Concerns about exorbitant executive compensation are making headlines, because executives receive lucrative packages despite state-owned enterprises (SOEs) performing poorly. It appears as if chief executive officers (CEOs) are not being held accountable for the performance of the SOEs.
The purpose of the study was to determine whether the size and the industry of an SOE had an impact on CEO compensation packages.
A greater understanding of the relationship between CEO remuneration and the size and type of industry of SOEs would assist with the standardisation of CEO remuneration and linking CEO pay to SOE performance.
A multiple regression analysis on a pooled dataset of 162 panel observations was conducted over a 9-year period. Financial data of 18 SOEs were extracted from the McGregor BFA database and the annual reports of SOEs.
The findings show that the size of an SOE does not influence the total compensation of CEOs. However, larger SOEs pay larger bonuses due to these SOEs being in a stronger financial position to offer lucrative bonuses. CEO’s remuneration was aligned within certain industries.
The findings emphasise the need to link CEO compensation with SOE performance. Standardisation in setting CEO compensation and implementing performance contracts should be considered.
The study confirms that CEO pay is not linked to performance and not justified when considering SOE size or industry.
Concerns about exorbitant executive compensation are not new, as noises were made back in 2008 about executives receiving lucrative packages despite state-owned enterprises (SOEs) performing poorly. The former Minister of Finance, Trevor Manuel, expressed the concern that excessive salaries were unjustified in the context of South Africa’s 23% unemployment rate (Theunissen,
State-owned enterprises, which are independent companies that are partially or wholly owned by the government, play a significant role in the South African economy. The four largest local SOEs are Transnet, Denel, Telkom and Eskom (Wendy Owens & Associates,
Despite the attention executive compensation has received, it appears as if government has become morally paralysed and unwilling to take action to ensure equality for all. The pay gaps between CEOs and average employees among the nine prominent SOEs showed that the average employee earned R58 9964.00, and the average CEO was paid R5.53 million (BusinessTech,
Given the widening gap between rich and poor, the disappearance of the middle class and the increasing number of people living below the breadline, it has become necessary to review those practices that threaten good governance and undermine equality (Mhlanga,
There is a general consensus that executive compensation has become excessive, given that (1) executive pay is inequitable relative to other employees’ pay and (2) the amounts are unjustified, compared to the SOEs’ performance (Nichols & Subramaniam,
Despite SOEs being the principal drivers of the formal sector of the economy, providing the bulk of economic growth, the performance of SOEs frequently comes under public scrutiny (Kanyane & Sausi,
Several South African SOEs had to be rescued by government to keep them afloat. Corruption and mismanagement have also been blamed for the billions of rand in losses which these companies have recorded in recent years (Mutiso,
In 2015, the government spent nearly 10% of its total annual budget in servicing debts and paying to help struggling SOEs. For example, South African Airways (SAA) reported a loss of R2.5bn during 2015. Smit (
It seems as if South Africa still lags behind in terms of implementing measures to curb exorbitant CEO compensation packages. Given the above challenges in local SOEs and unprecedented government policy interventions, the need for further research appears to be justified, especially as existing research has not managed to deliver a simplified understanding of the determinants of executive compensation in the South African context.
The first purpose of this research was to determine whether there is a relationship between CEO compensation and company size. This purpose is based on a PhD study by Bezuidenhout (
Sub-question 1: Is there a relationship between CEO
Sub-question 2: Is there a relationship between CEO
Duffhues and Kabir (
Sub-question 1: Is there a relationship between CEO
Sub-question 2: Is there a relationship between CEO
This article argues that the type of industry and company size in setting executive compensation within South African SOEs offers an explanation on how executive compensation is determined. It further provides insight on how challenges and constraints experienced in setting executive compensation could be overcome. The results should inform decisions about standard practices to control the perceived excessive pay of CEOs within South African SOEs.
The CEO, who is appointed by the board of directors, is responsible for leading the company in achieving its corporate goals (Shaw,
Compensation is a broad concept, but for the purposes of this study – and as defined by 21st Century Pay Solutions Group (
fixed pay – basic salary and employee benefits
variable pay – short-term incentives (STIs) (annual cash bonuses)
total compensation – fixed pay plus STIs.
Executive compensation packages usually consist of basic salary, benefits, STIs and long-term incentives (LTIs), and therefore a combination of fixed and variable pay (Bussin,
According to Bebchuk, Fried and Walker (
The wage gap continues to be a challenge in South Africa’s unequal society. In 2014, Mergence Investment Managers conducted an analysis of pay practices among the top 10 companies listed on the Johannesburg Stock Exchange (JSE) and found an upward trend over the past 5 years, with the gap between total compensation and average employee compensation increasing from just under 120 times in 2009 to over 140 times in 2013 (Lamprecht,
A study by Theunissen (
The Mergence Investment Managers’ analysis of variable compensation packages in 2012 and 2013 furthermore showed that approximately 50% of CEOs received 100% or more of the value of their fixed pay as variable compensation (Bezuidenhout,
Given this dire state of affairs, measures have been put in place to address concerns regarding CEO compensation. In this regard, the King IV report on Corporate Governance focuses on CEO compensation. Unfortunately, the report refers to fairness and responsibility but no clear guidelines are provided on exactly how CEO compensation needs to be determined. The report does, however, recognise that CEO compensation should be determined in the context of overall employee compensation in the organisation (Myburgh & De Costa,
In 2010, the Department of Public Enterprise (DPE) commissioned a compensation review with the purpose of determining the degree to which SOE compensation practices comply with DPE guidelines. Noteworthy findings revealed that SOEs do not follow the guidelines, and there is no standardisation in the way compensation is determined, nor were employment contracts, detailing the tasks and responsibilities of CEOs, compiled (Crafford,
A key problem with the current SOE compensation framework is the non-existence of a centralised authority to oversee such compensation, resulting in SOE boards and CEOs determining their own pay structures (Massie, Collier, & Crotty,
The allocation theory of control (Rosen,
In a market equilibrium, the most talented executives occupy top positions in the largest firms, where the marginal productivity of their actions is greatly magnified over the many people below them to whom they are linked’. (p. 182)
This implies that there should be a direct relationship between CEO compensation and the size of the company when the market is balanced. In much of the literature (Canarella & Gasparyan,
Various studies have outlined a positive correlation between executive compensation and company size (Lippert & Moore,
Chalmers, Koh and Stapledon (
The rule of thumb is that a CEO’s pay increases by 3% for every 10% increase in company size (Van Blerck,
Dai (
Research results linking CEO compensation to company performance differ across industry sectors. The results of related studies, particularly in South Africa, vary and are inconclusive, as many did not consider whether the company performance measures chosen bore any relation to executive compensation in different industries (Blair,
Dai (
The research made use of a positivistic philosophy and a deductive approach. The research methodology was, in essence, descriptive, exploratory and archival in nature, while the time horizon was longitudinal. The methodology was quantitative, as that allowed the researchers to identify relationships among two or more variables and, based on the results, to confirm or challenge existing theories or practices. The quantitative research approach made use of descriptive and inferential statistics.
Given that the researcher collected information from public companies’ annual reports, which had been subjected to financial audit, the data were regarded as accurate and credible.
The population of the study comprised 21 schedule 2 SOEs operating in South Africa. An SOE was included only if the annual reports were available on either the McGregor BFA database or the company website, and it had a 9-year financial history which revealed the CEO’s compensation. Eighteen SOEs were subsequently included in this study.
This study used two components of CEO compensation:
This study focused on company size and industry as independent variables. To determine the size of the SOEs, this study used the DPE’s organisation size grid (categorised according to revenue or assets) (see
State-owned enterprise categorisation – assets and revenue.
SOE size | Assets | Revenue | SOE category |
---|---|---|---|
A | > R16.3bn | > R2.54bn | Very large SOE |
B | R1.55bn – R16.3bn | R243.2m – R2.54bn | Large SOE |
C | R143.5m – R1.55bn | R22.8m – R243.2m | Medium SOE |
D | Up to R143.5m | Up to R22.8m | Small SOE |
SOE, state-owned enterprises; R, South African rand; m, million; bn, billion.
The target population for the study was schedule 2 SOEs. Using the definition of the
This study used the Statistical Package for the Social Science programme (SPSS version 22) for the descriptive analysis of the data. EViews (version 8), a software package for econometric analysis, forecasting and statistics, was used to run multiple regression models on the pooled dataset comprising a cross-section of 18 SOEs over a 9-year period. In an article, Polakow (
Multiple regressions were used to study the separate and collective contributions of organisation size and industry towards variances in CEO compensation components. According to Albright, Winston, Zappe and Broadie (
Secondary data, such as CEOs remuneration and the organisations’ financial performance, was collected from the annual reports of SOEs. Appropriate statistical techniques were used and information was not manipulated. Unisa provided ethical clearance for the study.
Chief executive officer compensation components for dataset (2006‒2014).
Variable | Fixed pay (R’000) | Total compensation (R’000) |
---|---|---|
Mean | 2863266.34 | 4663172.36 |
Median | 2582000.00 | 3989017.50 |
SD | 1348299.09 | 2863294.56 |
Skewness | 0.84 | 1.57 |
Kurtosis | 0.64 | 3.83 |
Minimum | 468000.00 | 636000.00 |
Maximum | 7751643.00 | 19108837.00 |
CEO, chief executive officer; SD, standard deviation.
According to the data in
The
The median of
Fixed pay (R‘ 000) (2006–2014).
Year | Mean | SD | Median | Percentage change |
---|---|---|---|---|
2006 | 1994250.19 | 1052027.05 | 1679000.00 | 23% increase |
2007 | 2372378.39 | 1242189.05 | 2062141.50 | |
2008 | 2509763.41 | 1325793.61 | 2044607.00 | - |
2009 | 2668468.03 | 1203410.04 | 2470000.00 | 3% increase |
2010 | 2769787.70 | 1034832.47 | 2550500.00 | |
2011 | 3160985.56 | 1394699.82 | 280850000 | - |
2012 | 3586606.11 | 1243883.04 | 331996400 | decrease |
2013 | 3184005.83 | 1459638.89 | 3182000.00 | |
2014 | 3523151.89 | 1487536.39 | 3063420.50 |
SD, standard deviation.
As can be expected,
The increase in the median of
Fixed pay (2006–2014).
Clearly, CEOs’
Total compensation (R‘ 000) (2006–2014).
Year | Mean | SD | Median | Percentage change |
---|---|---|---|---|
2006 | 3332067.96 | 2265677.94 | 2325750.00 | 35% increase |
2007 | 3807600.78 | 2136055.98 | 3132787.50 | |
2008 | 4237731.59 | 2744345.78 | 3970035.00 | - |
2009 | 4802590.06 | 2716499.95 | 4525037.50 | 13% decline |
2010 | 4531525.29 | 2300189.77 | 3959000.00 | |
2011 | 4868698.06 | 2666919.72 | 4111500.00 | - |
2012 | 5743642.19 | 3174628.91 | 4641500.00 | 12% decline |
2013 | 4577509.56 | 2634924.46 | 4072000.00 | |
2014 | 5241013.27 | 2695857.11 | 4490227.27 | - |
SD, standard deviation.
Total remuneration (2006–2014).
Clearly,
The first objective of this study was to determine whether there is a relationship between
The regression model included 119 unbalanced panel observations and 17 cross-sectional units over a period of 9 years. The regression model was run with an optimum model, where it was determined which (1) company performance measures (turnover, net profit and irregular, fruitless and wasteful expenditure) and (2) CEO demographic variables had an effect on fixed pay (Bezuidenhout,
Regression analysis –
Variable | Beta coefficient | Std. error | ||
---|---|---|---|---|
Constant | 3042350.00 | 2945625.00 | 1.03 | 0.30 |
Large company | 1413989.00 | 2452275.00 | 0.58 | 0.57 |
Very large company | 1368887.00 | 2434449.00 | 0.56 | 0.58 |
Weighted statistics:
Std., standard; DW, Durbin Watson Statistic.
The results in
A regression analysis was performed with the optimum model where it was determined which company performance measures had an effect on total compensation (operating profit, net profit, liquidity ratio, return on capital employed and irregular, fruitless and wasteful expenditure).
Regression analysis –
Models | 1 | 2 |
---|---|---|
Constant | 2307917.00 (1.26) | 3781641.00 (5.50) |
Large company | 1649044.00 (0.85) | - |
Very large company | 2796956.00 (1.48) | 1263352.00 (1.81) |
33.32 (0.00) | 38.06 (0.00) | |
DW stat | 2.73 | 2.74 |
0.68 | 0.67 | |
Adjusted |
0.6471 | 0.6478 |
DW stat, Durbin Watson Statistic.
Note: (1) Coefficients reported with
The last regression model, Model 2, was regarded as the optimum model, as the
The second objective of this study was to determine whether there is a relationship between
The regression model included 144 unbalanced panel observations and 18 cross-sectional data over a period of 9 years. A regression analysis was performed with the optimum model, where it was determined which company performance measures had an effect on
Regression analysis –
Models | 1 | 2 | 3 |
---|---|---|---|
Constant | 2786737.00 (5.27) | 2821658.00 (12.69) | 2909617.00 (13.80) |
Aviation and aerospace | 120566.30 (0.19) | - | - |
Development funding | 54069.37 (0.09) | - | - |
Energy | −957402.40 (−1.44) | −996300.00 (−1.55) | −1112492.00 (−2.26) |
Forestry | −145147.50 (−0.20) | - | - |
Telecommunications | 509854.40 (0.78) | 447263.70 (1.02) | |
28.60 (0.00) | 43.77 (0.00) | 52.28 (0.00) | |
DW stat | 2.50 | 2.50 | 2.48 |
0.68 | 0.68 | 0.65 | |
Adjusted |
0.63468 | 0.64217 | 0.64197 |
DW stat, Durbin Watson Statistic.
Note: (1) Coefficients reported with
The last regression model, Model 3, was regarded as the optimum model, as the
The regression model included 144 unbalanced panel observations and 18 cross-sectional data variables over a period of 9 years. A regression analysis was performed with the optimum model where it was determined which company performance measures had an effect on total compensation (operating profit, net profit, total irregular expenditure, liquidity ratio and return on capital employed).
Regression analysis –
Models | 1 | 2 | 3 |
---|---|---|---|
Constant | 4795968.00 (3.17) | 4903129.00 (7.40) | 4762020.00 (7.45) |
Aviation and aerospace | 101233.50 (0.06) | - | - |
Development funding | −500115.80 (−0.28) | - | - |
Energy | −921734.20 (−0.50) | −933546.50 (−1.07) | - |
Forestry | −1234459.00 (−0.60) | −1292823.00 (−0.80) | −1130791.00 (−0.69) |
Telecommunications | 811901.90 (−0.28) | - | - |
23.19 (0.00) | 32.38 (0.00) | 37.09 (0.00) | |
DW stat | 2.70 | 2.73 | 2.74 |
0.662 | 0.660 | 0.659 | |
Adjusted |
0.63380 | 0.64036 | 0.64184 |
DW stat, Durbin Watson Statistic.
Note: (1) Coefficients reported with
The last regression model, Model 3, was regarded as the optimum model, as the
Relationship between
Question | Compensation | |
---|---|---|
Fixed pay | Total compensation | |
Is there a relationship between CEOs’ compensation and the size of a South African SOE? | No | Yes – very large SOE |
Is there a relationship between CEOs’ compensation and the industry of a South African SOE? | Yes, within the energy industry | Yes, within the forestry industry |
CEO, chief executive officer; SOE, state-owned enterprise.
This study aimed to determine whether there was a relationship between the
in a market equilibrium, the most talented executives occupy top positions in the largest firms, where the marginal productivity of their actions is greatly magnified over the many people below them to whom they are linked. (Rosen,
This reasoning provides a theoretical basis for a positive relationship between CEO compensation and company size (Zhou,
Jeppson et al. (
Findings from this study show that industry affects both
The findings of this study offer new insights into the importance of company size and industry for the design, development and management of executive compensation practices, especially in the South African SOE environment. It further provides insight into research and practice in establishing CEO compensation in South African SOEs.
Even though CEOs want to be paid more for their skills, experience and performance, SOEs’ resources are limited and they face budgetary constraints. This forces compensation managers to address a basic economical fact, namely that of scarcity. In the war for talent and limited skills in the market, companies compete for competent CEOs. The reward committees of SOEs thus have to consider the following issues when setting executive pay levels:
The size of the organisation has an influence, depending on how the SOE compares in size to similar organisations.
The challenges presented by the industry within which the SOE operates.
It is acknowledged that South African SOEs have limited benchmarking opportunities when setting executive pay levels (Maloa,
In managing the compensation of CEOs in a fair and responsible manner, it is recommended that relevant stakeholders consider the following:
The development of an overarching framework for remunerating the CEOs of schedule 2 SOEs, in line with recommendations by the Presidential Review Committee on SOEs.
Inconsistencies and a lack of checks and balances exist in the implementation of transformation involving executive compensation. Compensation specialists should enforce compliance with the
That SOE remuneration committees continue to manage total compensation for SOE executives to ensure that company size is not used as the main reason for high total compensation.
This study aimed to contribute to a better understanding of CEO compensation. Much of the public debate on CEO compensation has highlighted the steady erosion of income equality, and the growing wage gap that has accelerated in recent years. Of great concern were not only the exorbitant compensation packages of CEOs, but also the poor performance of South African SOEs. There is an urgent need to hold CEOs accountable for SOE performance. High unemployment, the downgrading of South Africa to junk status, social unrest, service delivery strikes and (perhaps most importantly) the vast majority struggling to make ends meet are all indicators of a need for proper governance and ethical leadership in SOEs.
This research has contributed to a better understanding of measuring and setting CEO compensation against indicators that are significant to a particular industry as well as company size. However, the willingness and ability of SOEs to implement and apply the findings remains to be seen.
This article is based on the doctoral thesis ‘The relationship between CEO remuneration and company performance in South African state owned entities’, published in 2016 by Unisa, Pretoria.
The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.
M.C. wrote the article and was the co-supervisor for the study. M.L.B. provided assistance with the presentation of the results. This research is based on the PhD study of M.L.B.
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Data sharing is not applicable to this article as no new data were created or analysed in this study.
The views expressed in this article are the views of the authors and do not represent the position of the University of South Africa.