Insider perspectives on director remuneration governance deliberations

Abstract

Purpose

Companies around the globe increasingly receive immense shareholder scrutiny due to perceivably excessive executive director remuneration. The debate in South Africa intensifies due to severe pay inequality. The authors thus accounted for the perspectives of asset managers and listed financial services companies in South Africa pertaining to the impact of voting and engagement on director pay policies and practices.

Design/methodology/approach

Semi-structured interviews were conducted with selected asset managers, chief executive officers, chief financial officers and remuneration committee members of listed financial services companies to gauge their views on the impact of shareholder activism endeavours on remuneration governance. The qualitative data was analysed by conducting thematic analysis.

Findings

Most of the asset managers and financial services representatives preferred proactive, private engagement on pay concerns, given the impact thereof on voting outcomes, and ultimately director remuneration practices and policies. Independent remuneration committees have a prominent role in facilitating engagements with investors to ensure fair remuneration.

Research limitations/implications

The consequences should be clearer if organisations receive substantial votes against their pay policies and implementation reports. South African regulators can consider the “two-strikes” rule to ensure that action is taken in response to shareholder voting on director remuneration matters.

Originality/value

Representatives of asset managers and listed financial services investee companies offered valuable insights on remuneration governance deliberations in an emerging market. This in-depth analysis highlights the importance of proactive engagement to ensure that corporate leaders are paid fairly.

Citation

Van Zyl, M. and Mans-Kemp, N. (2023), "Insider perspectives on director remuneration governance deliberations", Personnel Review, Vol. 52 No. 3, pp. 799-816. https://doi.org/10.1108/PR-01-2021-0032

Introduction

Executive director remuneration is a highly contentious and evergreen human resource management topic (Proxy Insight, 2020). Practitioners, the media, shareholder activists and researchers increasingly place focus on director pay policies and practices around the globe (Stathopoulos and Voulgaris, 2016). Shareholder activists can be defined as investors who attempt to bring about change in investee companies, as they are unsatisfied with their management, operations and/or practices (Gillan and Starks, 2007). Shareholders can implement several mechanisms to focus attention on pressing financial, as well as environmental, social and governance (ESG) concerns, including voting and voicing concerns at annual general meetings (AGMs) and private discussions (Goranova and Ryan, 2014).

Say on pay is a widely researched remuneration governance consideration (Baixauli-Soler et al., 2020; Wu et al., 2020; Denis et al., 2020; Obermann and Velte, 2018; Stathopoulos and Voulgaris, 2016; Kimbro and Xu, 2016; Brunarski et al., 2015; Krause et al., 2014; Conyon and Sadler, 2010). Shareholders can accordingly express their views on director pay by either voting for or against the remuneration policies and reports of investee companies or abstaining (Baixauli-Soler et al., 2020).

Advisory say on pay voting has been implemented in several countries, including the United Kingdom (UK), United States of America (USA) and South Africa (Baixauli-Soler et al., 2020; Viviers et al., 2019). Researchers focused on the effectiveness of say on pay to encourage directorates, human resources committees and remuneration committees (REMCOs) to promote fair emolument that are aligned with a company’s interests, and by implication its shareholders’ welfare (Kimbro and Xu, 2016; Stathopoulos and Voulgaris, 2016).

Obermann and Velte (2018) emphasised that behavioural aspects, including the perception of fairness, considerably influence deliberations between investee companies and their investors on director remuneration. Given that South Africa has the highest wage inequality in the world, with a Gini coefficient of 0.63 (International Labour Organization, 2018), it is unsurprising that the remuneration practices and policies of companies listed on the Johannesburg Stock Exchange (JSE) receive considerable attention from shareholders and researchers [Padia and Callaghan, 2020; Viviers et al., 2019; PricewaterhouseCoopers (PwC), 2018a].

Since 1994, when the first democratic election took place in South Africa, the King Reports offer corporate governance guidance to listed companies in the country. The first King Report has been named after Judge Mervyn King, a former Judge of the Supreme Court of South Africa. Judge King was the first chair of the King Committee on Corporate Governance that was formed in 1992. This committee has developed four King Reports since 1994. The Institute of Directors in South Africa (IoDSA) has published four King Reports to date. The first King Report was revised in 2002 (King II), 2009 (King III) and 2016 (King IV) to align the local corporate governance guidelines with global developments and respond to changes in the local corporate landscape (IoDSA, 2016, 2009).

In essence, the King IV Report comprises a set of voluntary principles and leading practices that apply to all organisations in the country, including listed and unlisted entities (IoDSA, 2016). In line with the JSE Listings Requirements (2017), listed companies are expected to indicate how they have applied the King IV guidelines and explain deviations. The latest King guidelines, inter alia, cover board composition, director remuneration and board committee structures (IoDSA, 2016).

The manner in which institutional investors exercise their ownership rights can considerably enhance corporate governance practices in investee companies (Zhou et al., 2016). The King IV Report defines institutional investors as “the holders of beneficial interest in the securities of a company; [including] retirement funds and insurance companies as well as the custodians, nominees and service providers who act under mandate in respect of any investment decisions and investment activities exercised in relation to those securities” (IoDSA, 2016, p. 13). In contrast, retail investors invest on their own behalf.

South Africa is one of the few countries that officially encourage institutional investors to invest in a responsible manner. The Code for Responsible Investing in South Africa (shortly known as CRISA) was published in 2011 to guide institutional investors on the application of the King III Report. Regulation 28 of the local Pension Funds Act (No. 24 of 1956) was furthermore revised to ensure that institutional investors take ESG factors into account (IoDSA, 2011, 2009). Regulation 28 and CRISA were in the process of being revised at the time of publication.

Principle 17 in the King IV Report likewise encourages responsible institutional investment practices. This report also suggests that JSE-listed firms should engage with shareholders should their remuneration policies and/or implementation reports receive 25% or more dissenting votes. The King IV Report also places focus on two well-researched shareholder activism mechanisms, namely say on pay and engagement (IoDSA, 2016).

As REMCOs typically assist boards with emolument arrangements and engagements (IoDSA, 2016), they are deemed “the facilitators” of director pay for the purpose of this article. A major challenge that REMCOs face is how to design director emolument packages that align corporate governance principles and shareholders’ interests and also inspire directors to act in the best interests of their organisations (PwC, 2018a). As such, REMCOs should carefully apply their minds and engage with investors on board remuneration matters (IoDSA, 2016). Through questioning and improving remuneration approaches, REMCOs and institutional investors, including pension funds and asset managers can prevent corporate scandals and create fair director pay packages to facilitate sustainable value creation (PwC, 2018a).

Given considerable focus on the compilation of director pay packages in South Africa (Padia and Callaghan, 2020; Viviers et al., 2019), the country’s substantial wage gap (International Labour Organization, 2018) and well-developed corporate governance framework (IoDSA, 2016), the authors have accounted for the views of selected asset managers and financial services investee companies on director remuneration deliberations in South Africa. Semi-structured interviews were conducted with 12 asset managers and 11 representatives from JSE-listed financial services organisations, including leading executives and REMCO members. The article hence offers an insider view on remuneration governance issues. Pertinent attention was given to pay-related shareholder activism endeavours, as financial services investee companies’ REMCOs are likely to better understand the views, and effectively address the pay-related concerns and voting outcomes of asset managers if they engage in robust discussions. In turn, activism endeavours of asset managers can considerably impact director pay policies and practices in future.

The behavioural agency model was used as the theoretical lens, as explained next. An overview of shareholder activism mechanisms in the context of director emolument is then offered. Thereafter, the semi-structured interview process is discussed, followed by the thematic analysis findings. Based on the conclusions, recommendations are offered to improve director remuneration policies and practices in future.

A behavioural perspective on the agency theory

Investigations of director pay (Pepper et al., 2013) and shareholder activism on emolument matters (Goranova and Ryan, 2014) are often based on the agency theory. Jensen and Meckling (1976) explained that the separation of corporate ownership and control results in issues between shareowners (principals) and managers (agents) that need to be resolved.

Incentives could be offered to managers and executive directors in an attempt to ensure that they account for shareholders’ best interests (Goranova and Ryan, 2014). Yet, misalignment between executive pay and performance increasingly provokes discontent from shareholders, as it represents a lost opportunity to alleviate the agency problem (Goranova and Ryan, 2014; Ertimur et al., 2011). The effectiveness of long-term incentives to address the agency problem is hence questionable (Pepper et al., 2013).

In line with the agency theory, shareholders should disapprove high executive rewards when an investee company performs poorly, thereby symmetrically assessing gains and losses (Krause et al., 2014). The agency theory (Jensen and Meckling, 1976) hence restricts the risk-taking behaviour of agents to risk aversion or neutrality, while arguably ignoring risk-seeking behaviour (Wiseman and Gomez-Mejia, 1998).

However, corporate leaders are not necessarily rational decision-makers. Furthermore, human behaviour could be dependent on an individual’s unique interpretation, that is in turn experience dependent (Weick, 1970). Wiseman and Gomez-Meija (1998) thus argued that the agency theory should be amended to incorporate behavioural considerations. They combined the agency theory of Jensen and Meckling (1976) with the prospect theory of Kahneman and Tversky (1979) to derive the so-called behavioural agency model of managerial risk taking (hence forth referred to as the behavioural agency model).

The prospect theory postulates that the way in which an individual perceives possible outcomes of a decision situation is based on the framing of these outcomes as potential gains or losses. The outcomes of a scenario should thus be compared relative to a reference point, such as current wealth, rather than the absolute outcomes (Kahneman and Tversky, 1979). In this context, it can be demotivating for directors to receive deferred incentives, as such incentives might never realise. Deferred incentives are also considerably devaluated when being discounted (Pepper et al., 2013). Flammer and Bansal (2017, p. 1827) linked these issues to the so-called time-based agency problem.

Wiseman and Gomez-Meija (1998) therefore suggested that the agency theory’s assumption of risk aversion should be replaced with loss aversion in the context of director emolument. They reasoned that self-interested directors are less concerned about maximising their future wealth than minimising current losses. This view has substantial implications for the time-period that REMCOs set for performance-linked incentives (Pepper et al., 2013) and directors’ responses to shareholders’ emolument concerns.

Shareholder activism mechanisms and director pay

Institutional and retail investors have a range of mechanisms at their disposal to bring their discontent about corporate matters, including seemingly excessive pay packages, under directors’ attention. Details will be provided on how institutional investors can use the threat of exiting, voting and engagement to respond to remuneration governance concerns.

Application of Hirschman’s exit-voice framework in the investor context

Hirschman (1970) explained that if consumers note a deterioration in the quality of products, they can either exit (thereby refraining from buying products from the company in future) or bring their concerns under management’s attention (by using their “voices”) (Hirschman, 1970). This typology can also be applied by activist retail investors and institutional investors, such as asset managers (Goranova and Ryan, 2014). If investors are unsatisfied with the actions and/or policies of an investee company, they can sell their shares. Alternatively, they can raise their concerns with representatives of investee companies by using a range of private and public voice mechanisms (Viviers and Smit, 2015; Goranova and Ryan, 2014).

Private voice mechanisms include arranging discussions behind closed doors with corporate leaders and writing emails to explain pay-related concerns (Viviers and Smit, 2015). Timely private deliberations with dissident institutional investors can protect a company against considerable reputational damage, as the outcomes are typically not shared with other investors (Goranova and Ryan, 2014). Public voice mechanisms include casting say on pay votes at the AGM (Viviers and Smit, 2015).

In line with the behavioural agency model (Wiseman and Gomez-Meija, 1998), if managerial pay is linked to share price performance, managers are more likely to engage with key shareholders to prevent considerable financial losses (Admati and Pfleiderer, 2009). Managers might accordingly be more willing to discuss concerns with asset managers than retail investors to prevent a potential large loss, especially if a powerful institutional investor threatens to exit.

Say on pay voting

Shareholders can cast advisory votes on specific director pay aspects in several jurisdictions (Stathopoulos and Voulgaris, 2016), including South Africa (IoDSA, 2016). However, the effectiveness of this public shareholder activism mechanism to curb director pay is questionable, especially if the voting outcome is non-binding (Viviers, 2015).

Shareholders in multiple countries tend to exhibit limited interest (Chalaczkiewicz-Ladna, 2019). Hemphill (2019) remarked that most USA shareholders seem to be “satisfied” with executive pay practices, given the low percentage against votes. In addition, less than 10% of shareholders in the UK voted against investee companies’ remuneration policies and reports between 2016 and 2020. When accounting for abstentions, the annual figures remain below 10% (FTI Consulting and Proxy Insight, 2020). Similar results were reported in South Africa (Viviers and Smit, 2015).

Some scholars deem advisory say on pay a value-enhancing remuneration governance monitoring mechanism, given that some companies change their pay practices in response to negative voting results (Kimbro and Xu, 2016; Ferri and Maber, 2013; Ertimur et al., 2011). In contrast, other authors noted that voting outcomes had limited success in curbing executive pay packages (Wu et al., 2020; Conyon and Sadler, 2010). Denis et al. (2020, p. 3131) reported that say on pay can have “spillover effects”, as some companies respond to weak pay-related votes at peer companies by reducing executive pay.

Non-binding say on pay was introduced in the UK in 2002 (Baixauli-Soler et al., 2020). Since 2013, shareholders in this developed country can cast binding votes on executive remuneration policies and listed companies are required to enhance their remuneration disclosure (Wu et al., 2020). The level and quality of shareholder engagements considerably improved since the latest reform (Deloitte, 2016). However, Chu et al. (2021) found little evidence that this reform affected the level of executive pay, pay-performance sensitivity and the pay gap in the UK. Although Wu et al. (2020) reported that the binding vote resulted in greater accountability for poor performance, they noted that more should be done to align executive pay with multiple performance measures.

Krause et al. (2014) suggested that focus should be placed on “what” informs shareholders’ say on pay decisions. They argued that investors do not necessarily symmetrically access financial gains and losses when voting on executive pay, as suggested by the agency theory (Jensen and Meckling, 1976). Krause et al. (2014) applied the agency theory and Tversky and Kahneman’s (1991) model of loss aversion to the context of investor voting. Krause et al. (2014) accordingly postulated that shareholders are likely to respond more strongly to director pay concerns if they are in a loss than in a gain position.

Shareholders should therefore respond more negatively to an agency-related loss, such as high executive incentives despite low company performance, than to an agency-related gain, for instance low director rewards while the company had high performance (Krause et al., 2014). Obermann (2020) confirmed that voting dissent on director emolument is lower when companies meet or beat earnings forecasts. Fisch et al. (2017) also noted that shareholders are less critical of large executive pay packages if companies are performing well. Engagements with asset managers prior to AGMs can furthermore have a substantial impact on voting outcomes.

Proactive and reactive engagements

The actual power of say on pay arguably does not lie with the action of casting votes, but rather the negotiation power linked to influential shareholders’ right to vote on director emolument matters. Institutional shareholders can use their substantial negotiation power to request private engagements with managers, directors and/or REMCOs to discuss pay concerns prior to voting at the AGM. This type of engagement is defined as proactive, while reactive engagement occurs in response to shareholder voting outcomes (McCahery et al., 2016; Goranova and Ryan, 2014).

In line with the behavioural agency model (Wiseman and Gomez-Meija, 1998) and Krause et al.’s (2014) behavioural perspective on shareholder voting, institutional investors and corporate leaders are likely to address concerns before the AGM that might otherwise result in substantial welfare losses. Although proactive engagement might be effective to bring about corporate reform, it is not subject to shareholder approval. Furthermore, information discussed during private meetings are typically not shared with other investors (Goranova and Ryan, 2014).

Despite such concerns, a growing number of institutional investors are requesting proactive private meetings with representatives of investee companies to discuss performance-linked emolument and pay equity (Tonello and Gatti, 2019). Bauer et al. (2015) noted that institutional investors are more likely than retail shareholders to withdraw pay-related proposals following successful proactive engagements.

In the case of substantial votes casted against investee companies’ remuneration policies and implementation reports, the King IV Report recommends that proposed amendments to pay policies and practices should be discussed during reactive meetings (IoDSA, 2016). While such shareholder engagements are typically private, public platforms can also be used to explain reactive measures. Enhanced transparency on remuneration deliberations will enable shareholders of different sizes to make more informed voting and investment decisions (PwC, 2018a; Goranova and Ryan, 2014).

Although engagements can take place between investors and board members, management, remuneration and/or human resources committees, leading executives, such as chief executive officers (CEOs) and chief financial officers (CFOs) typically facilitate remuneration governance deliberations (PwC, 2018b). Engagements with South African institutional investors generally take place “behind closed doors” (Viviers and Smit, 2015. p. 30). However, there is inherent conflict of interest if corporate leaders should motivate their pay structures to institutional investors during private engagements.

Remuneration committees as emolument facilitators

The REMCO was introduced in corporate governance codes around the globe as a mechanism to address the agency problem and managerial power concerns (Main et al., 2008). The committee has three primary roles: to ensure that directors receive fair and responsible pay that is aligned with their firm’s long-term objectives; facilitate the publication of understandable, transparent and accurate remuneration reports and establish effective remuneration policies (IoDSA, 2016).

Information provided by REMCOs should by implication enable shareholders to make informed investment and voting decisions. Remuneration-related information can also be discussed during proactive and reactive engagements. However, if REMCO members do not have sufficient expertise and well-developed negotiation skills, pay arrangements might favour directors at shareholders’ expense (Jensen and Murphy, 2004). Committee members should thus undergo training to properly understand and fulfil their responsibilities (PwC, 2018b; Main et al., 2008).

Large executive incentives are often justified by REMCOs by stating that it is standard practice to reward performance. In addition, they typically argue that proper incentives should be offered to entice talent and that bonuses should reflect market-related compensation (Bohlander and Snell, 2010). Committee members should acknowledge, and challenge, this inherent tendency to follow the customary practices of peer REMCOs when establishing pay packages (Main et al., 2008). Luiz (2006) furthermore urged REMCOs to be cautious when obtaining external assistance, as advisors can contribute to excessive remuneration arrangements.

Pertaining to the composition of JSE-listed companies’ REMCOs, the King IV Report specifies that most of the members should be independent non-executive directors (NEDs) (IoDSA, 2016). There is a trade-off between experience and independence of directors. While longer tenured directors are typically more experienced, their independence might be impaired (Patro et al., 2018). Vafeas (2003) found that the involvement of senior directors on the REMCO is related to higher CEO pay and hence suggested that term limits should be set for REMCO members. The King IV Report suggests a nine-year term limit for independent directors (IoDSA, 2016). The study’s theoretical framework that was formulated based on the preceding discussion is displayed in Figure 1.

As shown in Figure 1, REMCOs should ideally facilitate emolument discussions between directors and institutional investors, specifically asset managers for the purpose of this study. By incorporating the views of Wiseman and Gomez-Meija (1998) and Krause et al. (2014), selected asset managers’ implementation of voting and engagement mechanisms and the possible responses of corporate agents have been illustrated in the context of director remuneration.

Research methodology

The perspectives of selected asset managers, REMCO members and CEOs and CFOs of selected listed financial services companies in South Africa were gauged on remuneration governance deliberations by conducting semi-structured interviews. Given the size of assets under their management and their related negotiation power, asset managers can more easily arrange engagements with company representatives to discuss emolument concerns than retail investors. In turn, REMCO members should facilitate remuneration discussions and decision-making, while executives are the receivers of the negotiated pay packages. Furthermore, CEOs and CFOs often conduct pay-related engagements with influential investors (PwC, 2018b).

Ahrens and Khalifa (2013, p. 24) provided support for the usage of a qualitative approach when conducting governance research by stating that it is “an important alternative to treating the various components and processes of corporate governance as black boxes whose key elements are assumed to be somehow ‘standard’ and thereby known without context specific inquiry”. The usage of semi-structured interviews allowed in-depth reflection on the views of influential investors and informed representatives of financial services investee companies on director remuneration governance deliberations, while accounting for the South African corporate context.

Development of the semi-structured interview guide

The semi-structured interview guide comprised open-ended questions based on the remuneration guidelines included in the King IV Report (IoDSA, 2016) and relevant literature. This guide was discussed with a corporate governance expert to ensure clarity and practical relevance before commencing with the interviews.

Questions covering the nature of the selected JSE-listed companies and asset managers were included in the first part of the interview guide. Company representatives were asked to indicate the approximate market capitalisation of their organisations. They were also requested to indicate whether their employers adhere to the Organisation for Economic Co-Operation and Development’s (OECD, 2015) Principles of Corporate Governance. Asset manager interviewees were asked whether their employers subscribe to CRISA. Other questions posed to the asset manager representatives included: What is the size of assets under your management; and is the asset manager that you represent a signatory of the United Nations Principles for Responsible Investment (PRI)?

In the second part of the interview guide, questions covered director pay-related deliberations, including voting and engagements. Specific questions were formulated for representatives of financial services companies, including:

1. Do you support a non-binding or binding vote on executive remuneration?

2. Have you ever obtained 25% or more dissenting votes on your remuneration policy and/or implementation report?

3. Have you experienced shareholder activism pertaining to director remuneration at AGMs?

4. Did engagements result in any change(s) regarding director remuneration practices and/or policies?

The questions that were formulated for asset managers included:

1. Do you support a non-binding or binding vote on executive remuneration?

2. How often have you witnessed investee companies receiving 25% or more dissenting votes?

3. Which concerns have resulted in voting against a remuneration policy and/or implementation report?

4. Do you engage with investee companies on behalf of your clients regarding director remuneration?

5. What were the objectives to engage and what were the outcomes?

Sample description

A combination of judgement and snowball sampling was employed to select the participants. An industry contact facilitated initial contact with South African asset managers, as well as banks and insurance companies listed on the JSE. Thereafter, interviewees provided contact details of other potential participants. Financials is a leading industry in South Africa that accounts for a considerable part of the JSE’s total market capitalisation (PwC, 2019).

Interviews were conducted with 12 representatives from ten asset managers that directly invest in equities/bonds listed on the JSE. All the selected asset managers invested funds on behalf of retail clients (individuals), businesses, institutions and pension funds. The approximate size of their assets under management ranged between ZAR500m and ZAR900bn.

According to Viviers (2015), institutional shareholder activism gained momentum when the PRI published their principles for responsible investment in 2006. Seven of the asset managers that participated in this study were signatories of these principles, while eight of them subscribed to CRISA. The industry experience of the asset manager participants ranged from seven to 33 years. They were well-informed on the King IV remuneration guidelines.

Interviews were also conducted with 11 representatives from five financial services companies that had a primary listing on the JSE. Their market capitalisation ranged between approximately ZAR27bn and ZAR309bn. A REMCO member and the CEO or CFO of each sample company participated in the study. An executive director suggested that an interview should also be conducted with their head of remuneration. The REMCO is a sub-committee of the board that typically comprises independent NEDs, while CEOs and CFOs are top-ranking, highly paid decision-makers (IoDSA, 2016). The industry experience of the financial services participants ranged from 21 to 47 years.

All the financial services interviewees indicated that their companies adhered to the OECD (2015) principles of corporate governance. These principles provide sound guidance to organisations to achieve economic objectives and enhance investor confidence. Two of the financial services participants specifically mentioned the importance of the OECD’s (2015) “disclosure and transparency” principle, given the relevance thereof to director remuneration.

Data analysis, trustworthiness and ethical considerations

Braun and Clarke’s (2006) six-step thematic analysis approach was adopted to analyse the transcribed interviews. Initial inductive coding was conducted on the transcribed data. Thereafter, preliminary themes were identified and reviewed. While finalising the themes and sub-themes, the number of times that responses related to the identified themes were taken into account. Conclusions were then drawn based on the finalised themes related to remuneration governance endeavours. Computerised software was not employed to analyse the collected primary data.

Applicable data extracts have been included in the results discussion to enhance thematic analytical validity. Pertaining to transferability and confirmability, a thorough description of the research context allows reflection on the applicability of the results in other settings. Some findings were clarified with participants to ensure that their views are accurately displayed.

Ethical clearance was obtained to conduct this study. The participants received the interview guide and consent form prior to each interview.. It was made clear that their participation was voluntary and anonymous. They could withdraw from the study at any time. One interviewee declined recording the interview, in which case substantial notes were made.

Findings and discussion

All asset manager representatives mentioned regular engagements on executive pay. The majority of the asset manager interviewees stated that they rarely engage on NED fees, as such fees are generally “fairly structured”. The interviewed CEOs, CFOs and REMCO members confirmed that they are frequently approached by asset managers to discuss executive pay practices and policies.

Although the asset manager representatives and financial services interviewees emphasised the importance of communication on remuneration matters, most of them deemed existing engagement platforms inadequate. A representative of a financial services company stated that the most common platform to discuss remuneration matters is AGMs. Yet these meetings are typically not well attended by shareholders. An asset manager representative agreed that AGMs are becoming less effective and urged companies to proactively reach out to shareholders to discuss pay-related aspects.

Details will now be provided on proactive engagement, say on pay, reactive responses to voting outcomes and the responsibilities and composition of REMCOs as emolument facilitators.

Preference for proactive engagement on director remuneration

In line with literature (Semenova and Hassel, 2019; Viviers and Smit, 2015), the majority of the asset manager participants and most of the financial services representatives preferred proactive private engagements to discuss director pay concerns above reactive engagements and public discord at an AGM. An asset manager representative remarked that “in South Africa there is pressure on asset managers to take a more proactive role”. Governing bodies were hence urged to engage with shareholders before finalising directors’ pay packages. Another asset manager interviewee suggested that a “governance roadshow or even a remuneration roadshow” could be a good opportunity to engage with key stakeholders prior to the AGM.

Representatives from the selected financial services companies indicated that proactive engagements provide REMCOs with valuable information on concerns based on remuneration policies. The asset manager interviewees agreed and explained that negative say on pay voting outcomes could be prevented by punctually resolving issues and clarifying uncertainties.

Deloitte (2015) likewise indicated that proactive engagement improves relations between companies and their shareholders, since trust and credibility could be enhanced. In addition, Bauer et al. (2015) reported that institutional investors often withdraw their remuneration proposals following successful proactive engagements. Based on the responses of the financial services interviewees, it was evident that the proactive claims of asset managers are prioritised above those of retail investors. This tendency largely relates to asset managers’ considerable negotiation power linked to their voting power, the threat of exit and potential negative publicity (Goranova and Ryan, 2014).

Decisions that are taken during private engagements are typically not subject to shareholder approval and details on such discussions are generally not shared with other shareholders (Yamahaki and Frynas, 2016). Enhanced transparency on remuneration deliberations with influential shareholders is likely to enlighten other investors’ decision-making.

Views on advisory say on pay votes

A leading executive director who participated in this study remarked that “King’s [IV] refreshed non-binding vote [linked to engagement] comes at a time when there is a lot of mistrust”. Two asset manager representatives added that governing bodies are accountable to shareholders to implement the approved remuneration policy without changes.

The opinion raised by Viviers (2015) that voting on director pay in South Africa should rather be binding was supported by the majority of the asset manager representatives. They explained that a binding vote is likely to have more onerous consequences than a non-binding one. Skovoroda et al. (2018) warned that the potential “threat” of binding say on pay could weaken over time if shareholders do not actually exercise their voting power.

The asset manager interviewees were of the view that there are not adequate consequences following disapproval of remuneration. An asset manager representative stated that without clear consequences, the outcome would be a “tick box exercise”. This participant hence suggested that the Australian “two-strikes” rule should be implemented in South Africa in future to ensure that action in taken by JSE-listed companies in response to substantial say on pay voting dissent. This rule entails that if a company receives 25% or more votes against its remuneration report for two consecutive years, the entire board (excluding the CEO) may be considered for re-election or removal by shareholders (Australian Securities Exchange Corporate Governance Council, 2019).

However, an asset manager interviewee argued that shareholders do not sufficiently apply their minds to the remuneration policy and “rather focus on the quantum”. This participant added that even if considerable value was created, shareholders might still vote against the remuneration policy. Another asset manager interviewee cautioned that if a binding vote would be implemented, the minority shareholders might “become insignificant in the equation”.

All the leading executives and REMCO members who were interviewed supported the non-binding say on pay vote. They indicated that such a vote “keeps you aligned to what your shareholders want, and it keeps you informed of how your shareholders feel” and allows the “full shareholder base, not just the major shareholders” to voice their consent or dissent.

As all individuals who participated in this study have been involved in the financial services industry group, either as asset managers, REMCO members or CEOs at selected financial services companies, they can be deemed insiders. The emolument of both groups (asset managers and company representatives) depends on remuneration decisions regarding senior financial positions. In this context, the choice between binding or non-binding say on pay arguably impacts the power balance between asset managers and financial services investees. The interviewed asset managers and financial services interviewees disagreed about whether votes casted on pay at AGMs should be binding or non-binding. Yet these insiders shared a preference for proactive discussions behind closed doors. These results thus point towards a “power struggle” between the considered insider groups.

Similarly, private discussions are likely to keep proxy advisors that financial services investees dislike outside “the inner circle”. Two REMCO interviewees argued that it would be difficult to implement a binding vote in South Africa, due to the “overreliance” of asset managers on proxy advisors to assist with voting decisions. The view was raised by a leading executive director that proxy advisors follow a “formulaic tick box type approach without necessarily understanding the complexity and nuances of an organisation”. In addition, this CEO warned against potential conflict of interest by stating that proxy advisors often offer advisory services to a company after recommending shareholders to vote against its remuneration policy. Researchers concur that institutional investors prefer to obtain information from proxy advisors, given the considerable costs related to conducting independent research on each proposal in their substantial portfolios (Malenko and Malenko, 2019; Tingle, 2016).

Responses to voting dissent on remuneration policies and implementation reports

All the asset manager representatives stated that proactive engagements before companies release their remuneration policies are preferred above reactive responses to voting outcomes. However, if JSE-listed investee companies receive 25% or more votes against their remuneration policies and/or implementation reports, they must reactively engage with investors (IoDSA, 2016). One of the asset manager interviewees raised the view that reactive measures can cause “a lot of unnecessary tension with the relationship that a board should have with shareholders”.

A representative of a financial services company stated that shareholders often vote against a remuneration policy or implementation report due to “ignorance” as they “either did not understand it completely or they did not have all the information or the background”. A REMCO member added that, following engagements in response to receiving more than 25% against votes, they realised that the “communication in the policy and implementation reports [was] not clear enough”. As a result, shareholders misinterpreted the information. This financial services participant hence urged REMCOs to ensure “transparent and uncomplicated” remuneration explanations. Ertimur et al. (2013) likewise reported that engagements with key shareholders in response to substantial votes casted against director pay at S&P 1500 companies considerably improved shareholders’ understanding of remuneration policies.

The view was raised by an asset manager interviewee that the King IV Report has “brought shareholders and investee companies into greater dialogue”, as companies increasingly invite dissenting shareholders to a public call to address emolument concerns. However, the majority of the asset manager representatives, REMCO members and CEOs/CFOs who participated in this study deemed teleconferences ineffective for this purpose. An asset manager interviewee recommended that organisations should rather engage individually with large shareholders, while a teleconference could be considered for other (minority) shareholders.

Views on the responsibilities and composition of remuneration committees

Focus was placed in this study on the REMCO as a key emolument facilitator. In line with literature (Appiah and Chizema, 2015), most of the asset manager and financial services participants deemed REMOCs’ duties quite complex. An asset manager interviewee stated that REMCOs should understand “their specific industry’s issues”. A REMCO member added that this committee should provide the governing body with a “level of comfort that the company’s overarching approach attract, retain and motivate talent”. A REMCO chairman elaborated by stating that REMCOs have a duty towards shareholders and stakeholders to ensure “transparent and uncomplicated” explanations within the remuneration policy. This participant explained that such explanations could substantially reduce misinterpretation of remuneration information.

Representatives of asset managers and financial services companies highlighted that REMCOs should account for aligning shareholders’ and other stakeholders’ interests when reflecting on director pay. Asset manager participants remarked that “if the company looks after the other stakeholders as well, it would mean that shareholders actually would generate a higher return” and “company stakeholder engagement sits at the heart of understanding societal and environmental concerns”.

Obermann and Velte (2018) confirmed that shareholder activism can have a considerable impact on the wider stakeholder community if institutional investors influence REMCOs to incorporate sustainability aspects, such as the six capitals, into remuneration policies. The International Integrated Reporting Council (IIRC, 2021) refer to six capital sources, namely financial, manufactured, intellectual, human, social and relationship and natural capital. Capital sources are defined as “stocks of value that are increased, decreased or transformed through the activities and outputs of the organization” (IIRC, 2021 p. 6).

Despite encouragement to link director pay to broader social goals, and by implication divergent capital sources (Yarram and Adapa, 2020), director incentives are typically linked to short-term financial gains (Padia and Callaghan, 2020; Kakabadse et al., 2004). Scholars and practitioners (Schoenmaker and Schramade, 2019; Melloni, 2018) hence increasingly focus on how director emolument can be linked to sustainable value creation by accounting for the IIRC’s (2021) six capitals. Instead of focusing on short-term financial targets, REMCOs are encouraged to set more balanced pay-performance objectives (Melloni, 2018).

Asset manager interviewees were of the view that REMCOs are often unsuccessful in explaining executives’ key performance indicators in a manner that reassures shareholders that long-term value has been created. One of them mentioned that REMCOs should actively engage with shareholders “to get their buy-in and inputs before actually finalising and formulating the remuneration policy”.

Another asset manager urged REMCOs to ensure that long-term incentives comprise the largest portion of executive pay packages. However, executives’ time preferences for receiving incentives sooner rather than later might be misaligned with those of shareholders, in line with the so-called time-based agency problem (Flammer and Bansal, 2017, p. 1827). As such, REMCOs should account for discounting of deferred incentives (Pepper et al., 2013).

The interviewees representing asset managers and financial services companies had differing opinions regarding whether it is truly important that the majority of the REMCO members should be independent NEDs, as suggested by the King IV Report (IoDSA, 2016). Four asset manager representatives mentioned engagements with REMCOs on director independence concerns.

An asset manager interviewee indicated that independence gives investors some assurance that their objectives “will be considered in certain remuneration [practices] to incentivise management, without any bias, or any influence from management teams”. A REMCO member, however, stated that “it is far more important that [directors] have the right skills and background”. This interviewee suggested that organisations should appoint directors with “independence of mind” to challenge the information on which remuneration is based.

An asset manager representative stated that there should be a balance of REMCO members with a short tenure that could challenge executives and experienced, long-tenured members. A “forced rotation policy” whereby “indispensable” REMCO members could be nominated to serve on the committee for an additional term was thus suggested. A REMCO member furthermore believed that a long tenure could be beneficial as it “ensures knowledge and makes sure there is not an over-reliance on advisors”. PwC (2018b) likewise stated that the REMCO should caution against overreliance on remuneration consultants when contemplating director emolument polices and packages.

Conclusions and recommendations

South Africa is classified as the country with the highest wage inequality in the world (International Labour Organization, 2018). It is thus not surprising that shareholders increasingly criticise the size and compilation of South African directors’ remuneration packages (Viviers et al., 2019). The King IV Report on corporate governance pays considerable attention to director emolument considerations. This report recommends that shareholders should pass non-binding votes on the remuneration policies and implementation reports of investee companies. Engagements should take place with shareholders if at least a quarter of them voted against the remuneration policy and/or implementation report (IoDSA, 2016).

Given their substantial assets under management and related negotiation power, asset managers can have a considerable impact on emolument deliberations and say on pay voting outcomes. In turn, REMCO members should facilitate the determination of director pay packages and compile emolument reports, while leading executives receive large remuneration packages for managing organisations. Semi-structured interviews were hence conducted with representatives of selected JSE-listed financial services companies and South African-based asset managers to gauge their views on remuneration governance deliberations. As the King IV Report focuses on voting and engagement on remuneration matters, pertinent attention was given to participants' views on these shareholder activism mechanisms.

The CEOs, CFOs and REMCO members who participated in this study gave priority attention to asset managers’ director remuneration concerns. Private engagements offer the opportunity to resolve uncertainties that might otherwise have resulted in significant say on pay voting dissent. The asset manager representatives likewise preferred proactive private engagements with company representatives to discuss pay policies above reactive engagements. All South African asset managers are therefore encouraged to proactively join the dialogue on director remuneration to ensure that emolument packages are fairly structured. They should also caution against overreliance on proxy advisors when casting their votes and making investment decisions.

The key finding that asset managers and financial services investees prefer proactive discussions behind closed doors might be ascribed to the fact that both groups are insiders that might benefit from refraining from public discord at AGMs on pay-related matters. In turn, their divergent views on non-binding say on pay could possibly be ascribed to a potential power struggle.

Based on the findings, REMCO members should ensure that remuneration reports offer understandable information in an uncomplicated format to ensure informed engagement and voting decisions. In addition, REMCOs should focus on improving alignment between shareholders’ and other stakeholders’ interests, specifically employees, when determining pay packages by accounting for the six capitals, sustainable value creation and fair pay across the corporate spectrum.

The representatives of asset managers and financial services companies disagreed on whether binding or non-binding say on pay is preferable. Although, the CEOs, CFOs and REMCO members largely supported the non-binding vote explained in the King IV Report, the consequences if the remuneration policy and/or implementation report is rejected should be clearer. The South African government could consider implementing the Australian “two-strikes” say on pay rule. This rule might encourage companies to give more attention to shareholders’ pay-related concerns in a timely manner. Organisations could furthermore create electronic systems to encourage feedback on voting outcomes and pay concerns. Other governance matters could also be reported more easily via an electronic channel.

Based on the reported findings, the actual power of say on pay prescribed in the King IV Report does not per se lie with the action of casting votes. This shareholder activism mechanism offers asset managers considerable negotiation power to impact director pay policies and practices. In future, researchers can thus focus on outcomes of voting and engagements by comparing data for several emerging countries, such as Brazil, Russia, India, China and South Africa.

Reference to the PRI and responsible investment in the interview guide might have biased the information provided by respondents. Future scholars can thus distribute a survey to a broader stakeholder group, including proxy advisors and employees’ pension fund representatives to gauge their views on fair and responsible pay practices in an emerging market context. The scope of a future study could be expanded beyond the financial services industry group by determining the opinions of representatives of listed companies operating in different industries in South Africa and Australia on the effectiveness and application of say on pay and engagement practices. These countries’ divergent say on pay guidelines should be considered when comparing the findings.

Timely engagements among asset managers, REMCO members and corporate leaders can substantially impact director pay practices and policies. As such, asset managers can play a considerable role to enhance debate on fair emolument in the country with the highest pay inequality globally.

Figures

Figure 1

Theoretical framework