POWELL, D. 1991. Counter Revolution: The Tesco Story, London, Grafton Books, Harper Collins. PREBLE, J. F. 1984. The selection of Delphi panels for strategic planning purposes. Strategic Management Journal, 5, 157-170. PRESIDENT-ICAEW 2014. Governance & Risk in Retail Accounting Disclosures (expert interview). In: TEJI, T. (ed.) Doctoral Research Programme Manchester Business School. University of Manchester. RANDALL, J. 2003. Record fine for toy price-fixing. BBC News Channel, 19 February 2003 17:08 GMT. RANKIN, J. 2015. Marks and Spencer bosses get bonuses for first time in two years. The Guardian, 4th June 2015. REINARTZ, W., THOMAS, J. S. & KUMAR, V. 2005. Balancing acquisition and retention resources to maximize customer profitability. Journal of Marketing, 69, 63-79. RETAILWEEK 2015. Retail Week Property April 2015. London: Retail Week. RINK, D. R. & SWAN, J. E. 1979. Product life cycle research: A literature review. Journal of business Research, 7, 219- 242. ROBERTS, J. & SCAPENS, R. 1985. Accounting systems and systems of accountability — understanding accounting practices in their organisational contexts. Accounting, Organizations and Society, 10, 443-456. ROBRECHT, L. C. 1995. Grounded theory: Evolving methods. Qualitative Health Research, 5, 169-177. ROE, M. J. 2006. Political determinants of corporate governance: political context, corporate impact, Oxford, Oxford University Press. ROSS, S. A. 1973. The economic theory of agency: The principal's problem. The American Economic Review, 63, 134- 139. ROTEM-MINDALI, O. & WELTEVREDEN, J. W. 2013. Transport effects of e-commerce: what can be learned after years of research? Transportation, 40, 867-885. ROTH, V. J. & KLEIN, S. 1993. A Theory of Retail Change. International Review of Retail, Distribution and Consumer Research, 3, 167-183. ROWE, G. & WRIGHT, G. 2001. Expert opinions in forecasting: The role of the Delphi technique. International Series in Operations Reasearch and Management Science, 125-144. RUST, R. T., MOORMAN, C. & BHALLA, G. 2010. Rethinking Marketing. Harvard Business Review, 88, 94-101. RYAN, B., SCAPENS, R. & THEOBALD, M. 2002. Research Method and Methodology in Finance and Accounting, Andover, Hampshire Cengage Learning EMEA. SAMPSON, S. D. & TIGER, D. J. 1994. The impact of warehouse membership clubs: the wheel of retailing turns one more time. The International Review of Retail, Distribution and Consumer Research, 4, 33-59. SARBANES-OXLEY 2002. Sarbanes-Oxley Act. Washington, USA: United States Government. SARGUT, G. & MCGRATH, R. G. 2011. Learning to live with complexity. Harvard Business Review, 89, 68. SAUNDERS, M., LEWIS, P. & THORNHILL, A. 2012. Research Methods for Business Students, Harlow, Essex, England, Pearson Education Limited. SAYER, A. 2010. Method in Social Science: A Realist Approach, Oxford, Routledge, Taylor & Francis Group. SCAPENS, R. W. 1990. Researching management accounting practice: the role of case study methods. The British Accounting Review, 22, 259-281. SCAPENS, R. W. 1994. Never mind the gap: towards an institutional perspective on management accounting practice. Management Accounting Research, 5, 301-321. SCAPENS, R. W. & BROMWICH, M. 2001. Editorial Report—Management Accounting Research: the first decade. Management Accounting Research, 12, 245-254. SCAPENS, R. W. & BROMWICH, M. 2010. Management accounting research: 20 years on. Management Accounting Research, 21, 278-284. SCHATZMAN, L. 1991. Dimensional analysis: Notes on an alternative approach to the grounding of theory in qualitative research. In: MAINES, D. R. (ed.) Social Organization and Social Process: Essays in honor of Anselm Strauss. New York: Aldine de Gruyter. SCHERER, A. G. & PALAZZO, G. 2008. Handbook of Research on Global Corporate Citizenship, Cheltenham, UK, Edward Elgar Publishing. SCHIRESON, B. 1957. Towards a New Accounting. The Accounting Review, 32, 253-257. SCHLEICHER, T., HUSSAINEY, K. & WALKER, M. 2007. Loss firms’ annual report narratives and share price anticipation of earnings. The British Accounting Review, 39, 153-171. SCHUMPETER, J. A. 1980. Business Cycles: A Theoretical, Historical and Statistical Analysis of Capitalist Process Kent, UK, Atlantic Books. SEAL, W. 2010. Managerial discourse and the link between theory and practice: From ROI to value-based management. Management Accounting Research, 21, 95-109. SENN, J. A. 2000. The emergence of m-commerce. Computer, 33, 148-150. SFO 2003. WICKES Plc - Leslie Rosenthal acquitted by direction. London: Serious Fraud Office. SHAH, A. K. 1998. Exploring the influences and constraints on creative accounting in the United Kingdom. European Accounting Review, 7, 83-104. SHARPE, W. F. 1964. Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk. Journal of Finance, 19, 425-442. SHARPE, W. F. 2012. Capital Asset Prices: A Theory of Market Equilibrium Under Conditions of Risk. The Journal of Finance, 19, 425-442. SHEPHERD, D. A. 2003. Learning from business failure: Propositions of grief recovery for the self-employed. Academy of Management Review, 28, 318-328.

222

SHEPHERD, D. A. 2004. Educating entrepreneurship students about emotion and learning from failure. Academy of Management Learning & Education, 3, 274-287. SHEPPARD, D. & HUME, N. 2015. A prison ID in one hand - and a CFO award in the other. Financial Times, 1 July 2015. SHEPPARD, J. P. 1994. Strategy and bankruptcy: An exploration into organizational death. Journal of Management, 20, 795-833. SHEPPARD, J. P. & CHOWDHURY, S. D. 2005. Riding the Wrong Wave: Organizational Failure as a Failed Turnaround. Long Range Planning, 38, 239-260. SHUMWAY, T. 2001. Forecasting Bankruptcy More Accurately: A Simple Hazard Model*. The Journal of Business, 74, 101-124. SIMMONDS, K. 1989. The Fundamentals of Strategic Management Accounting, London, Chartered Institute of Management Accountants. SIMONS, R. 1990. The role of management control systems in creating competitive advantage: new perspectives. Accounting, Organizations and Society, 15, 127-143. SIMONS, R. 1994. How new top managers use control systems as levers of strategic renewal. Strategic Management Journal, 15, 169-189. SIMONS, R. 2013. Levers of control: how managers use innovative control systems to drive strategic renewal, Boston, Harvard Business Press. SIMONS, R., DÁVILA, A. & KAPLAN, R. S. 2000. Performance measurement & control systems for implementing strategy: text & cases, Prentice Hall Upper Saddle River, NJ. SIRDESHMUKH, D., SINGH, J. & SABOL, B. 2002. Consumer trust, value, and loyalty in relational exchanges. Journal of Marketing, 66, 15-37. SMITH, A. 1776. The Wealth of Nations. London: Penguin Books, ed. SMITH, A. 2015. Ways to count lile-for-like sales must reflect shopping changes. Financial Times. SMITH, D. & SPARKS, L. 2004. Logistics in Tesco: past, present and future. Logistics and Retail Management, 2, 101- 120. SMITH, D. & SPARKS, L. 2007. Temperature controlled supply chains. Food supply chain management, 179. SMITH, M. 2011. Research methods in accounting, Sage Publications Limited. SMITH, T. 1992. Accounting for Growth: Stripping the camouflage from company accounts, London, Century Business SOARS, B. 2009. Driving sales through shoppers' sense of sound, sight, smell and touch. International Journal of Retail & Distribution Management, 37, 286-298. SOLOMON, J. F., SOLOMON, A., NORTON, S. D. & JOSEPH, N. L. 2000. A conceptual framework for corporate risk disclosure emerging from the agenda for corporate governance reform. The British Accounting Review, 32, 447-478. SOLOMON, M., RUSSELL-BENNETT, R. & PREVITE, J. 2012. Consumer Behaviour, NSW, Australia, Pearson Higher Education SOOPRAMANIEN, D. G. & ROBERTSON, A. 2007. Adoption and usage of online shopping: An empirical analysis of the characteristics of “buyers”“browsers” and “non-internet shoppers”. Journal of Retailing and Consumer Services, 14, 73-82. SPARKS, L. 1990. Spatial-structural relationships in retail corporate growth: a case-study of Kwik Save Group plc. Service Industries Journal, 10, 25-84. SPARKS, L. 1994. The logistics transformation of British retailing: concepts and questions. International Journal of Logistics Management, The, 5, 53-62. SPARKS, L. 1995. Reciprocal retail internationalisation: the Southland Corporation, Ito-Yokado and 7-Eleven convenience stores. Service Industries Journal, 15, 57-96. SPARKS, L. 2014. Turmoil at Tesco. Stirling Retail [Online]. [Accessed 30 September 2014 2014]. SRINIVASAN, S. & HANSSENS, D. 2009. Marketing and firm value: Metrics, methods, findings, and future directions. Journal of Marketing Research, 46, 293-312. STAHLBERG, M. & MAILA, V. 2012. Shopper Marketing: How to Increase Purchase Decisions at the Point of Sale, Kogan Page. STRAUSS, A. & CORBIN, J. 1994. Grounded Theory Methodology. Handbook of Qualitative Research. Thousand Oaks, California Sage. STRAUSS, H. J. & ZEIGLER, L. H. 1975. The Delphi Technique and Its Uses In Social Science Research*. The Journal of Creative Behavior, 9, 253-259. SUDARSANAM, P. S. & TAFFLER, R. J. 1995. Financial ratio proportionality and inter-temporal stability: An empirical analysis. Journal of Banking & Finance, 19, 45-60. SULL, D. 1999. Why good companies go bad. Harvard Business Review, 77, 42-56. SULL, D. 2005. Why Good Companies Go Bad and How Great Managers Remake Them, Boston , MA, Harvard Business Press. SWANSON, E. & TYBOUT, J. 1988. Industrial bankruptcy determinants in Argentina. Journal of Banking and Finance, 7, 1-27. SWINYARD, W. R. & SMITH, S. M. 2003. Why people (don't) shop online: A lifestyle study of the Internet consumer. Psychology & Marketing, 20, 567. TAFFLER, R. & TISSHAW, H. 1977. Going concern, gone-four factors which predict. Accountancy, 50-4. TAFFLER, R. J. 1982. Forecasting Company Failure in the UK Using Discriminant Analysis and Financial Ratio Data. Journal of the Royal Statistical Society. , 145, 342-358. TAFFLER, R. J. 1983. The assessment of company solvency and performance using a statistical model. Accounting and Business Research, 13, 295-308. TAFFLER, R. J. 1984. Empirical models for the monitoring of UK corporations. Journal of Banking & Finance, 8, 199- 227. TAMARI, M. 1966. Financial ratios as a means of forecasting bankruptcy. Management International Review, 6, 15-21. TANG, T.-C. & CHI, L.-C. 2005. Neural networks analysis in business failure prediction of Chinese importers: A between-countries approach. Expert Systems with Applications, 29, 244-255. TESCO 1995. Annual Report and Accounts. Cheshunt, Hertfordshire. TESCO 2014a. Annual Report and Accounts. Cheshunt, Hertfordshire. TESCO 2014b. Tesco and Society Report 2014. Cheshunt, Hertfordshire.

223

TESCO 2015. Annual Report and Accounts. Cheshunt, Hertfordshire. THALER, R. H. 2015. Misbehaving: The Making of Behavioral Economics, New York, WW Norton & Company. THORNBERG, R. 2012. Informed grounded theory. Scandinavian Journal of Educational Research, 56, 243-259. TRUSOV, M., BUCKLIN, R. E. & PAUWELS, K. 2009. Effects of word-of-mouth versus traditional marketing: findings from an internet social networking site. Journal of marketing, 73, 90-102. TUGBY, L. 2015. Discounters outpace supermarket and convenience store growth. Retail Week. London: retailweek.com. TURNBULL, S. 1997. Corporate governance: its scope, concerns and theories. Corporate Governance: An International Review, 5, 180-205. UOO 2005. Oxford Dictionary of English, Oxford, Oxford University Press. VAISHNAVI, S., CONNOR, K. & DAVIDSON, J. R. T. 2007. An abbreviated version of the Connor-Davidson Resilience Scale (CD-RISC), the CD-RISC2: Psychometric properties and applications in psychopharmacological trials. Psychiatry Research, 152, 293-297. VAN FREDERIKSLUST, R. A. I. 1978. Predictability of Corporate Failure: Models for prediction of corporate failure and for evaluation of debt capacity, Leiden, Martinius Nijhoff Social Sciences Division,. VEBLEN, T. 1908. On the nature of capital. The Quarterly Journal of Economics, 22, 517-542. VENKATACHALAM, M., LINSMEIER, T. J., THORNTON, D. B. & WELKER, M. 2001. Do FRR 48 Disclosures Reduce Investors' Uncertainty and Diversity of Opinion about Firms' Market Risk Exposures?: A Trading Volume Analysis, Graduate School of Buisness, Stanford University. VENKATESAN, R. & KUMAR, V. 2004. A customer lifetime value framework for customer selection and resource allocation strategy. Journal of marketing, 68, 106-125. VIRÁG, M. & KRISTÓF, T. 2005. Neural networks in bankruptcy prediction - A comparative study on the basis of the first Hungarian bankruptcy model. Acta Oeconomica, 55, 403-426. WALSH, J. & GODFREY, S. 2000. The Internet: a new era in customer service. European Management Journal, 18, 85- 92. WALTERS, D. 1977. Approaches to retail management. Retail and Distribution Management, 5, 7-11. WALTERS, D. & WHITE, D. 1987. Retail Marketing Management, Basingstoke, Macmillan. WALTERS, R. G. & JAMIL, M. 2003. Exploring the relationships between shopping trip type, purchases of products on promotion, and shopping basket profit. Journal of Business Research, 56, 17-29. WANG, Z. J., LI, H. X. & DENG, X. L. Empirical study of business failure prediction based on rough set. 2007 Lille. IEEE, 848-852. WARD, K. 1992. Strategic Management Accounting, Oxford, Butterworth-Heinemann WARD, K. 2012. Strategic Management Accounting, Oxford, Routledge. WARNABY, G. & MEDWAY, D. 2004. The role of place marketing as a competitive response by town centres to out-of- town retail developments. The International Review of Retail, Distribution and Consumer Research, 14, 457- 477. WARREN, C. 1935. Bankruptcy in United States History, Washington, Beard Books. WATSON, C. J. 1990. Multivariate Distributional Properties, Outliers, and Transformation of Financial Ratios. The Accounting Review, 65, 682-695. WATSON, R. T., PITT, L. F., BERTHON, P. & ZINKHAN, G. M. 2002. U-commerce: expanding the universe of marketing. Journal of the Academy of Marketing Science, 30, 333-347. WELLS, W. D. 2011. Life Style and Psychographics, USA, Marketing Classics Press Edition. WHETTEN, D. A. 1980. Organizational decline: A neglected topic in organizational science. Academy of Management Review, 577-588. WHITTINGTON, G. 1980. Some basic properties of accounting ratios. Journal of Business Finance & Accounting, 7, 219-232. WHITTINGTON, G. & TIPPETT, M. 1999. The Components of Accounting Ratios as Co‐integrated Variables. Journal of Business Finance & Accounting, 26, 1245-1273. WILLIAMS, H. 2014. Preface In: WRIGLEY, N. & BROOKES, E. (eds.) Evolving High Streets: Resilience & Reinvention: Persepectives from Social Science. Southampton: ESRC. WIND, J. 2009. Rethinking marketing: Peter Drucker’s challenge. Journal of the Academy of Marketing Science, 37, 28- 34. WOLFINBARGER, M. & GILLY, M. C. 2001. Shopping online for freedom, control, and fun. California Management Review, 43, 34-55. WOOD, Z. 2012. Discount stores boom as upmarket shoppers boast of saving money. The Guardian, 26 August 2012. WOODHEAD, L. 2012. Shopping, Seduction & Mr Selfridge, Profile Books. WOOLWORTHS 2008. Annual Report and Accounts 2008. London: Woolworths Group PLC. WRIGLEY, N. 2014. Evolving high streets: in what ways can social science research add value? In: WRIGLEY, N. & BROOKES, E. (eds.) Evolving High Streets: Reilience & Reinvention. Southampton: ESRC. WRIGLEY, N. & BROOKES, E. 2014. Evolving High Streets: Resilience and Reinvention perspectives from social science. In: WRIGLEY, N. & BROOKES, E. (eds.) Evolving High Streets: Resilience and Reinvention

Southampton: ESRC. WRIGLEY, N. & DOLEGA, L. 2011. Resilience, fragility, and adaptation: new evidence on the performance of UK high streets during global economic crisis and its policy implications. Environment and Planning-Part A, 43, 2337. WRIGLEY, N. & LAMBIRI, D. 2014. Convenience culture and the evolving high street. In: WRIGLEY, N. & BROOKES, E. (eds.) Evolving High Streets: Resilience & Reinvention. Southampton: ESRC. WU, W. W. 2010. Beyond business failure prediction. Expert Systems with Applications, 37, 2371-2376. WU, Y., GAUNT, C. & GRAY, S. 2010. A comparison of alternative bankruptcy prediction models. Journal of Contemporary Accounting & Economics, 6, 34-45. WYATT, A. 2005. Accounting recognition of intangible assets: theory and evidence on economic determinants. The Accounting Review, 80, 967-1003. XU, X. & WANG, Y. 2009. Financial failure prediction using efficiency as a predictor. Expert Systems with Applications, 36, 366-373. YANG, K. 2010. Determinants of US consumer mobile shopping services adoption: implications for designing mobile shopping services. Journal of Consumer Marketing, 27, 262-270.

224

YANG, Z. R., PLATT, M. B. & PLATT, H. D. 1999. Probabilistic neural networks in bankruptcy prediction. Journal of Business Research, 44, 67-74. YIN, R. K. 2009. Case Study Research: Design and Methods, London, Sage Publications, Inc. YOU, H. & ZHANG, X. 2009. Financial reporting complexity and investor underreaction to 10-K information. Review of Accounting Studies, 14, 559-586. ZAHAY, D. & GRIFFIN, A. 2010. Marketing strategy selection, marketing metrics, and firm performance. Journal of Business & Industrial Marketing, 25, 84-93. ZAVGREN, C. 1983. The prediction of corporate failure: the state of the art. Journal of Accounting Literature, 2, 1-37. ZAVGREN, C. V. 1985. Assessing the vulnerability to failure of American industrial firms: a logistic analysis. Journal of Business Finance & Accounting, 12, 19-45. ZAVGREN, C. V., DUGAN, M. T. & REEVE, J. M. 1988. The association between probabilities of bankruptcy and market responses—a test of market anticipation. Journal of Business Finance & Accounting, 15, 27-45. ZAVGREN, C. V. & FRIEDMAN, G. E. 1988. Are bankruptcy prediction models worthwhile? An application in securities analysis. Management International Review, 28, 34-44. ZEITHAML, V. A., BERRY, L. & PARASURAMAN, A. 1991. Understanding customer expectations of service. Sloan management review, 32, 42. ZHOU, J. G. & TIAN, J. M. Predicting corporate financial distress based on rough sets and wavelet support vector machine. Wavelet Analysis and Pattern Recognition, 2008 Beijing. IEEE, 602-607. ZMIJEWSKI, M. E. 1984. Methodological issues related to the estimation of financial distress prediction models. Journal of Accounting Research, 22, 59-82. ZOTT, C. & AMIT, R. 2013. The business model: A theoretically anchored robust construct for strategic analysis. Strategic Organization, 11, 403-411. ZOTT, C., AMIT, R. & MASSA, L. 2011. The business model: recent developments and future research. Journal of Management, 37, 1019-1042.

225

APPENDICES

Appendix 1: List of 194 retailers in distress 2005-2010

Retailer Retailer Retailer

A.G. Stanley Limited Evesham Technology LMB Realisations Limited Adams Childrenswear Express Gifts Limited Maison Blanc Limited AGA Rangemaster Limited F&M Sales Group UK Malachite 1 Limited Albert Fisher Group Faith Group Limited Marston Mills Limited Alexon International Famous Army Stores Matte PLC Allders Limited Farepak Food & Gifts MFI Furniture Centres Allied Carpets Group PLC Fenwick,Limited Miller Brothers Group Allsports Limited Fhsc Limited Miss Sixty UK Ltd Appollo Discount Limited Findel P.L.C. MKD Holdings Limited Atlantic Fashions Limited Fired Earth Limited Mosaic Fashions Austin Reed Limited First Quench Retailing Mothercare Plc. Balfour Convenience Focus (DIY) Limited Music Zone Services Barry Prince & CO. Limited Fraser Hart Limited MVC Entertainment Limited Basebuy Limited Freemans Public Limited Netto Foodstores Limited Baugur UK Limited French Connection Group Night Realisations PLC Beale PLC Furnitureland Limited Oasis Fashions Limited BHS Limited Game Stores Group Limited One Stop Phone Shop BIR Realisations Limited Games Station Limited Original Shoe Company Blacks Leisure Group PLC GCG Holdings Ltd OS Realisations (2009) Blockbuster Entertainment Glyn Webb Wallpapers Ossian Retail Group Limited Borders (UK) Limited Granville Technology Group Our Price Holdings Limited Box Clever Holdings Grattan Public Limited Owen Owen Limited Brantano (UK) Limited H Realisations Limited Paperchase Products Bridge Realisations Limited Hamleys Of London Limited Passion For Perfume Brighthouse Group PLC Henson NO. 1 Limited Past Times Trading Brightside And Carbrook Heron Food Group Limited Pilot Clothing Limited C. S. Lounge Suites Limited Historical Collections Group Powerhouse Holdings Card Warehouse Group HMV Group PLC PRG Powerhouse Limited Cath Kidston Group Limited Holland & Barrett Retail Quantum Clothing Group Cellar 5 Limited Homeform Group Limited R. And M. Swaine Limited Chelsea Stores Limited HPJ UK Limited Ray Alan Limited Choicesuk PLC Hurley's Menswear Limited Redcats (Brands) Limited Ciro Citterio Menswear PLC Hurrans Garden Centres Retail Variations PLC Clinton Cards PLC Iceland Foods Group Robert Dyas Holdings Coast Fashions Limited Ilva Furniture Limited Russell & Bromley Limited Coco Ribbon Limited Intersport Great Britain Ryman Group Limited Comet Group Limited Intex Computers Limited Safeway Stores Limited Computer 3000 Limited Jacques Vert Group Limited SCS Upholstery PLC Country Casuals Limited James Beattie Limited Select (Retail) Limited Courts PLC Jessops PLC Shoe Zone Group Limited Cranham Group PLC JG Realisations Limited Shop Electric Limited Cromwells Madhouse PLC JJB Sports PLC Sit-Up Limited DFS Trading Limited JN Realisations Limited SLB Holdings Limited Diamonds & Pearls Limited John Dennis Smallbone PLC Digital POS Ltd John Menzies PLC Somerfield Stores Limited Direct Wines Limited Joy General Stores Limited Sound Control Holding Dolcis Limited K.F. Group Limited Speciality Retail Group

226

Early Learning Centre Karen Millen Fashions Stand ME Shoes Limited Effective Cosmetics Retail Kingsway Beds Limited Stead & Simpson Limited Empire Direct PLC Kwik Save Stores Limited Steinhoff UK Retail Limited Erewash Upholstery La Senza Limited Style Menswear Limited Etam PLC Land Of Leather Holdings Stylo PLC Ethel Austin Retail Group Laura Ashley Limited Superdrug Stores PLC Everything But The Music Lillywhites Limited T. J. Morris Limited Tempo Limited The Works Stores Limited Waterstones Booksellers Texstyle World Limited The new name Holdings WEW Group PLC The Bear Factory Limited Thorntons PLC Whistles Limited The Furnishings Place TJX UK Whittard Trading Limited The Furniture Factory Topps Tiles (UK) Limited Wickes Building Supplies The Gadget Shop Limited Toys "R" Us Limited WOC Realisations Limited The Garden Centre Group Trident Fashions Woolworths Group PLC The Marvellous Group UNO PLC Your More Store Limited The Officers Club Limited Unwins Wine Group Zavvi Retail Limited The Pier Retail Group USC Group Limited The Sofa Workshop Limited Walmsley Furnishing PLC

Source: Mintel 2005-2010 & FAME database

227

Appendix 2: Ratios used - 82 variables in failure prediction models Note: (xn and numbers in the table below are a cross reference to the variable annotation in these original papers)

Ratio/ Variable Beaver Altman Taffler Ohlson McGurr Bhargava Charitou Argawal (1966) (1968) (1982) (1980) and et al et al and Devaney (1998) (2004) Taffler (1998b) (2007) 1 Cash flow to sales 1 10 2 Cash flow to total assets 2 7 3 Cash flow to net worth 3 9 4 Cash flow to debt 4 7 11 5 Net income to sales 5 6 Net income to total assets (ROA) 6 6 1 1 7 Net income to net worth 7 8 Net income to total debt 8 9 Current liabilities to total assets 9 4 16 X3 10 Long-term liabilities to total assets 10 5 11 Current plus long-term liabilities 11 2 5 (TL) to total assets (TA) 12 Current plus long-term plus 12 preferred stock to total assets 13 Cash to total assets 13 2 14 Quick assets to total assets 14 X3 19 (QA)/(TA) 15 Current assets to total assets 15 X1 13 16 Working capital to total assets 16 3 8 21 17 Cash to current liabilities 17 18 Quick assets to current liabilities 18 20 19 Current ratio (CA/CL) 19 4 7 14 20 Cash to sales 20 21 Accounts receivable to sales 21 9 22 Inventory to sales (stock turn) 22 X5 10 3 37 23 Quick assets to sales 23 39 24 Current assets to sales 24 36 25 Working capital to sales 25 26 Net worth to sales 26 14 38 27 Total assets to sales 27 X5 12 6 28 Cash interval 28 29 Defensive interval 29 30 No credit interval 30 X4 31 Retained earnings to total assets X2 7 1 32 Earnings before interest and tax X3 X1 4 23 (EBIT) to total assets (TA) 33 Market value of equity to book X4 5 value of total debt 34 Total liabilities (TL) to net capital X2 6 employed (NCE) 35 Working capital (WC) to net worth X4 (NW) 36 Size (log(TA/GNP price-level 1 index)) 37 OENEG = One if TL>TA and zero 5 otherwise 38 INTWO = one if net income was 8 negative for last two years, otherwise zero 39 CHIN = Change in net income 9 22 40 Inventory per cent (stock / total 3 assets) 41 Change in long term debt % 6 42 Change in current ratio % 8 43 Fixed assets to sales 11 44 Change in total asset turnover % 13 45 Sales growth 15 46 Sales per employee 16 47 Gross margin % 17 48 Gross margin return on inventory 18 49 Gross margin return on stock 19 investment 50 Change in return on sales % 20 51 Change in working capital % 21 52 Change in working capital $ % 22

228

Ratio/ Variable Beaver Altman Taffler Ohlson McGurr Bhargava Charitou Argawal (1966) (1968) (1982) (1980) and et al et al and Devaney (1998) (2004) Taffler (1998b) (2007) 53 Change in working capital to sales 23 per cent 54 Change in equity to debt 24 55 Cash flow 2 56 Shareholders’ equity /total assets 2 57 Shareholders’ equity / total debt 3 58 Shareholders’ equity / total 4 liabilities 59 Cash flow to total assets 8 60 Debtors to cash flow 12 61 Current liabilities / current assets 15 62 Current liabilities to net worth 17 63 Current liabilities to total liabilities 18 64 EBIT /current liabilities 24 65 EBIT /fixed assets 25 66 EBIT / shareholders’ equity 26 67 EBIT /total liabilities 27 68 Income before extraordinary items/ 28 fixed assets 69 Income before extraordinary items/ 29 sales 70 Income before extraordinary items/ 30 total liabilities 71 Income before extraordinary items/ 31 total assets 72 Income before extraordinary items/ 32 shareholders’ equity 73 Working capital from operations / 33 total assets 74 Working capital from operations / 34 net worth 75 Working capital from operations / 35 sales 76 Sales / current assets 40 77 Sales / total assets 41 78 Sales / fixed assets 42 79 Market value of equity / total debt 43 80 Market value of equity / 44 shareholders’ equity 81 Profit before tax / current liabilities X1 82 Current assets / total liabilities X2

229

Appendix 3: Exploratory test survey

230

231

232

Appendix 4: Original template coding structure Coda Chairman’s Director’s Report Chief Executive’s Finance Total Coda Chairman’s Director’s Report Chief Executive’s Finance Director’s Total Statement Report Director’s Statement Report Report Report Page Number 03 04-09 10-15 16-19 03-19 1. Financial Measures

Number of pages 1 6 6 4 17 4.1 Group Sales 3 3 Named Yes Yes Yes 4.1.1 Retail 2 2 Signed 4.1.2 Entertainment 1 4 5 Date 2nd April 2008 2nd April 2008 2nd April 2008 4.1.3 Publishing 2 2 Photographs or Pictures* Neutral 21 21 1. Success 6 21 37 5 69 4.2 Group Profits 3 1 2 6 1. Failure 1 0 3 2 6 4.2.1 Retail 1 1 1 3 Paragraphs (number) 7 21 40 28 96 4.2.2 Entertainment 1 1 1 3 4.2.3 Publishing 1 1 2 1. Retail Metrics** 3.1 Stores 2 (#818) 2 4.3 Exceptional items 1 1 2 4 3.1.1 Square foot 1 1 4.4 Intangible Assets 1 1 2 3.2 Brands 1 1 4.5 Fixed Assets 1 1 3.2.1 Ladybird 1 (kids 1-10 yr.) 1 2 4.5.1 Property Portfolio 1 2 3 3.2.2 Chad Valley 1 1 4.6 Loan facility 1 1 3.2.3 Worthit’ 1 4 5 4.7 Working capital 1 1 2 3.2.4 Licences 1 1 1 1 3.3 Customers 1 2 (4m per week) 4 7 4.7.1 Cash flow 3.4 Like for like sales 1 4 5 4.8 Dividend 3 2 5 3.5 People 1 1 (30,000) 2 4.9 Shareholder return 1 1 3.6 Website (Multi channel) 1 2 3 4.10 Gross margins 7 7 3.7 Catalogue 1 (Big Red Book) 1 2 4.11 Stock control 2 1 3 3.8 Products 3 2 5 4.12 Capital Expenditure 1 1 2 3.8.1 Category 1 1 4.13 Integration Costs 3 1 4 3.8.1.1 Computer Games 1 1 4.14 Shrinkage 1 1 3.8.1.2 Toys 1 1 4.15 Unit handling cost 1 1 3.8.1.3 Confectionery 1 1 4.16 Dividends Rec’d (Royalties) 1 1 3.9 Supply Chain 1 2 3 4.17 Earnings per share 2 2 3.9.1 Order Lead Times 1 1 4.18 Debt 2 2 3.9.2 Transport 1 1 2 2 3.10 Market Reach (B2B) 2 4 6 4.19 Tax 3.11 Value Added Services 2 2 4.20 Pensions 3 3 3.12 Supplier relationships 2 2 4 4.21 Treasury 1 1 3.13 Partner of choice 1 1 4.22 Funding 3 3 3.14 Expertise - Pk. Media 1 1 4.23 Currency 1 1 3.15 Vouchers 1 1 4.24 Interest 1 1 3.16 Advertising 1 1 4.25 Credit Risk 1 1 3.17 Christmas 4 4 4.26 Going Concern 1 1 3.18 Price 1 1 3.19 Service (call)Centre 1 1 Sub total 15 0 34 33 82 3.20 Self Audit 1 1

3.21 Mystery Shopper 1 1

3.22 Competition 1 1 2 Commission Retail metrics 3 25 43 1 72

Sub total 3 25 43 1 72 Financial measures 15 0 34 33 82 Totals 18 25 77 34 154

Appendix 5: Semi-structured interview meeting agenda

AGENDA – MEETING WITH [------] [------DIRECTOR]

[RETAILER X] [DATE & TIME]

 Recording of Meeting o Ethical standards

 Formal Thank You o Time o Survey completion

 Confidentiality o Privacy of individual o Privacy and confidentiality of company information

 Review survey responses o Information content o Decision context

 Performance measurement discussion o Financial measures o Retail marketing measures o Risk measures – company beta o Other o Internal v published

 Academic ‘Case Study’ methodology o Qualitative research (no right or wrong answers) o Multi-retailer case study o Ten year history - historic data o Interviews with: Chairman, CEO; Finance Director; Marketing Director, Operations Director etc.

 AOB

Background This research is a study across multiple retailers on the content of retail performance metrics and the boardroom decision context aimed at identifying actual retail performance metrics used by retailers in the UK.

Research commenced in 2010 and is planned to complete in 2015.

MBS/TNT/ [Date]Wednesday, 13 January 2016 [email protected]

234