6-Bussin.qxd At a time when hunting for the ’next Enron’ is an international sport, companies are uncertain as to what governance decisions they should be making. The clean-up has extended to insider selling, financial disclosure, even Chief Executive Officer (CEO) pay – all issues that feed the image of corporate corruption (Useem, 2002). In South Africa many stakeholders and the media are also baying for blood and newspaper headlines like, “The fat cats are still grabbing all the cream,” (Sunday Times Business Times, April 27 2003, p. 1) hit sensitive nerves. There is a rising tide of resistance against executives who take huge pay rises despite poor earnings, both nationally and internationally. The Johannesburg Stock Exchange (JSE) will be getting tough on complying with King II (Shevel, 2003), which means that organisations are in the spotlight too. Yet Watson Wyatt’s 1999–2000 CEO Pay Study reveals a strong, positive correlation between changes in pay and corporate performance (Kay and Graskamp, 2001). The dichotomy of attraction, motivation and retention of good executives versus tough corporate governance and media spotlights, places remuneration decision-makers in a difficult position (Merchant, 1989). A company’s ability to understand the ‘drivers’ of remuneration policy and CEO pay is therefore a critical component in determining its present and future success in good remuneration governance. This study seeks to understand the factors that drive changes to remuneration policy and what the outcomes are. Well designed remuneration systems play a strategic role by promoting organisational success in highly competitive markets in which technological change constantly influences how employees perform their jobs (Martocchio, 1998). Indeed, some go so far as to argue that there are strong links between remuneration system design and organisational performance (Greenhill, 1988; Modise, 1993; Rodgers, 1999; Rynes and Gerhart, 2000; Young, 2002). Problem definition Committees meet several times a year to make decisions about the organisation’s remuneration policy, philosophy, CEO and director pay. The committees that make remuneration policy decisions range in nature from Remuneration Committees, Human Resource Committees, Finance Committees, Audit Committees, Boards of directors, Owners, Advisers and Consultants (Chingos, 1997; Corporate Leadership Council, 2001b; Fisher, 1991; Reda, 2002; Williams, 1994). Decisions are made on the basis of experience or market benchmarking, but not really on empirical evidence. This study seeks to define the empirical relationship. Scope of research Three main constructs have been isolated for this research as shown in figure 1. Figure 1: Inter-relatedness of research constructs The dependent variable is outcomes in the form of perceived impact on the organisation. Respondents indicated whether the remuneration policy changes had a positive or negative impact on the organisation. A list of factors driving remuneration policy decisions was presented to respondents who indicated the extent to which each factor drove remuneration policy decisions in their organisations. Motivation for the study Six reasons can be offered regarding the motivation for the study. These are ordered in the same order as the research constructs shown previously in figure 1, and flow from external to the organisation to internal as follows: MARK BUSSIN DEON HUYSAMEN Department of Human Resource Management Rand Afrikaans University ABSTRACT The aim of this study was to develop an in-depth understanding of the relative importance of the factors driving change to remuneration policy decision making and the impact on organisations. Data from 148 organisations was analysed and subjected to rigorous statistical analysis. The results show that the most potent drivers of remuneration policy are retention of key staff, financial results and organisation strategy. The greatest changes to remuneration policy were in the areas of variable pay, merit/performance related pay, market position, total package and job evaluation/ broadbanding policy. A strong correlation was found between the extent of change in Remuneration policy and impact on the organisation. This suggests that the greater the change in Remuneration policy the greater the impact on the organisation. OPSOMMING Die doel van hierdie studie is om die relatiewe belangrikheid van die faktore wat verandering in die vergoeding besluitnemingsbeleid dryf asook die impak wat dit op organisasie het in diepte te verstaan. Data van 148 organisasies is geanaliseer onderworpe aan streng statisiese analises. Die resultaat wys dat die mees kritieke drywers vir vergoedingsbeleid die volgende is: retensie van sleutelpersoneel, finansiële resultate en organisasie strategie. Die grootste veranderinge in vergoedsbeleid was in die volgende areas: veranderlike betaling, meriete/prestasie verwante betaling, markposisie, totale pakket en rolevaluasie/‘broadband’ beleid? Sterk korrelasie is gevind tussen die vlak van verandering in die vergoedingsbeleid en die impak op die organisasie. Dit wys onder andere uit hoe groter die verandering in vergoedingsbeleid, hoe groter die impak op die organisasie. FACTORS DRIVING CHANGES TO REMUNERATION POLICY AND OUTCOMES Requests for copies should be addressed to: M Bussin, Department of Human Resource Management, RAU, PO Box 524, Auckland Park, 2006 45 SA Journal of Human Resource Management, 2004, 2 (2), 45-54 SA Tydskrif vir Menslikehulpbronbestuur, 2004, 2 (2), 45-54 Construct 1 – Forces driving change 1. Changing nature of work 2. Investor confidence 3. Lack of knowledge Construct 2 – Remuneration policy 4. Governance Construct 3 – Outcomes – perceived impact on the organisation 5. Attracting and retaining high-performing employees 6. Return on investment Factors driving and influencing remuneration policy Remuneration policy is influenced by many factors. Decision- makers usually do not know how much ‘weight’ to give each factor, and are often reactive to situations that develop. Detecting environmental signs before others is what often gives the competitive advantage. Often, the driver is based primarily on what the committee can ‘sell’ to shareholders, and hence may not be the most appropriate solution. Understanding the extent to which each of the factors drives remuneration decision-making will inform the relative weight that needs to be attached to the driver when making the remuneration changes. The main components of remuneration policy Organisations all have different remuneration policies with different content. It is what gives them their competitive advantage (Fay, Knight and Thompson, 2001). There were, however, common themes derived from the literat ure review as indicated in A rmstrong and Murlis (1998), Bergman and Scarpello (2001), Burnett (2000), Chingos (1997), Corporate Leadership Council (2001d), Fay et al., (2001), Fisher (1991), King (2002), Martocchio (1998), White and Druker (2000), and Young (2002) and this review of the literat ure yielded 18 remuneration policy headings (with definitions). Outcomes from changes to the components Milkovitch and Newman (1999) and Lawler (1990) suggest that the impact of changes to remuneration policy on organisations can be grouped under the headings of positive and negative impact. From a remuneration point of view, Armstrong and Murlis (1998), the Corporate Leadership Council (2002), Lawler (1990) and Milkovitch and Newman (1999) define the following as positive impacts on the organisation and these are sorted into ‘employee’ and ‘company’. Table 1 provides a summary of the constructs which formed the basis of the questionnaire. The link between remuneration and company performance has been researched and a positive correlation has been found (Armstrong and Murlis, 1998; Lawler, 1990; Milkovich and Rabin, 1991). Given the background and problem statement sketched above, the specific research question for each construct was: Construct 1 – Factors driving change to remuneration policy Question 1: What factors are important in driving remuneration policy decisions? Question 2: To what extent do these factors influence decision- making? Construct 2 – Remuneration policy components Question 3: Which components of the remuneration policy were changed and to what extent were they changed? The null hypothesis is that components of the remuneration policy identified in the literature study are valid components. BUSSIN, HUYSAMEN46 TABLE 1 SUMMARY OF CONSTRUCTS Input: Stimulus Throughput: Organismic Output: Response Forces driving change Remuneration policy Impact on organisation External Forces Fixed Pay Employee 1. Advanced technological developments 1. Base pay management policy 1. Compensation arrangements are provided so 2. Competitors 2. Competence based pay policy that money issues do not become a distraction 3. Corporate failures, e.g. Enron 3. Fixed period Employment Contracts (i.e. 2. Employees are attracted, motivated, retained and 4. External advisers, e.g. Legal, Auditors, for a fixed period of time) policy understand the remuneration system Consultants 4. Fringe benefits policy 3. Productivity is positively affected 5. Investment/Stock exchange analysts 5. International remuneration (i.e. all or part of 4. There is a strong link between performance 6. Legislation, e.g. Employment Equity remuneration paid overseas) policy and pay Act, SARS audit 7. Publicity, e.g. media, customers Variable pay Company 8. Remuneration Committee 9. Shareholder expectations 6. Long term incentives (e.g. EVA, Banking of bonus 5. Governance policies are sound 10. Social upheaval/trade union over several years – not shares) policy 6. Publicity is positive 11. Surveys/benchmarking 7. Share schemes policy 7. Remuneration costs are maintained 12 Turbulence in business environment 8. Short term incentives (e.g. profit share, gain 8. Reward underpins the business strategy 13. War for talent and retention share, bonus schemes, commission) policy. 9. Shareholders, business analysts and stakeholder, are satisfied with remuneration levels, and see Internal factors Remuneration process, systems and policy the company as a good investment 10. The compensation system is legally compliant 14. Affordability/Rising costs 9. Job evaluation or broad banding policy 11. The organisation performs better financially 15. Board of directors 10. Legal compliance (e.g. Employment Equity 16. Change in culture Act) policy 17. Development/Career progression 11. Market position policy, Benchmarking practice 18. Economic restructuring/different work patterns policy 19. Financial results/organisation success 12. Merit pay/individual performance leading to a 20. Internal advisers, e.g. HR department, higher annual increase/performance related pay stakeholders (PRP) policy 21. Governance and King II report 13. Rand hedging of salaries policy 22. Strategic thrust 14. Remuneration governance policy 23. Productivity 15. Remuneration mix (the ratio of guaranteed to 24. Remuneration committee variable pay) policy 25. Retention of key staff 16. Retention strategy policy 26. Staff loyalty 17. Retirement funds policy 18. Total package concept/all inclusive remuneration/my pay -my way policy Construct 3 – Impact of remuneration policy changes on organisation Question 4: What impact did the changes to the remuneration policy make on the organisation? The null hypothesis is that a correlation exists between the extent of changes in remuneration policy and the impact on the organisation. Question 5: Are there differences in the responses for different participants? The null hypothesis is that the mean extent of change in remuneration policy is equal across categories of characteristics of organisations. METHOD Research approach The ‘blending’ methodology as a way to resolve the qualitative / quantitative dichotomy of the components of this study was used. Creswell (2003, p.10) refers to this dual approach as “triangulation” and the dual approach is sequential and dominant-less dominant design. The qualitative, less dominant phase preceded the dominant quantitative phase. The triangulated research model that was adopted, comprised two phases: � Phase 1 – A pilot study that was qualitative, involving individual interviews with experts in the field and content analysis. This qualitative step was carried out to define valid constructs for the questionnaire for phase 2 � Phase 2 – The main study that was quantitative and involved questionnaires and statistical analysis Definition of variables The dependent variable, perceived impact on the organisation, was operationalised via the five-point Likert Scale on a continuum of extremely negative impact to extremely positive impact. To operationalise the impact on the organisation, Knapp (Corporate Leadership Council, 2002), board chairman of INOVA (an American based company), has a three-fold goal in assessing compensation at the company: � Fulfil the responsibilit y for meeting any regulatory requirements � Provide compensation arrangements so that money issues do not become a distraction � Design compensation arrangements that align actions to required performance These three goals can be associated with a positive impact of remuneration on the organisation. The converse is also true. In another study by the Corporate Leadership Council (2002) the success of a compensation strategy is defined under the following elements: � Retention � Profitability � Employee feedback � Benchmarking data � Financial performance � Enforcement of business goals The primary independent variable, remuneration policy, was to be operationalised by listing each remuneration policy component and gauging the extent of the change via a four- point scale. The secondary independent variable factors driving changes was to be operationalised through the use of a semantic differential scale ranging from 1 = to no extent to 7 = to a very large extent. All major variables were obtained through personal interviews and a comprehensive review of the literature. Population and sampling Phase 1 – Pilot study (Qualitative) The purpose of the pilot was to obtain valid constructs for the questionnaire and to supplement the literat ure review. Interviews were held and the questionnaire was amended to incorporate all variables not covered in the literature review (Cooper and Schindler, 1998). The purposive method of sampling was used to identif y the sample for the pilot study (Zikmund, 1997). The outcome of this phase was to integrate the research methodology and the research questions. The population needed to be more narrowly defined where the following mix of 10 interviewees were selected: � Listed companies employing more than 5000 employees – 2 interviews with dedicated remuneration professionals � Companies employing less than 5000 employees – 2 interviews with dedicated remuneration professionals � Remuneration committee members who are not dedicated remuneration experts – 4 interviews � Managing Directors – 2 interviews This sample was to conform to criteria that serve the purpose of the study – a method commonly used in exploratory research. Welman and Kruger (2001) state that it is not possible to evaluate the extent to which such samples are representative of the relevant population. Phase 2 – Main study (Quantitative) The size of the population for the main quantitative phase of the study were people responsible for remuneration in all registered companies in South Africa. Several mail lists were used, namely South African Chamber of Commerce, all listed companies in South Africa and the 21st Century Business and Pay Solutions client database. These combined databases consist of 1700 companies. The response was considered a representative sample of the companies operating in South Africa. Measuring instrument/Data-gathering methods Two measuring instruments were used. Phase 1 – Pilot study For the first phase of the study an open-ended questionnaire from the research topic was used to generate the variables from the selected sample. Respondents were asked what the factors were that drove remuneration policy decision-making and what the components were of their remuneration policy. This qualitative method provided valid results if the questions in the interview elicited the type of information that was sought (Leedy, 1997). This data was content analysed to generate research constructs and variables in addition to the ones obtained from the literature review. Phase 2 – Main study For the second phase a closed-ended quest ionnaire constructed from theory and results of the pilot study was used. Biographical data was collected and then the constructs (independent variables) were measured to determine the relative importance of each factor in the remuneration decision. The quest ionnaire was sent to part icipants electronically and data was capt ured electronically. The data was verified on capt uring and suspicious data was telephonically confirmed. Research process Phase 1 – Pilot study The individual interviews had open-ended questions to elicit further constructs and to let respondents interpret and to avoid REMUNERATION POLICY AND OUTCOMES 47 bias based on the researchers questions. Appointments were set up with a sample of 10 interviewees. These interviewees were selected through quota sampling from the mail list. Self-administered questionnaires were used (Leedy, 1997) and data was gathered using the intercept method (Cooper and Schindler, 1998). Any constructs identified in the literature research were included at this point. Phase 2 – Main study Questionnaires were e-mailed to 1700 companies. The person who is the custodian of the remuneration policy was asked to complete the questionnaire. Creswell (2003) discusses how the items in the survey instrument must be closely aligned to the research questions, the variables from the literature and the proposed statistical analyses. The questionnaire asked closed- ended questions through an electronic form which allowed for the efficient capturing and processing of the data, and which was suitable for highly educated respondents (Cooper and Schindler, 1998). The questionnaire comprised 3 main sections. Demographical data was collected at the beginning of the questionnaire (section 1) to enable data to be analysed by classification to identif y differences relative to outcomes. Section 2 comprised factors driving change to remuneration policy using a 7-point scale to determine the extent of each factor. The extent to which the components of the remuneration policy changed since the year 2000 were tested using a 4-point scale ranging from ‘to no extent’ to ‘to a large extent’ in section 3. The impact of these changes on the organisation were also tested in this section using a 5-point scale and the option to indicate if the change was not applicable to impact on the organisation. Participants had the option of completing their details (if they want to receive an executive summary) or remaining anonymous. Statistical analysis Phase 1 – Pilot study The data was analysed for content to define valid constructs. Cooper and Schindler (1998) define content analysis as a research technique for the objective, systematic and quantitative description of the manifest content of a communication. The unit of analyses was identified as factors driving changes to remuneration policy and remuneration policy components as prescribed by Welman and Kruger (2001) for this type of study. The resultant constructs were visible in the questionnaire. The outcome of the pilot study formed the basis for phase 2 by identif ying the interviewees’ beliefs and values regarding remuneration policy and factors driving it. Consequently their behaviour was viewed as intentional and creative and it could be explained but not predicted in the outcome (Schurink, 2003). Phase 2 – Quantitative study Phase 1 resulted in certain themes and/or categories of factors driving change in remuneration policy and remuneration policy components, and phase 2 then studied these constructs in four steps from a quantitative perspective. This was done using a detailed questionnaire sent out to the participants in the study. The data was analysed by Statkon (the ‘Statistics Consultancy’ at Rand Afrikaans University). The methodology is discussed in more detail. Recorded frequencies The background information was analysed for recorded frequencies. Information on respondent position, organisation type, organisation structure, organisation size, industry sector, organisation age and whether the organisation is listed or not, was elicited by the questionnaire. Where recorded frequencies of data were too low, the data was reduced into broader categories. The background information of the recorded frequencies is referred to as the characteristics of the recorded frequencies hereafter. The broader categories of recorded frequencies of background information will be referred to as the categories of the characteristics of the recorded frequencies. Factors driving change in remuneration policy – construct 1 The Semantic Differential technique was used to illicit the strength of each factor driving change to the company’s remuneration policy over the past 3 years (i.e. since 2000). Bipolar rating scales were developed on a seven-point scale (from 1 to 7) for the constructs. Whilst the seven-point Semantic Differential scales can be highly generalisable, specific tailor- made bipolar rating scales were used that permit connotative perceptions. The chi-square test was used to determine the dependence of the categorical variables (background information) and the independent variables (factors driving change). In other words, the null hypothesis was that the extent of the factor driving change is independent of the characteristics of recorded frequencies. Testing was done at a significance level of 0.05 after the degrees of freedom for each construct had been determined. The calculated chi-square value was compared to the critical chi- square values. The outcome of this identified the factors driving changes in remuneration policy that were not independent of the characteristics of recorded frequencies. Extent of changes in components of remuneration – construct 2 A first order factor analysis was done on the 18 components of remuneration. The method used was principal factor analysis. The diagnostics performed were the Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy. Measures of Sampling Adequacy (MSA) were calculated per component. A Varimax (orthogonal) rotation was used and then the Kaiser criterion to decide on the number of components of remuneration with an extent of change in remuneration policy. The first order factor analysis was repeated on the remaining number of components to yield a number of first order factors. Various components of remuneration were grouped in each factor through the process. A reliability analysis on the first order factors was performed and tested using Cronbach’s Alpha. First order factors with a value < 0,6 were excluded. Second order factor/factors with an extent of change in components of remuneration of the remuneration policy were identified through the reliability analysis on the first order factors. Impact of changes on the organisation – construct 3 The extent of change of components of remuneration in the remuneration policy was cross-tabulated with the impact of the change on the organisation. The hypothesis was that there is not independence between the extent of change of components of remuneration in the remuneration policy and the impact of change on the organisation. The Chi-squared value was used to check that there is not independence. A correlation was assumed if Cramer’s V is greater than 0.2. The impact of the change on the organisation was grouped into 3 measures, namely, negative impact, no impact and positive impact. If ‘to no extent’ was marked for the extent of change in a component of remuneration, it was assumed to be ’not applicable’ to the impact of change on the organisation. Identify the strength of the relationship between the extent of changes and recorded frequencies of the sample data The Anova test for continuous variables was used. The null hypothesis assumed that the mean extent of change of the components of remuneration in first and second order factors were equal across the broader categories of characteristics of the sample data. The significance of the relationship between the extent of change of the components of remuneration in first and second order factors and the broader categories of the characteristics of the sample was tested using Cramer’s V. BUSSIN, HUYSAMEN48 A gap analysis was then performed on the data to illustrate the relationship between the extent of change of components of remuneration and the impact of these changes on the organisation. A comparison of the gap analysis was made between the Private sector and the Public sector broader categories of characteristics. Analysis of factors driving change to remuneration policy and outcomes of changes in remuneration policy A first order factor analysis was done on the 24 factors driving change to remuneration policy. The method used was principal factor analysis and followed the same diagnostics as that of the analysis on the extent of change in components of remuneration. A Varimax (orthogonal) rotation was used and then the Kaiser criterion to decide on the number of factors driving change to remuneration policy. Various factors driving change to remuneration policy were grouped in each factor through the process. A reliability analysis on the first order factors was performed and tested using Cronbach’s Alpha. First order factors with a value < 0,6 were excluded and then a second order factor analysis was performed on the first order factors using the same method and diagnostics. Finally the correlation between the resultant second order factor/factors and extent of change to remuneration policy and impact of change on the organisation was tested using a Pearson Correlation. Reliability and validity of the research results The central consideration concerning the process of data collection is that of reliability (Mouton and Marais, 1996). Reliability of observations or data is influenced by four variables; namely, the researcher, the participants, the measuring instrument and the research context or circumstances under which the research is conducted. In this study compliance with reliability and validity were attained through the detailed research approach, design and execution that followed to ensure that the study could be replicated and results in the same findings. Reliability tests were performed at all stages of the analysis to test the validity of the Repertory Grid (Fransella and Bannister, 1977). The Grid revealed a pattern of relationships between the constructs by revealing a pattern in the way in which the person has ranked elements. RESULTS Organisation profiles The demographic details of the survey respondents are reported from the questionnaire as shown in table 2. Analysis of research propositions and results Table 3 indicates the various factors driving change to remuneration as to the extent that they influence the remuneration policy of the organisation. It is rank ordered by the percentages in the last column – ‘To a large extent.’ The top 5 factors that drive change to remuneration policy to a large extent in terms of recorded frequency are retention of key staff, financial results, strategic thrust, surveys/benchmarking and internal advisers. Corporate failures, investment/stock exchange analysts, publicity, governance/King II report, social upheaval/trade unions drive change to remuneration policy to a lesser extent. Factors driving change in remuneration policy Results of the analysis The results of the hypothesis testing are shown in table 4. The bold figures indicate for which broader categories of the characteristics there is a dependence and is listed in increasing p-values. Numbers in bold indicate a rejection of the null hypothesis that the extent of factors driving change in remuneration policy is independent of the broader categories of the characteristics of the data. TABLE 2 CATEGORIES AND CHARACTERISTICS OF SURVEY RESPONDENTS Characteristic Broader categories of characteristic Industry sector Resources Financials Non-cyclical Cyclical Other consumer services goods 26,5% 24,3% 20,9% 15,5% 12,8% Type of Private Public company sector sector/ Parastatal 75% 25% Listing of Not listed Listed company 60,1% 39,9% Organisational Holding Subsidiary (Missing structure (local or data) international) 33,1% 43,2% 23,7% Organisation Up to >R300m >R1,51bn financial size* R300m 27,7% 29,1% 43,2% Number of Small (up Medium (501 Large employees to 500) to 5000) (>5000) 32,4% 36,5% 31,1% Number of 50 years Over 50 years or less years incorporated 48,6% 51,4% Remuneration Yes No Do not committee know 62,2% 34,5% 3,3% Position HR Other manager/ 52,7% director 47,3% * Financial size is determined by turnover, sales or budget TABLE 3 FACTORS DRIVING CHANGE TO REMUNERATION POLICY RANK ORDERED BY LARGE EXTENT Factors driving change to To a small To a moderateTo a large remuneration policy extent extent extent Count % Count % Count % Retention of key staff 8 5,4% 64 43,2% 76 51,4% Financial results 13 8,8% 66 44,6% 69 46,6% Strategic thrust 16 10,8% 67 45,3% 65 43,9% Surveys/benchmarking 18 12,2% 78 52,7% 52 35,1% Internal advisers 17 11,5% 84 56,7% 47 31,8% Affordability/rising costs 13 8,8% 91 61,5% 44 29,7% Productivity 17 11,5% 88 59,4% 43 29,1% Legislation 35 23,6% 72 48,7% 41 27,7% Board of directors 32 21,6% 76 51,4% 40 27,0% Change in culture 34 23,0% 76 51,3% 38 25,7% Your competitors 43 29,1% 69 46,6% 36 24,3% Development/career progression 19 12,8% 97 65,6% 32 21,6% Economic restructuring/ 26 17,6% 90 60,8% 32 21,6% different work patterns Remuneration Committee 60 40,5% 60 40,5% 28 19,0% Shareholder expectations 51 34,5% 74 50,0% 23 15,5% Staff loyalty 32 21,6% 95 64,2% 21 14,2% Turbulence in business 45 30,4% 82 55,4% 21 14,2% environment Advanced technological 65 43,9% 66 44,6% 17 11,5% developments External advisers 53 35,8% 79 53,4% 16 10,8% Social upheaval/trade union 73 49,3% 62 41,9% 13 8,8% Governance/King II report 46 31,1% 91 61,5% 11 7,4% Publicity 89 60,1% 51 34,5% 8 5,4% Investment/stock exchange 106 71,6% 37 25,0% 5 3,4% analysts Corporate failures 112 75,7% 33 22.,3% 3 2,0% REMUNERATION POLICY AND OUTCOMES 49 TABLE 4 COMPONENT LOADINGS ON 7 FIRST ORDER POSTULATED FACTORS Factor Extent of 1 2 3 4 5 6 7 change Internal advisers 0,691 0,122 0,177 Governance/King II 0,592 0,230 0,186 0,297 report Strategic thrust 0,518 0,384 0,201 -0,187 0,234 Remuneration 0,486 0,101 0,255 0,163 committee External advisers 0,466 0,287 0,153 Investment/Stock 0,811 0,205 exchange analysts Corporate failures 0,125 0,627 0,149 0,105 Shareholders 0,101 0,487 0,368 expectations Publicity 0,383 0,209 0,107 0,252 -0,150 Productivity 0,596 0,371 0,251 Economic 0,167 0,567 restructuring/d ifferent work patterns Change in culture 0,285 0,407 -0,158 0,229 Advanced 0,340 0,384 0,223 technological developments Development/Career 0,131 0,315 0,311 0,284 progression Financial results 0,239 0,160 0,199 0,615 0,192 Board of directors 0,145 0,440 0,165 0,214 Affordability/Rising 0,392 costs Turbulence in -0,112 0,289 0,311 0,107 0,308 0,111 business environment Social upheaval/trade 0,103 0,138 0,141 0,605 unions Legislation 0,202 0,183 0,52 Your competitors 0,120 0,201 0,653 0,106 Surveys/benchmarking 0,354 -0,126 0,446 Retention of key staff 0,256 0,163 -0,162 0,268 0,620 Staff loyalty 0,137 0,222 0,122 0,616 Seven factors are postulated according to Kaiser’s (1970) criterion (Eigenvalues greater than unit y) that explain approximately 58% of the variance in the factor space. The component loadings on these five factors are shown in table 4. Therefore the null hypothesis is accepted. First order factor analysis suggests that there is an interdependence of the components of each factor that could be assigned single descriptive labels as shown in table 5. TABLE 5 FIRST ORDER FACTORS Factor Descriptive Label 1 Governance 2 Shareholders/stakeholders 3 Organisation and work design 4 Financial and leadership 5 Legislation and agreements 6 Competitors 7 Staff retention It is now possible to understand the groupings of the forces that affect remuneration policy. Five factors are postulated according to Kaiser’s (1970) criterion (Eigenvalues greater than unity). These five factors explain approximately 58% of the variance in the factor space. These are the most important components when developing remuneration policy. The component loadings on all the factors are shown in table 6. TABLE 6 COMPONENT LOADINGS ON 5 FIRST ORDER POSTULATED FACTORS Component loadings Remuneration policy component 1 2 3 4 5 Remuneration mix 0,721 0,203 0,195 -0,114 Short term incentive 0,718 2,26 Remuneration governance 0,669 0,149 0,285 0,134 Legal compliance 0,548 0,105 0,151 Share scheme 0,547 0,38 Fixed period employment contracts 0,725 -0,107 0,323 Total package 0,347 0,651 0,134 -0,177 Fringe benefits 0,189 0,597 0,22 Base pay management -0,175 0,503 0,408 0,249 Rand hedging of salaries 0,142 0,843 0,136 International remuneration 0,107 0,14 0,768 0,102 Job evaluation/broadbanding 0,118 0,124 0,825 Market position 0,142 0,173 0,596 0,485 Merit pay/PRP 0,166 0,338 0,372 0,531 Retention strategy 0,284 0,132 0,119 0,765 Long term incentives 0,446 -0,103 0,295 -0,179 0,494 First order factor analysis suggests that there is an interdependence of the components of each factor and these have been grouped into 5 descriptive labels given in table 7. TABLE 7 FIRST ORDER FACTORS Factor Descriptive Label 1 Remuneration mix and governance 2 Guaranteed pay 3 International remuneration 4 Pay progression/PRP 5 Retention strategy These 5 descriptive labels are the major headings of remuneration policy design and are what create competitive advantage between organisations. Extent of change in components of remuneration policy and impact of changes A gap analysis was then performed on the data to illustrate the relationship between the extent of change of components of remuneration and the impact of these changes on the organisation. A comparison of the gap analysis was made between the private sector and the public sector broader categories of characteristics. Figures 2 and 3 illustrate the gap analysis for both private and public sectors, showing the relationship between the extent of remuneration policy changes and the impact on the organisation. The remuneration policy components are rank ordered by extent of change. BUSSIN, HUYSAMEN50 Results of the analysis Significance that the mean extent of change in components of remuneration policy is not equal across the categories of characteristics is evident for the following: � Holding and subsidiary organisational structures � HR Manager/Director and other positions Significance that the mean extent of impact of change in components of remuneration policy on the organisation is not equal across the recorded frequencies of characteristics is evident for the following: � Holding and subsidiary organisation structures � Different industry sectors � HR manager/director and other positions DISCUSSION This study attempted to establish and quantif y the relationship, dependence and/or correlation that exists between factors driving change in remuneration policy, the extent of changes in components of the remuneration policy and the impact of these changes on the organisation. To distil meaning from a large number of drivers and components of remuneration, the descriptive factors are submitted in table 8. The linkage between each construct is now known. Construct 1 – Stimulus/Input The factors driving change in remuneration policy are driving different extents of change in the components of remuneration policy. The extent of change of these components are interdependent and make up the remuneration policy. Factors driving change to remuneration policy (stimuli) are informed by business strategy including governance, stakeholders, organisation and work design, finance and leadership, legislation and staff retention. These factors driving change of a component of remuneration policy will have a significant impact on the extent of change of remuneration policy in an organisation. 1.00 1.50 2.00 2.50 3.00 Fringe benefits Total package Merit pay / PRP Short-term incentives Long term incentives Job evaluation Market position Base pay management Share schemes International remuneration Retirement funds Retention strategy Remuneration mix Remuneration governance Legal compliance Competence based pay Rand hedging of salaries Fixed period employment contracts 1.00 1.50 2.00 2.50 3.00 Job evaluation Short-term incentives Competence based pay Market position Merit pay / PRP Legal compliance Fixed period employment contracts Remuneration mix Total package Retention strategy Long term incentives Retirement funds International remuneration Base pay management Remuneration governance Fringe benefits Rand hedging of salaries Share schemes MO IMP 3 = Figure 2: Gap analysis of extent of change of remuneration policy and impact of change on the organisation – private sector Figure 3: Gap analysis of extent of change of remuneration policy and impact of change on the organisation – public sector REMUNERATION POLICY AND OUTCOMES TABLE 8 SUMMARY OF ANALYSIS Stimulus/Input Organismic/Throughput Response/Output (Construct 1) (Construct 2) (Construct 3) Factors driving change Components of Impact on remuneration policy organisation Governance Remuneration mix/ Positive governance Shareholders/stakeholders Guaranteed pay Neutral Organisation and work International remuneration Negative design Financial and leadership Pay progression/PRP Legislation and agreements Retention strategy Competitors Staff retention Construct 2 – Organismic/Throughput These organismic throughputs result in a response or output. If a component of remuneration policy was changed, the impact of change on the organisation has a strong dependence on the extent of the change of the component of remuneration policy. This relationship is not equal across holding or subsidiary structures, different industry sectors and may be position dependent. The extent of change of remuneration policy has a high correlation with the impact of changes in remuneration policy on an organisation (response or outputs). The literature review and pilot study yielded 18 components of remuneration policy. What was not known was the extent on change of these policies in South Africa over the past 3 years. Table 9 shows each remuneration policy component. Firstly, they are rank ordered by mean extent of change. Secondly, they are ranked by the mean impact of the remuneration policy change on the organisation. A score of 1 = small extent, 2 = moderate extent, and 3 = large extent. TABLE 9 MEAN EXTENT OF CHANGE OF REMUNERATION POLICY AND IMPACT ON ORGANISATION Sorted by extent Sorted by impact Extent Impact Extent Impact Short term incentives 2,13 2,63 2,10 2,68 Total package Job evaluation 2,11 2,52 1,91 2,68 Competence based pay Total package 2,10 2,68 2,09 2,64 Merit pay/PRP Merit pay/PRP 2,09 2,64 2,13 2,63 Short term incentives Market position 2,08 2,59 1,95 2,63 Retention strategy Long term incentives 2,06 2,60 2,06 2,60 Long term incentives Fringe benefits 2,01 2,17 2,08 2,59 Market position International 1,97 2,47 1,94 2,54 Remuneration mix remuneration Base pay management 1,96 2,44 2,11 2,52 Job evaluation Retirement funds 1,96 2,38 1,97 2,47 International remuneration Retention strategy 1,95 2,63 1,96 2,44 Base pay management Remuneration mix 1,94 2,54 1,85 2,44 Share schemes Competence based pay 1,91 2,68 1,96 2,38 Retirement funds Legal compliance 1,89 2,35 1,89 2,35 Legal compliance Share schemes 1,85 2,44 1,81 2,32 Remuneration governance Remuneration 1,81 2,32 1,76 2,25 Fixed period governance employment Fixed period 1,76 2,25 1,70 2,18 Rand hedging of employment salaries Rand hedging of salaries 1,70 2,18 2,01 2,17 Fringe benefits Construct 3 – Response/Output The greater the extent of change in remuneration policy, the greater the impact on the organisation. The policy components that had the most positive impact are total package, competence based pay, long term incentives, market position, merit pay/PRP, total package, remuneration mix, short term incentives and retention strategy. The results of this study support the relationship between the factors driving change to remuneration policy and the impact of the changes of the remuneration policy on the organisation. This finding has closed the gap in our current level of knowledge. The literature review is validated in that the components of remuneration policy were identified and not one participant added another component to the ‘other’ column in the questionnaire. Linkage to research questions A synopsis of the major findings linked to the research questions is provided below in summary format. This closes a gap in current knowledge and provides a guide to remuneration decision makers. Question 1: What factors are important in driving remuneration policy decisions? A review of the literature and pilot study yielded 24 factors that drive remuneration policy. The strength of each was determined and the top 5 drivers in order of strength are: � Retention of key staff � Financial results � Strategic thrust � Surveys/benchmarking � Internal advisers Question 2: To what extent do these factors influence decision- making? The research has shown that there is a strong correlation between the factors driving change and the extent between the factors driving change and the impact of change. The factors therefore influence decision making directly. Question 3: Which components of the remuneration policy were changed and to what extent were they changed? The research provided a good indication of which components were changed and the extent to which they were changed. For private sector participants, the top 5 changes in order of extent of change were in: � Fringe benefits policy � Total package policy � Merit pay/performance related pay policy � Short term incentives policy � Long term incentives policy In public sector and parastatals, the top 5 changes in order of extent of change were in: � Job evaluation /broadbanding policy � Short term incentives policy � Competency based pay policy � Market position policy � Merit pay/performance related pay policy Common to both public sector/ parastatals were short term incentives, and merit pay/ performance related pay policies. Variable pay continues to be an important component of the remuneration mix with it receiving the most attention in participating organisations as most organisations wish to tie the onerous salary and wage bill to the fortunes of the organisation. Wealth sharing is a concept that has gained momentum over time, especially since organised labour has warmed to the concept. BUSSIN, HUYSAMEN52 The need to review market position policy is possibly linked to the retention driver. There is also a move in the market to focus on market related guaranteed packages pegged at the median or upper quartile. This is a shift from the common wisdom over the past 10 years where organisations aimed for a lower guaranteed package relative to the market and used to aim for lower quartile to the median. Question 4: What impact did the changes to the remuneration policy make on the organisation? The greatest positive impact in the private sector in order of strength for the top 5 was in the following areas: � Total package policy � Merit/Performance related pay policy � Retention strategy policy � Short term incentives policy � Competence base pay policy For the public sector/parastatals, the greatest positive impact in order of strength for the top 5 was in the following areas: � Competency based pay policy � Long term incentives policy � International remuneration policy � Total package policy � Market position policy Common to both is the total package concept which demonstrates the need for equity, cost containment and flexibility. International remuneration in the parastatals shows a significant interest in cross border expansion. While an anticipated domain for private sector, the parastatals are facing the challenge of international pay. The strength of the impact of competency based pay in the public sector and parastatals perhaps demonstrates the usefulness of this reward vehicle where capacity building and training are big issues. Question 5: Are there differences in the responses for different participants? There are distinct differences in the extent of change and impact of change across organisation structure and industry sector and particularly between private and public sector. This is telling from a survey and benchmarking point of view, in that the best data for comparative purposes will come from organisations structured similarly and in the same industry sector. Value add: what is now known The results of this research provide useful insights to remuneration policy decision makers. Firstly, the forces driving change to remuneration policy are known as well as the strength of each force. The forces dependent on the type of organisation are also known. Secondly, the changes made to remuneration policy over the last 3 years is also known as well as the extent of the changes. This provides guidance to the focus areas and indicators as to which components need to be watched over the next few years. Thirdly, the impact of change of each component of remuneration policy on the organisation is known. This guides remuneration policy makers as to the most positive impacts on the organisation. More important is the discovery of those components that have a neutral impact on the organisation. Expectations can therefore be managed accordingly. Finally, the differences between the various t ypes of organisation are known and that has implications for the remuneration strategists regarding what to keep on the ‘radar screen.’ The estimated value for remuneration policy makers and organisations is that: � Guidelines are provided as to the forces driving remuneration policy change � The strength of each force provides a guide as to how much research and emphasis should be placed on each driving force � Guidelines as to the changes made to remuneration policy are reported and it is now known where the focus areas have been. This informs remuneration policy decision making and provides competitive advantage � The impact of each remuneration policy change on the organisation is known and this provides clear guidance on how to manage the implementation of the policy. Those policies that have a neutral impact should not be hyped up by management and implemented routinely without unrealistic expectations � Guidelines as to which driving forces play a role by organisation demographics are reported. This allows a focused approach when doing the annual compensation review Limitations of this research The following limitations need to be considered relative to this study: � The main focus was on the quantitative study and the completion of the questionnaire. This results in a respondents perception, which would have been better captured in conjunction with a more thorough qualitative analysis � The dependent variable – impact on organisation – was described in broad terms. This needs to be expanded on so as to provide deeper understanding of the impact of remuneration policy changes on organisations � This study captures management’s view of the situation. It would be useful to complement this with the views from employees Future research Emanating from this study, the following research is suggested: � There is an increasing interest in the needs of generation X and Y employees (Huysamen, 2003). It would be useful to repeat this study with a special emphasis on these categories of employees � An in-depth study of other important components of remuneration policy i.e. remuneration mix and retention is recommended. It may include design issues and application of the strategy and implementation � It is recommended that a more detailed understanding of the impact on organisations is researched. This study was limited to a positive, neutral or negative impact, which could be expanded to provide richer understanding � Lastly, the linkages between retention of key staff and the organisation scorecard measures like customer loyalty, finance, learning and growth and organisation processes could be explored Conclusion Whilst the literature review provided a useful insight into all the drivers and remuneration components, little was known about the strength, extent and links between these. This research provided answers on which to base remuneration decision making. There were expected and unexpected findings, both of which provided rich insight and scrambled some assumptions. It is hoped that this research will provide a useful base from which to build and further enhance our understanding of a subject that touches all working people’s pockets. Money matters. REFERENCES Armstrong, M. & Murlis, H. (1998). Reward management. London: Biddles Ltd. Bergman, T. & Scarpello, V. (2001). Compensation decision making. Fort Worth: Harcourt College Publishers. Burnett, J. (2000). The relative importance of the components of remuneration structures for managers. Unpublished MBA research report. Johannesburg: Universit y of the Witwatersrand. REMUNERATION POLICY AND OUTCOMES 53 Chingos, P. (1997) Paying for performance – The current climate. In P. Chingos and KPMG Peat Marwick LLP Compensation and Benefits Consultants (Eds.), Paying for performance: A guide to compensation management. (pp. 242 – 256) New York: John Wiley and Sons. Cooper, D.R. & Schindler, P.S. (1998). Business research methods. Boston: McGraw-Hill. Corporate Leadership Council (2001b, November). Legal responsibilities of compensation committees. Washington DC: The Corporate Adviser’s Board. Corporate Leadership Council (2001d, May). Executive compensation philosophies. Washington DC: The Corporate Adviser’s Board. Corporate Leadership Council (2002, March). Executive compensation metrics and selection. Washington DC: The Corporate Adviser’s Board. Creswell, J. W. (2003). Research design. California: Sage Publications, Inc, Thousand Oaks. Fay, C., Knight D. & Thompson, M. (2001). The executive handbook on compensation. New York: The Free Press. Fisher, J. (1991). The Role of the Compensation Committee. Executive compensation: A strategic guide for the 1990s. Boston, Massachusetts: Harvard Business School Press. Fransella, F. & Bannister, D. (1977). A Manual for repertory grid technique. London: Academic Press. Greenhill, R. (1988). Performance related pay. England: Director Books. Huysamen, D. (2003, 30 June – 4 July). Module One: Executive Human Resources Update. Lecture presented for ‘The 5-day MBA: What every executive needs’ conference. Kay, I. & Graskamp, E. (2001, February 1). Year 2000 executive pay: Superior pay for superior performances. ACA News, (pp. 29 – 30). King, M. (2002). The King report on corporate governance for South Africa. Institute of Directors. Lawler, E. (1990). Strategic pay. San Francisco: Jossey-Bass Publishers. Leedy, P. (1997). Practical research planning and design. (6th Ed.). Upper Saddle River, New Jersey: Prentice Hall, Inc. Martocchio, J. (1998). Strategic compensation. New Jersey: Prentice-Hall, Inc. Merchant, K. (1989). Rewarding results. Boston: Harvard Business School Press. Milkovich, G. & Newman, J. (1999). Compensation. Boston: Irwin Mc Graw-Hill. Milkovich, G. & Rabin, B. (1991). Executive Compensation and Firm Performance. In F.K. Faulkes (Ed.). Executive compensation: A strategic guide for the 1990s. (pp. 81 – 97). Boston, Massachusetts: Harvard Business School Press. Modise, J. (1993). Executive compensation and corporate performance, Unpublished MBA report. Johannesburg: University of the Witwatersrand. Mouton, J. & Marais, H. (1996). Basic concepts of the social sciences. Pretoria: HSRC publishers. Reda, J. (2002). Compensation committee handbook. New York: John Wiley and Sons, Inc. Rodgers, C. (1999). Managers’ perceptions of a total package compensation system. Unpublished Master of Management Research Report, Johannesburg: University of the Witwatersrand. Rynes, S. & Gerhart, B. (2000). Compensation in organisations. San Francisco: Jossey-Bass. Schurink, W.J. (2003 March). Workshop on Qualitative Research. Lecture presented at Rand Afrikaans University, Johannesburg. Shevel, A. (2003, April 13). JSE to get tough on King II compliance. Sunday Times, Business Times. Times Media Limited: Johannesburg, (pp. 22). Useem, J. (2002, September 16). In corporate America it’s cleanup time. Fortune, (pp. 23 – 28). Useem, J. (2003, April 28). Have they no shame. Fortune, (pp. 23 – 27). Welman, J.C. & Kruger, S.J. (2001). Research methodology. Cape Town: Oxford. White, G. & Druker, J. (2000). Reward management: A critical text. London: Routledge. Williams, A. (1994). Just reward? The truth about top executive pay. London: Kogan Page. Young, Warren. (2002). A post-mortem of corporate failure – An analysis of the effect that remuneration policy has on directors and consequent corporate failure, Unpublished MBA research report, Pretoria: Gordon Institute of Business Science. Zikmund, W. (1997). Business research methods. Fort Worth: The Dryden Press. BUSSIN, HUYSAMEN54