Understanding Post-Privatisation Performance of Statutory Bodies Subject to Government Shareholding—A Suggested Theoretical Framework, for Malaysian Researchers

Abstract

The purpose of this concept paper is to suggest a theoretical framework for understanding the post-privatisation performance of statutory bodies, subject to government shareholding. We identify a suitable model, from the analytical economics literature. We argue that this model is a manifestation of agency theory. Our proposed framework for using this theory is replete with examples from Malaysia. We conclude that in Malaysia, the principal determinant of whether government subsidisation enhances or erodes shareholder wealth is the level of government shareholding. We also predict that in Malaysia, the relation between shareholder wealth and government shareholding follows an “inverted U” shape. However, the turning is likely to vary, cross-sectionally and temporally. We believe that the framework presented within this paper can be used to understand empirical results reported by other Malaysian studies into the shareholder wealth effects arising from economic policies featuring close co-operation between the public and private sectors.

1. Introduction

Boycko et al. (1996) present an analytical model of shareholder wealth impacts resulting from the privatisation of statutory bodies. The model is a suitable tool for researchers from developing countries, particularly where government policies are based on the ideology of New Public Management (Siddiquee 2006). The current concept paper argues that the Boycko et al. (1996) model would be a suitable lens for Malaysian capital markets researchers. This paper critiques the theory, with specific reference to the Malaysian setting. It then provides guidance to assist Malaysian researchers in implementing the Boycko et al. (1996)model in the context of the Malaysian institutional environment. In particular, we exemplify our view that the framework we present can be used to interpret previous Malaysian evidence.1

The principal objective for this concept paper is to enhance the rigor of capital market research from Malaysia in response to the calls from review papers for more corporate governance research in Asia. These review papers have urged Asian researchers to adapt theories to their countries’ institutional settings and acknowledge differences from Anglo-American environments (Claessens and Fan 2002; Claessens and Yurtoglou 2013). A secondary objective is to illustrate the process of critiquing a theory and assessing the theory’s strengths and weaknesses. This critique focuses on the three principal dimensions of a “theory”: its assumptions (abstractions from reality), the analyses and the predictions (Perkins 2010). In achieving these aims, this paper facilitates capital market research in other developing countries, particularly from the Association of Southeast Asian Nations region (Sinnadurai 2018). The discussion should be read in conjunction with Sinnadurai et al. (2021). This concept paper provides Malaysian examples to illustrate the arguments presented in Sinnadurai et al. (2021) regarding the applicability of the Boycko et al. (1996) model to Malaysia.

There are two principal conclusions of this paper. The first conclusion is that in Malaysia, there is one principal determinant of whether government shareholding in (and extension of subsidies to) listed companies enhances or erodes shareholder wealth. This determinant is the level of government shareholding. The second conclusion is that the association between shareholder wealth and the level of government shareholding, in Malaysia, follows an “inverted U” shape.

The Boycko et al. (1996) model may be regarded as a manifestation of agency theory (Jensen and Meckling 1976). The Boycko et al. (1996) model predicts parameters and settings, conducive to government share ownership in privatised, listed companies (and hence subsidisation of these companies) constituting an agency mechanism, rather than an agency cost. The suggestions in this concept paper facilitate interpretation and synthesis of prior Malaysian evidence regarding the shareholder wealth effects of government share ownership in, and subsidisation of, listed companies. To the authors’ knowledge, this paper is the first attempt to facilitate usage of the Boycko et al. (1996) model, in this regard, for subsequent empirical researchers.

2. Outline and Purpose of the Underpinning Analytical Economic Model

The Boycko et al. (1996) model is premised on agency theory. This theory regards a listed company as a “nexus of contracts” between principals and agents. A “contract” is any arrangement, expressed or implied, in which parties have reciprocal mutual obligations. In these contracts, principals delegate decision-making authority to agents to act on the principals’ behalf. However, it is assumed that both principals and agents act in their self-interests, possibly to the detriment of the well-being of other stakeholders and possibly in breach of their contractual obligations. In listed companies, characterised by separation of management and ownership, a key contract is between shareholders (as principals) and managers (as agents). Managers are employed by shareholders to select corporate policies that maximise shareholder wealth. In situations where the interests of managers do not align with shareholder wealth maximisation, managers are predicted to act dysfunctionally, selecting suboptimal corporate policies, with respect to maximising shareholder wealth. This results in a “residual loss”, meaning lower shareholder wealth than if the managers had selected optimal policies. This residual loss is borne by both managers and shareholders. The shareholders, acting rationally, price-protect via downward adjustment to managers’ remuneration, so that the managers bear the residual loss of their dysfunctional behaviour (Shirley and Walsh 2000; Sinnadurai 2018).2

However, both managers and shareholders have an incomplete information base. Shareholders cannot accurately observe the extent to which management have engaged in dysfunctional behaviour and therefore cannot be sure that they have price-protected sufficiently. Similarly, managers cannot observe the extent to which shareholders have price-protected and cannot be confident that they have not been excessively penalised for dysfunctional behaviour. Hence, both parties have incentives to institute agency mechanisms: devices that re-align the interests of managers and shareholders (Jensen and Meckling 1976; Shirley and Walsh 2000; Sinnadurai 2018).

Most influences on the corporate environment may either exacerbate agency costs or constitute an agency mechanism, depending on the circumstances. For example, product market competition creates more survival difficulty for management of incumbent players. This would increase the likelihood of corporate failure and hence, bankruptcy costs, amplifying the agency problem. Conversely, product market competition may increase management’s exposure to the disciplinary threats of the markets for managerial labour and corporate control. In this regard, product market competition would constitute an agency mechanism (Babar and Habib 2021).

This assessment applies to government ownership in, and subsidisation of, listed companies. This situation would magnify the agency problem in circumstances where the government shareholders acted as a “grabbing hand”, expropriating shareholder wealth (Boubakri et al. 2020). Alternatively, government-related shareholders may have enhanced authority to implement controls that protect shareholder wealth by curbing the “grabbing hand” of managerial wealth expropriation. In this scenario, government share ownership would constitute an agency mechanism (Yu and Wang 2020). The purpose of the theory postulated by Boycko et al. (1996) is to explain the circumstances under which privatisation is conducive to achieving both shareholder wealth maximisation and public policy objectives. The Boycko et al. (1996) model may be interpreted as predicting the policy parameters and other settings conducive to government share ownership in (and by extension subsidisation of) listed companies constituting an agency mechanism, rather than an agency cost.

It has been argued that privatisation lacks sound theoretical underpinnings from the viewpoints of the government relinquishing ownership and management of the privatised entity (Kay and Thompson 1986; Thillainathan and Cheong 2019). Boycko et al. (1996) respond to this call. The model predicts that a rationale for privatisation is to simultaneously achieve public policy objectives and shareholder wealth maximisation in the postprivatised entity. In particular, Kay and Thompson argue that privatisation may seldom be desirable—from the viewpoint of management, as rational players—since it reduces their protection from private sector disciplinary forces, such as the markets for managerial labour and corporate control. Boycko et al. (1996) address this gap in the literature. Their model acknowledges that management accrue additional private benefits, as partial shareholders in the postprivatised entity. These benefits would constitute a motivation for management, as economically rational players, to support privatisation. Concurrent achievement of shareholder wealth maximisation and public policy objectives would result in both the improvement of quality of life for society as a whole and the maximisation of the wealth of equity market investors. Boycko et al. (1996) make a contribution via elucidating the circumstances conducive to these goals converging rather than diverging.

Achieving this goal continues to be a longstanding objective of economic policy (Megginson et al. 1994). For example, across the globe, policymakers have embraced New Public Management as an ideology. This refers to the belief that the public sector can be improved via importing private sector mechanisms into the management of public sector entities. New Public Management results in blurred boundaries between the private and public sectors. Examples include corporatisation, out-sourcing, privatisation and other manifestations of laissez-faire economics (Siddiquee 2006; Gomez 2009). In particular, public–private partnerships, also referred to as “political connections”, are used in developing countries to facilitate economic development and nation-building (Sinnadurai 2018). These policymakers need rigorous theory for guidance and to assist in formulating policy settings conducive to achieving their objectives without incurring disenchantment of shareholders seeking to maximise shareholder wealth. Hence, a strength of the Boycko et al. (1996) model is its sound motivation.

Consistent with agency theory, the Boycko et al. (1996) model assumes that parties are economically rational (Jensen and Meckling 1976). Utility functions are also assumed to be linearly related to shareholder wealth. The analyses use microeconomics and differential calculus as tools. They distinguish several scenarios according to whether the employment decision is made by the politician or the corporate manager (assumed to be distinct shareholders in the corporation) and whether it is possible for the politician to pay a subsidy to the company. The principal scenario is one in which the politician pays a subsidy to the company and the employment decision is made by the politician rather than the manager. For this scenario, the theory calculates the optimal subsidy by maximising the product of the utility functions of the manager and politician.

The model predicts that it will be optimal to subsidise when the marginal benefit to the politician exceeds the marginal cost. There are two components to the marginal cost of subsidising: the cost associated with the politician bearing some of the loss of shareholder value, from choosing the high employment state (rather than the low state) and the cost associated with actually making the payment. It is predicted that both of these costs are ameliorated by the fact that the politician is a partial shareholder rather than the sole shareholder.

3. Assumptions Underpinning the Theory

It is assumed that there are two shareholders: the manager of the corporation and a politician. They respectively own α and (1 − α) of the shares. The decision being modelled is the amount that the corporation will spend on employment. It is assumed that this variable, Ě, has two states, H (high) and L (low). The high employment state achieves the public policy objective of full employment. It is further assumed that the politician may pay a subsidy of t to the entity.

The aforementioned assumptions are oversimplifications. Naturally, listed corporations have more than two shareholders. Similarly, many corporate decisions are made jointly. It is not possible to isolate the employment decision from other decisions, such as the amount of capital expenditure, dividend policy and debt policy (Farinha 2003). Furthermore, the amount of employee-related expenditure is continuous rather than a discrete variable with two states. This situation is illustrated via the complexity of the Malaysian accounting standard on employee entitlements and the factors that determine the measurement of many different types of employee expenditure (e.g., annual leave, long-service leave and sick leave). For example, in estimating the provisions for long service leave and long-term disability benefits, corporate management is required to include estimates of the following components in the income statement: the current service cost, the interest cost, past service costs and the effects of any curtailments or settlements (Malaysian Accounting Standards Board 2002).

Both parties are assumed to be self-interested and pursue the goal of maximising the utility of their own wealth. It is important to acknowledge that the politician, in addition to the manager, is economically rational. Shleifer (1998) argues that the desire of politicians to consume perquisites and other private benefits explains why government control has continued to prevail in some scenarios when this is inefficient from the viewpoint of satisfying public policy objectives. Shirley and Walsh (2000) further argue that public sector entities also face an agency problem, albeit different from the agency relationships of equity experienced in private sector corporations. Public sector entities are encumbered by an agency relationship between politicians (as agents) and voters (as principals). This relationship is also characterised by information asymmetry. It follows that the contracts between voters and politicians are also incomplete and that monitoring, by voters, of politicians is imperfect. This indicates that by acting in their own self-interests, politicians will sometimes make public policy decisions that erode the well-being of voters and society at large. Hence, a strength of Boycko et al. (1996) is that they formally model the politician, in addition to the manager, as a rational, self-interested player.

An ideological consequence of the self-interest assumption is that it presents an unflattering view of human nature and legitimises corporate policies that achieve shareholder wealth to the detriment of social and/or environmental well-being (Hines 1992). While this criticism is valid, a theory that starts with the assumption that human beings are self-interested may have greater explanatory power for the policies of corporations and politicians than a theory that assumes people are altruistic and motivated by the public interest (Barbalet 2012).

The model also assumes that the employment decision is made unilaterally by either the manager or the politician. This assumption is an oversimplification. Managers are employed as agents of all shareholders (Jensen and Meckling 1976). Hence, the “manager” in the Boycko et al. (1996) model would have a responsibility to consider the objectives of the politician in making the employment decision. Strong, ethical managers would have regard for the fact that they are not the only shareholder in the company (Hu and Kumar 2004).

The principal parameters of the model are as follows. q is the marginal benefit to the politician of spending an extra dollar on labour. q captures factors such as higher net present value of remuneration from making a decision that pleases the election and hence has a higher chance of being re-elected. m is the marginal cost to the politician of spending an extra dollar on labour. m captures factors such as the politician having less remuneration due to reduced expected fiscal revenue for the government, as a result of the company having a larger allowable deduction for employment expenditure. k is the marginal cost to the politician of actually paying the subsidy. k encompasses items such as political costs, from voter disenchantment at issuing government securities to fund the subsidy. All three of these parameters are assumed to be between 0 and 1 (Boycko et al. 1996).

The model assumes that the two shareholders’ wealth utilities have negative linear relations to the amount spent on employment. In the case of the politician, the model assumes that there are two offsetting components of the politician’s wealth utility: the politician’s marginal benefit from obtaining voter approval for spending more on employment within the corporation (qĚ), and a cost from the bearing, as a partial shareholder, of the reduction in share value from choosing the high employment state. The second component is equal to m(1 − α)Ě, the loss experienced by the politician as a partial shareholder (Boycko et al. 1996). It is noteworthy that the politician’s marginal benefit is assumed not to depend on the portion of his/her shareholding. However, his/her marginal cost is assumed to depend on this variable. This inconsistency may be regarded as a limitation of the model. Furthermore, the utility functions may actually be nonlinear in specification, which is more complicated than the model assumes.

However, most of these are simplifying assumptions rather than critical assumptions. The Boycko et al. (1996) model would still be valid if it were assumed that there may be more than two shareholders or more than two states of the employment decision. Similarly, the model would still be valid if the utility functions were nonlinear but monotonic. These assumptions ensure that users of the theory do not have to grapple excessively with complicated algebra and hence may focus on the economic substance (Prasch 2006)3.

4. Analyses of the Model

The model analyses consider several different scenarios. The first is one in which it is not possible for the manager to pay a bribe to the politician, it is not possible for the politician to subsidise the company and the manager makes the employment decision (Scenario 1). The second situation is identical to the first, except that the politician makes the employment decision (Scenario 2). The next two scenarios cover situations in which management pays a fixed bribe to the politician. Scenario 3 occurs in a world with bribery in which the manager makes the employment decision. Scenario 4 refers to a world that is identical to Scenario 3, except that the politician makes the employment decision. For Scenarios 3 and 4, the model defines the combined wealth utility as the sum of the wealth utilities of the manager and politician. The model proceeds to determine the conditions for which Ě = L maximises this combined utility. The analogy to bribery may be regarded as a weakness of the model. This analogy legitimises bribery and corruption (Hines 1992). The analyses would have standalone validity without reference to bribery.

The analyses then acknowledge that bribery is illegal. The authors explain that payment of a subsidy by the politician to the firm would affect the utility functions in a similar manner to a bribe. The model considers how the wealth utilities of the two parties would change via payment of a fixed subsidy of t. (Naturally, it is also assumed that the manager, rather than the politician, makes the employment decision.) The manager’s wealth utility would increase by αt. (As owner of α of the shares, the manager has a claim over this portion of the subsidy.) Since the subsidy is an “out-of-pocket” expense of the politician, payment reduces her wealth utility by αt. The politician, as owner of (1 − α) of the shares, has a partial claim over the subsidy she pays; hence, her net expenditure is (t − (1 − α)t = αt). The manager’s decline in wealth utility, from paying the subsidy, is assumed to be linearly related to αt via the parameter k.

The analyses proceed to model how the utility functions of the two parties would change if the decision was made to select Ě = H rather than Ě = L. The analyses use economic theory, which postulates that the equilibrium subsidy, t^*, is represented via Equation (1).

t^* = {∆Ĕ * [q + k α − m (1 − α)]}/(2 k α)

(1)

The algebraic steps, progressing from the utility functions of the two investors, to Equation (1), are shown in Sinnadurai et al. (2021, p. 7). A strength of the analysis is its use of rigorous mathematics and microeconomics.

5. Predictions of Underpinning Analytical Economic Model

Following from Equation (1), the principal prediction of the model relates to a world in which bribery is illegal and it is possible for a politician to pay a subsidy of t, and the corporate manager, seeking to maximise her own wealth utility, makes the employment decision. It will be optimal to subsidise when Inequality (2) is satisfied.

q > m (1 − α) + k α

(2)

Inequality (2) captures the situation in which the marginal benefit to the politician, accruing from paying the subsidy, exceeds the marginal cost.

A strength of the predictions is that α and (1 − α) are easy to observe. Empirical measures of the percentage of shares owned by different stakeholders are available in annual reports of listed companies. Malaysian companies are required to provide annual report disclosure of details pertaining to shareholders with the twenty largest holdings of each class of equity (Commissioner of Law Revision, Malaysia 2018). Unfortunately, the other parameters are unobservable and difficult to estimate. The mechanisms affecting these parameters are nebulous and would be determined by country-specific factors. This assessment is illustrated with reference to the Malaysian setting in the following section.

6. Applicability of the Underpinning Analytical Economic Model in the Malaysian Context

6.1. Illustrations of Conflicts of Interest Faced by Malaysian Companies Subject to Government Shareholding and Types of Subsidies Forwarded to These Companies

Politically connected companies in Malaysia, subject to government share ownership, have dual missions to maximise shareholder wealth and implement public policy (Gomez and de Micheaux 2017). These two goals frequently conflict (Boycko et al. 1996). Two examples follow. Tenaga Nasional Berhad is an electric utility. This company is subject to substantial share ownership by Khazanah Nasional Berhad (the investment arm of the Malaysian Ministry of Finance). Tenaga Nasional Berhad has been prevented from raising the price of electricity in the interest of making energy affordable to residents. Furthermore, Tenaga Nasional Berhad is compelled to purchase a specified amount of power from designated generators, irrespective of the amount needed to sustain business (Lai 2012). Petronas Gas Berhad is a company responsible for gas processing and transmission throughout Malaysia. This company is subject to substantial share ownership by Petroliam Nasional Berhad (a fully government-owned oil and gas company), Employees’ Provident Fund Board (a Government-Linked Investment Corporation) and Kumpulan Wang Persaraan (another Government-Linked Investment Corporation). Petronas Gas Berhad has a monopoly on gas processing and transmission across Peninsula Malaysia. The fees charged to users for gas transmission are governed by a Throughput Fee Agreement, binding until the end of 2033 (Dynaquest Sendirian Berhad 2015).

The subsidies paid by Malaysian governments to politically connected entities may be categorised into several types. The first type of subsidy is an award of lucrative capital investment projects related to economic development and nation-building. These projects have long time horizons. Empirical evidence shows that some types of politically connected companies in Malaysia enjoy lower costs of equity capital and higher abnormal earnings growth rates than other companies (Boubakri et al. 2012; Sinnadurai 2016) than other companies. This indicates that the nation-building capital investment projects may be net present value accretive. Two examples follow. The Federal Government has awarded Gamuda Berhad, a listed construction company, the contract as turnkey developer for a tunnelling project in construction of the Klang Valley Rapid Mass Transit railway. Similarly, the Penang State Government has awarded Gamuda Berhad the contract to be the Principal Delivery Partner for road and public transport projects within this state (Dynaquest Sendirian Berhad 2015; Gomez et al. 2018, p. 100).

The second type of subsidy arises from government-directed restructuring of industries. It is quite common in Malaysia for government-related investors to engineer horizontal and vertical integration as a strategy for achieving public policy objectives (Gomez and de Micheaux 2017). This has the effect of reducing product market competition and increasing the market share of at least one incumbent. The threat of new product market entrants would be reduced due to increased market power and dominance of the affected incumbent. Vertical integration reduces the competitive threat from the bargaining power of suppliers and consumers via blurring the distinction between the incumbent company and entities at different levels of the supply chain (Porter 1979; Karuna 2007; Ali et al. 2014).

In some instances, politically connected companies are created by Malaysian governments and ensured monopoly (or oligopoly) power via regulations. This situation typically prevails in industries related to logistics, utilities and communications, regarded as critical for the national well-being (Gomez et al. 2018, p. 26). For example, Telekom Malaysia was incorporated in October, 1984 as a privatisation exercise to take over operations formerly conducted by the Federal Ministry of Energy, Telecommunications and Post. In the postprivatisation era, the telecommunications industry within Malaysia has become more competitive. Notably, in April, 2008, Telekom Malaysia divested its mobile and fixed-line businesses to list Axiata Group Berhad as a separate company (Dynaquest Sendirian Berhad 2008). However, the extent to which this divestiture has eroded Telekom Malaysia’s market power is arguably minimal. Firstly, Axiata Group Berhad provides a different type of service from Telekom Malaysia. Secondly, both companies are subject to substantial ownership by three Government-Linked Investment Companies: Khazanah Nasional Berhad, Employees’ Provident Fund Board and Skim Amanah Saham Bumiputera (a subsidiary of Permodalan Nasional Berhad). The latter entity is also a Government-Linked Investment Corporation. Telekom Malaysia is still the largest provider of telecommunication services in the country. Its share of the fixed-line and broadband businesses exceeds 95% (Dynaquest Sendirian Berhad 2015, 2022). Other politically connected companies with monopoly or near-monopoly product market power include Tenaga Nasional Berhad (electricity provision) (Lai 2012), Pos Malaysia Berhad (provision of postal services) (Gomez et al. p. 26) and Petronas Gas Berhad (gas processing and transmission across Peninsula Malaysia) (Dynaquest Sendirian Berhad 2015).

Vertical integration enhances shareholder wealth via reducing the competitive threats from the bargaining power of suppliers and consumers (Porter 1979). For example, Petroliam Nasional Berhad (Petronas), the national oil and gas provider, is a company wholly owned by the Ministry of Finance (Dynaquest Sendirian Berhad 2022). Petronas is the largest shareholder of companies within the Petronas Group: Petronas Chemicals Berhad, Petronas Gas Berhad and Petronas Dagangan Berhad. Petronas Chemicals Berhad is a subsidiary of Petronas and an integrated chemicals producer, manufacturer and exporter. Petronas Dagangan Berhad was incorporated in 1982 as the marketing arm of Petronas (Gomez et al. 2018, pp. 130–2). Hence, the Petronas group enjoys a high degree of product market power within the gas transmission industry. The competitive threat from the bargaining power of suppliers is reduced by Petronas’ ownership of Petronas Chemicals Berhad. The competitive threat from bargaining power of consumers is reduced via Petronas’ ownership of Petronas Dagangan Berhad (Porter, 1979). These three companies jointly account for eight percent of the market capitalisation of the Malaysian stock exchange (Gomez et al. 2018, p. 130).

In other situations, government-related institutional investors implement public policy objectives by intervening in pre-existing product markets. The investors engineer horizontal integration within the industries, resulting in some of the incumbents experiencing increased market share. This increases shareholder value via the reduction of intraindustry competition, both from other incumbents and potential new entrants (Porter 1979). An example follows. In 2007, guided by the objective of transforming Government-Linked Corporations into “global champions”, Permodalan Nasional Berhad oversaw the merger of three large Malaysian plantation companies: Sime Darby Berhad, Golden Hope Plantations Berhad and Kumpulan Guthrie Berhad. The postmerger company, retaining the name “Sime Darby Berhad”, was one of the largest plantation companies in the world (Gomez et al. 2018, p. 57).

The provision of management services by the Malaysian Government to politically connected companies constitutes another mode of subsidisation. For example, in 2018, Kumpulan Prasarana Rakyat Johor Sendirian Berhad, an investment company wholly owned by the Johor State Government, owned shares in Iskandar Waterfront City Berhad, a property development and construction company involved in Iskandar Malaysia (Dynaquest Sendirian Berhad 2022). The latter plan was announced by the Malaysian Government in 2006. The principal objective is to develop Southern Johor into a hub of economic, social and educational activities. Khazanah Nasional Berhad and the Johor State Government have been actively involved in Iskandar Malaysia via providing planning services and assisting in project implementation for companies awarded development contracts (Lai 2012). These services would facilitate the execution of corporate policies, such as the investment, financing and dividend decision, with a view to maximise shareholder wealth. Hence, the services may be regarded as a type of subsidy (Boycko et al. 1996).

Another mode of subsidisation of politically connected companies is the availability of extra sources of finance (Fraser et al. 2006). For example, as part of its macroeconomic response to the Asian Financial Crisis of 1997–1998, some Government-Linked Companies received “bail-out” packages funded by Khazanah Nasional Berhad. Two of these companies were Bank Bumiputera Malaysia Berhad and Time Engineering Berhad (Lai 2012). These extra financial resources enable politically connected companies to afford larger-scale capital investment projects, resulting in greater shareholder wealth.4

Malaysian capital market researchers face some challenges implementing the Boycko et al. (1996) model. It would be difficult to estimate the dollar value amount of some of these subsidies—the parameter t—in the model. It would also be difficult to quantify the change in shareholder wealth due to the subsidies. The dollar value of financial packages associated with the award of nation-building contracts and the dollar of “bail-out” packages could be readily measured. For example, CIMB Bank Holdings Berhad is a politically connected bank, subject to substantial share ownership by three Government-Linked Investment Companies: Khazanah Nasional Berhad, Employees’ Provident Fund and Kumpulan Wang Persaraan (diperbadankan). This bank experienced financial distress during the Asian Financial Crisis of 1998–1999. It received a bail-out package from Khazanah Nasional Berhad specified at RM1.1 billion (Lai 2012). In this case, the value of t in the Boycko et al. model would be RM1.1billion. It would be more difficult to quantify the dollar value amount of the other types of subsidies (i.e., facilitation of vertical and horizontal integration and provision of management services). Similarly, it would be difficult to isolate the impact on shareholder wealth of these subsidies from other wealth drivers.

6.2. Relative Magnitudes of the Underpinning Analytical Economic Model Parameters in Malaysia

The incremental wealth accruing to the politician, q, from implementing public policy imperatives would be cross-sectionally high in Malaysia. Boycko et al. (1996) identify generic components of q that apply in any jurisdiction. These include garnering more votes (from meeting expectations of the electorate) and less pressure from trade unions. It is very common for Malaysian politicians to serve as independent directors on boards of listed companies during their careers and after retirement from politics (Gomez et al. 2018, pp. 190, 194, 203). The politicians would accrue high political visibility from these activities. They would reduce this visibility and the resulting political costs via actions that demonstrate they are discharging their accountability to the electorate (Lim and McKinnon 1993).

m is the incremental cost to the politician, as a shareholder of the subsidised listed company, from the company selecting a policy that achieves public policy objectives rather than shareholder wealth maximisation (Boycko et al. 1996). m would also be cross-sectionally high in Malaysia. Corporate decisions that achieve public policy objectives, to the detriment of shareholder wealth maximisation, are likely to entail a large opportunity cost. An example follows. UEM Edgenta Berhad, in 2013, was subject to 34% shareholding by Khazanah Nasional Berhad via the subsidiary UEM Group Berhad. UEM Edgenta Berhad was originally awarded a 15-year concession to provide healthcare support services to government hospitals across three states in Malaysia. The period of the concession was subsequently shortened to 10 years, resulting in reduced earnings for UEM Edgenta Berhad (Dynaquest Sendirian Berhad 2013). Hence, shareholders in UEM Edgenta Berhad would have incurred the opportunity cost from management choosing these investment projects related to public policy rather than other investment projects not involving the public sector. Another dimension of m would be UEM Edgenta Berhad’s economic dependence on public health policy. Khazanah Nasional Berhad, to the extent of its share ownership of UEM Edgenta Berhad, would have borne these costs.5

In Malaysia, there are at least two mechanisms affecting k, the cost to the politician of making the subsidy. Both of these mechanisms are indirect. Firstly, the subsidy would result in a wealth transfer to the target members(s) of society from other members of society. The politician, as a member of either or both sections of society, would partially bear this wealth transfer. In cases where the politician belongs to the section of society making the transfer, this component of k would be positive. In cases where the politician belongs to the section of society receiving the wealth transfer, this component of k would be negative. The latter scenario frequently prevails in Malaysia. This would contribute to k being lower than q and m. An example follows. In 2015, Cahaya Mata Sarawak Berhad, an infrastructure conglomerate, secured a five-year contract to construct the Sarawak Museum Campus and Heritage Trail (Dynaquest Sendirian Berhad 2015). Procurement of this contract may be regarded as a subsidy from the Sarawak Economic Development Corporation (an investor representing the State Government of Sarawak and principal shareholder of Cahya Mata Sarawak Berhad) to Cahaya Mata Sarawak Berhad (Boycko et al. 1996). This heritage asset would attract tourists to Sarawak, effecting a wealth transfer from these tourists to the people of Sarawak. The Sarawak Economic Development Corporation would be among the recipients of this wealth transfer.

Political costs are another mechanism affecting k in Malaysia. Subsidisation of listed companies is frequently accompanied by public criticism and objections from the members of society who bear the costs of funding the subsidy. The implementation of Iskandar Malaysia was initially accompanied by scepticism from the investment community about the ability of the players to deliver. The degree of scepticism was exacerbated during the Global Financial Crisis of 2008–2009 (Lai 2012). This fall-out may be regarded as a form of political visibility and may have generated political costs, such as a lower chance of being re-elected (Panchapakesan and McKinnon 1992). The lower expected present value of the remuneration of politicians employed by Khazanah Nasional Berhad may be regarded as a political cost associated with paying the subsidy and hence a component of k in the Boycko et al. (1996) model.

6.3. Synthesis and Implications of Underpinning Analytical Economic for Malaysian Capital Market Research

In synthesis, application of the Boycko et al. (1996) model to Malaysia suggests the following. The model accurately characterised companies with a shareholder political connection.6 The Boycko et al. (1996) parameters of q and m are uniformly high in Malaysia. k is also uniformly high but lower than q and m. Hence, variation in (1 − α)—the level of government shareholding—is likely to be the main determinant of whether Inequality (2) is satisfied in Malaysia. Further application of the Boycko et al. (1996) model demonstrates the following.

MC (politician) = m(1 − α) + kα

where MC (politician) = the marginal cost to the politician from making a corporate investment inconsistent with shareholder wealth maximisation and paying the subsidy.

δ [MC (politician)]/δ (1 − α) = m − k

(3)

Since (m − k) would be positive in Malaysia, an increase in the level of government shareholding would always increase the marginal cost, to the politician, of subsiding listed companies. There would be a threshold level of (1 − α) which, if exceeded, would result in the marginal cost to the politician exceeding the marginal benefit, causing Inequality (2) to no longer be satisfied. For levels of (1 − α) in this range, shareholder wealth would be maximised by choosing Ě = L, which is the choice that does not achieve the public policy objective. However, since management is compelled to choose Ě = H due to the input of the political connection, shareholders suffer wealth deterioration.

In implementing the Boycko et al. (1996) model, Malaysian capital markets researchers should treat government-related institutional investors as “the politician”.7 They should treat the aggregate shareholdings in a listed company by all the government-related shareholders as a proxy for (1 − α), which is the total shareholding by the politician. They should also regard this estimate of (1 − α) as a proxy for the extent to which a listed company is subject to government input into its decision making and the extent to which it is the recipient of government subsidies. It would not be feasible to attempt to quantify the value of subsidies extended since some of the subsidies are indirect. Similarly, it would be infeasible to attempt to quantify q, m and k, which are the incremental costs and benefits incurred by the politician. Some of the components of these parameters are indirect and unobservable. Malaysian researchers should treat the total shareholding (direct and indirect) by a listed company’s Chief Executive Officer (CEO) and his/her family as a proxy for α (the portion of shares in the company held by the manager). The balance of shareholdings (held by neither proxies for the politician nor the manager) should be treated as a third category. Investors in this category include private sector institutional investors, foreign investors and individual magnates. This would not affect the validity of the analyses, even though this bloc of shareholders is not considered in Boycko et al. (1996).

The aforementioned application of Boycko et al. (1996) would be a suitable underpinning for Malaysian researchers to investigate economic consequences (particularly for shareholders) of the mix of two ideologies, used by the government, in formulating economic and social development policies. The ideologies are the development state and laissez-faire economics (Siddiquee 2006; Gomez 2009; Sinnadurai 2018). Similarly, the suggested framework would be a suitable underpinning to investigate economic consequences of paradigm shifts in the mix of these two ideologies. For example, Johnson and Mitton (2003) and Mitchell and Joseph (2010) investigated the impacts of firm performance because of the removal of capital controls by the Malaysian Government after the end of the Asian Financial Crisis. Their studies also examined whether these impacts differed between politically connected and other companies. The research questions, investigated by Johnson and Mitton (2003) and Mitchell and Joseph (2010), could be reinvestigated using the framework suggested in this paper. The suggested framework would also be appropriate for investigation of the economic consequences of many dimensions of corporate policy and corporate governance outputs. These include dividend policy (Benjamin et al. 2016), cost of debt (Bliss and Gul 2012), audit quality (Gul 2006) and capital expenditure policy (Phan et al. 2020).

6.4. The Suggestions in Our Paper Prvide a Framework for Synthesising and Reconciling (Somewhat Contradictory) Prior Malaysian Evidence

The purpose of simultaneously maximising shareholder wealth and achieving public policy objectives is unlikely to be satisfied if the fundamental assumptions underpinning the Boycko et al. (1996) model are not met. For example, Private Financial Initiatives are a type of Public–Private Partnership which, prima facie, would satisfy the assumptions of the Boycko et al. (1996) model. (These are arrangements in which the government provides fundamental services. However, the premises in which these services are provided and/or the facilities are used for construction are provided by private sector entities.) In Malaysia, these initiatives involve shareholding by government-related investors. The contracts to provide the facilities may be regarded as subsidies.8 The Malaysian Private Financial Initiatives have been criticised for allocating a disproportionate share of the projects’ risk to the public sector participants. Empirical evidence indicates that the projects were substantially overvalued in bids mounted by the private sector participants. This is consistent with (but does not prove) planned wealth expropriation from taxpayers and shareholders to the favoured parties, which are affiliated with private sector participants (Thillainathan and Cheong 2019).

Our application of the Boycko et al. (1996) model would explain this finding as follows. The “grabbing hand” (Boubakri et al. 2020; Yu and Wang 2020) prevented the assumptions underpinning the Boycko et al. (1996) model to be satisfied in the context of Private Financial Incentives. An unstated assumption is that the politician and manager, as shareholders in the privatised statutory body, bear risk in proportion to their percentage shareholdings.

Thillainathan (2021) investigated the effectiveness of privatisations within the Malaysian toll-road industry with respect to the extent to which they had successfully achieved the New Economic Policy goal of facilitating the emergence of a class of Malay managers. In the context of the Boycko et al. (1996) model, award of the infrastructure project to a private sector company would constitute payment of the subsidy. Thillainathan argues that prior to the Asian Financial Crisis, the implementation of the New Economic Policy, particularly within this industry, had facilitated the emergence of a class of wealthy, politically connected Malay businessmen. However, the goal of producing a class of Malay business managers had not been satisfied. In the context of the Boycko et al. (1996) model, the latter goal would constitute the public policy objective analogous to choosing the high employment state.

Thillainathan (2021) observes that Malay businessmen were complaining that as a result of this implementation mode, State-Owned Enterprises were “crowding out” the private sector, inhibiting the emergence of a class of Malay managers. Thillainathan also notes that the aforementioned implementation of the New Economic Policy was inequitable; it benefitted a minority of politically connected Malay individuals, rather than the majority. Naturally, fallout from this public dissatisfaction would adversely affect the wealth of the management and directors of Khazanah Nasional Berhad, the relevant Government-Linked Investment Corporation. Khazanah Nasional Berhad would be alleviating this effect via using competitive tendering to allocate nation-building infrastructure projects to private sector bidders. Hence, the relevant politicians would have a higher likelihood of re-election and postretirement appointment to boards of directors of listed companies and hence, higher expected present value of remuneration (Gomez et al. 2018, pp. 190, 194, 203). In applying the Boycko et al. (1996) model, Malaysian researchers would treat this as a component of m, the personal benefit to the politician, of extending the subsidy.

Thillainathan (2021) concludes that the privatisation policy has failed to achieve the aforementioned public policy objective due to contracts being awarded on the basis of patronage rather than merit. Application of the Boycko et al. (1996) would categorise this conclusion as follows. The “grabbing hand” (Boubakri et al. 2020; Yu and Wang 2020) expropriated the subsidy and diverted the politician’s shareholding from the intended statutory body to different entities.

7. Conclusions

This paper presents a critique of the Boycko et al. (1996) analytical model of shareholder wealth impacts resulting from the privatisation of statutory bodies. Whilst the model has sound motivation, its underpinning assumptions are oversimplified. Fortunately, these are mostly simplifying, rather than critical, assumptions. A strength of the analysis is its rigorous use of mathematics and microeconomics to develop the propositions. A weakness of the analysis is that one of the scenarios presents an analogy to bribery. A strength of the model’s predictions is that two of the parameters—the level of shareholding by the manager and politician—would be easy to observe by empirical researchers. A weakness is that the other parameters—the amount of the subsidy and the marginal costs and benefits experienced by the politician—would be difficult to proxy.

The model accurately characteristics companies subject to shareholding by government-related institutional investors in Malaysia. These companies have a dual mission (maximising shareholder wealth and implementing public policy). These objectives are sometimes conflicting. Companies subject to shareholding by government-related institutional investors receive subsidies, directly and indirectly. The application of Boycko et al. (1996) to Malaysia suggests the principal determinant of whether subsidisation enhances or erodes shareholder wealth is the level of shareholding by government-related investors, which is a proxy for the extent to which the government has input into the company’s decision making. Furthermore, our critique suggests that in Malaysia, the association between shareholder wealth and the degree of government shareholding, in politically connected companies, follows an “inverted U” shape.

These suggestions could be used by Malaysian researchers to investigate the economic consequences, for listed corporations, of government policies that differ with respect to the degree of underpinning by laissez-faire economics versus the development states, as ideologies (Gomez 2009; Sinnadurai 2018). This framework could be used to investigate an array of corporate policies. It could also be used as an alternative underpinning for re-examination of research questions, addressed in previous research, related to corporate governance outputs by politically connected companies. The framework proposed in the paper may facilitate policymakers’ assessment of the reasons for failures and successes of economic policies that involve co-operation between the public and private sectors. The manner in which this paper relates the findings of Thillainathan and Cheong (2019) and Thillainathan (2021) to the Boycko et al. (1996) theory provides a role model in this regard.

A limitation of this paper is that it may be difficult for researchers from Malaysia (and elsewhere) to utilise the suggestions. Many of the parameters (including the value of the subsidy and the marginal costs and benefits to the politician) would be difficult to quantify. Furthermore, it would be difficult to identify the theoretical apex of the “inverted U”. This situation is exacerbated by the fact that these parameters (and hence, the turning point) may vary temporally and cross-sectionally.

Naturally, this paper also has implications for capital markets researchers from other emerging markets, across the globe, in which government policies reflect a marriage between the two aforementioned economic ideologies. These researchers should assess the extent to which the assumptions underpinning the Boycko et al. (1996) model characterise their country’s environment. If they regard the assumptions as feasible, they should assess the relative magnitudes of the Boycko et al. (1996) parameters with reference to the institutional features of their countries. Following the role model of this paper, the researchers would then be positioned to mount conjectures regarding the relative importance of these parameters, as determinants of the wealth impact of government share ownership and subsidisation, in their particular countries. A limitation of this paper, for non-Malaysian researchers, is that the conclusions may not fully generalise to their countries due to institutional differences. However, global evidence from 45 countries (excluding Malaysia) suggests that the association between shareholder value and state ownership, in listed companies, follows an “inverted U”. This finding preliminarily suggests that the arguments in this paper may have applicability to countries other than Malaysia.