1

MANAGEMENT ACCOUNTING AND REGIMES OF CONTROL IN ASHANTI GOLD FROM COLONIAL TIMES1

Mathew Tsamenyi University of Birmingham,

Trevor Hopper University of Sussex,

and

Shahzad Uddin University of Essex, UK

Address for correspondence: Mathew Tsamenyi Birmingham Business School University of Birmingham University House Birmingham, B15 2TT UK Tel: +441214158439 Email: m.tsamenyi@bham.ac.uk

1 The authors wish to thank the CIMA Research Foundation for supporting this research

2

MANAGEMENT ACCOUNTING AND REGIMES OF CONTROL IN ASHANTI GOLDFIELDS FROM COLONIAL TIMES

ABSTRACT

This paper examines the evolution of management accounting control systems (MACS) in a large gold mining company in Ghana, namely Ashanti Goldfields Corporation (AGC). The paper is informed by a cultural political economy of management accounting summarised in Hopper et al. (2009), which stems from Burawoy’s (1979, 1985) notions of ‘factory regimes’ applied to MACSs in LDCs in Uddin and Hopper (2001), and the ‘cultural political economy’ in Wickramasinghe and Hopper (2005) and Wickramasinghe et al. (2004). The paper identifies how accounting and control systems in AGC are shaped by different regimes of control. Under colonial despotism, control was personal and coercive, with physical controls that exploited ethnic divisions amongst indigenous miners. During the state and politicised state capitalism regime of AGC, the state intervened into operational controls for political advantage, culminating in deteriorating productivity and finances. After a ‘privatisation’ financial controls remained centralised and focussed on head office reporting and production orientations predominated management. It was not until the financial crisis that major MACS changes occurred, namely decentralised management accounting, tight budgetary control, and cost-oriented management. Viewed from the miners’ perspective, market capitalism has brought little improvement: controls may have changed but remain essentially despotic: politicised market capitalism (or neo-colonialism) has arguably replaced colonialism. Thus, the paper argues, a cultural political economy of management accounting is particularly useful to provide us an understanding of the evolution of controls in AGC. Key Words: Accounting, Management Control Systems, Less Developed Countries, Ghana, Labour Process Theory

3

1. INTRODUCTION

This paper examines the evolution of management accounting control systems (MACS)

in a large gold mining company in Ghana, namely Ashanti Goldfields Corporation

(AGC). Several reasons motivated the study. First, more individual country studies from

less developed countries (LDCs) are necessary to trace patterns and exceptions of MACS

development. Second, the study focuses on Ghana, an early Sub-Sahara African adopter

of IMF and World Bank reforms to privatise state-owned enterprises (SOEs) and

deregulate markets. Ghana, like many LDCs, being dependent upon external capital, must

heed IMF and World Bank economic advice, and the latter often cite Ghana as a success

story.2 Whatever, the reforms changed the institutional environment - accountability and

efficiency are now important parts of organisational vocabularies following capital

shifting from public to foreign-private hands. Third, the findings test and develop a

framework for tracing the evolution of MACSs in poor ex-colonies summarised in

Hopper et al. (2009), which stems from Burawoy’s (1979, 1985)3 notions of ‘factory

regimes’ applied to MACSs in LDCs in Uddin and Hopper (2001), and the ‘cultural

political economy’ in Wickramasinghe and Hopper (2005) and Wickramasinghe et al.

(2004). Its fundamental premise is that MACSs practiced in LDCs are not merely a

consequence of efficiency and market pressures but also other factors like political

instability, trade union struggles and economic uncertainties over time.

We observed that MACSs were not initially instrumental in production control and

decisions. This was partly attributable to: the previous dominance of foreign capital -

which resulted management ‘at a distance’ during colonialism; politicisation when the

mine was nationalised and controlled by the Ghana government after independence;

employee beliefs that they were exploited; lax operational controls; miners’ militancy and

2 Fortune Magazine (2001: s9) observed that: ‘’Ghana, and former President Rawlings in particular, were hailed as examples to be emulated across the continent”. 3 Burawoy (1985) offers a coherent analysis of production, management controls and state politics based on research on mines in Zambia, which shares with Ghana a legacy of British colonialism. He criticises economists and sociologists for neglecting LDCs - ‘the hidden abode of production’ - and, when they do, their over-emphasis on market relations and neglect of labour processes, class struggles, and state politics.

4

resistance; and racial/ethnic divisions between Europeans and Africans, and Southern and

Northern Ghanaians. It was not until the company was ‘privatised’ following a flotation

of government shares and a subsequent collapse of the gold market that it significantly

changed its MACS, with limited success due to the historical legacy of controls bearing

on employee reactions.

The paper proceeds as follows. The next section describes the theoretical model

governing the subsequent analysis. Then it discusses the methods used to collect and

analyse data. Next, the empirical findings are analysed drawing on our theoretical model

and the final section provides conclusions.

2. A CULTURAL POLITICAL ECONOMY OF MANAGEMENT ACCOUNTING

MACS research from West Africa is sparse but that which exists emphasises how socio-

political and economic context shapes practices (e.g. Rahaman and Lawrence, 2001a, b;

Tsamenyi and Mills, 2003; Olowo-Okere and Tomkins, 1998). Moreover, as Uddin and

Tsamenyi (2005) argue, reforms like those in Ghana may not serve the public interest as

expected by the World Bank and other Western donors for the ensuing regulators lack

resources, become politicised and act largely to maintain external legitimacy. These are

useful studies but, apart from Olowo-Okere and Tomkins (1998), they are essentially

historical fragments lacking an overall theoretical analysis - hence our pursuit of a

‘cultural political economy’ to examine control transformations at different stages of

development. This portrays issues as complex, interrelated, and chronological, though

neither deterministic nor inevitable – MACS developments are a product of the interplay

of politics, culture, state actions, bureaucracy, and mode of production (MOP). We

sympathise with allegations that this carries the danger of meta-analyses relying on

overly general and deterministic analytical categories doomed to empirical failure

(Thompson, 1993), for outcomes of struggles are indeterminate. Nevertheless, that they

can fall into patterns, and micro-analyses of controls must iterate with macro-

theorisations of context for theoretical insight, and the framework’s constructs helped

identify factors impinging on the transformation of controls at AGC.

5

[Insert Figure 1 and Table 1 near here]

Our cultural political framework provides a dialectic explanation of MAS transformation

due to changing social, economic, and political factors. It identifies five regimes of

control: colonial despotism, state capitalism, politicised state capitalism, market

capitalism, and politicised market capitalism (see Figure 1). Each regime is rendered

unstable by contradictions and conflicts that fuel political struggles nationally and within

production that lay the basis for new regimes (see Figure 1). State and market capitalism

represent ‘idealised’ regimes of control, being dominant espoused ideologies that

underpin socio-economic reforms commended by mainly external agencies, whereas

colonial despotism, and politicised state and market capitalisms represent actual

outcomes. Each regime is brought about by force, manipulation, persuasion, and

authority in political and economic struggles. Controls veer between coercion and

consent following interactions between key dimensions of each regime - MOP, culture,

ethnicity and race, the state, regulation and the law, political parties, industrial relations,

and international finance. MACS practices in each regime reflect prevailing managerial

strategies of control. Features of each regime are summarised in Table 1, and explained

more fully in Hopper et al. (2009) and, to a limited degree, the subsequent analysis.

Colonial Despotism is despotic for force prevails over consent: it is colonial because one

racial group dominates through political, legal and economic rights denied to others.

Traditional non-capitalist agricultural and domestic MOP within traditional, ethnocentric

cultures tended to be left intact. Rather colonial rulers fostered capitalist enterprises

financed by colonial capital to exploit primary sectors of the economy. Although a small

indigenous capitalist merchant/landowner class often existed prior to colonisation this

tended to remain underfinanced, family-based, and operating outside capital markets.

Politics based on imperialism often employed divide and rule tactics based on ethnicity to

weaken domestic resistance and create indigenous ruling agents. State regulation was

minimal, enabling company states to flourish and control their employees at work and in

their surrounding communities. Trade unions were initially illegal though nascent

6

unionism brought some state regulation of industrial relations. The arbitrary power of the

colonial boss (usually white) to exercise coercive controls based on physical and financial

penalties, and often overt and explicit racism and exploitation of ethnic differences, was a

principle of organizing production (Spearpoint, 1937). Accounting was rudimentary and

insignificant for control, confined largely to financial accounting for stewardship and

tracking remittances to head offices in the imperial power: a MACS for local control was

largely absent: controls were direct and physical – labour was cheap and lacked power in

company states. Not surprisingly, this helped precipitate independence struggles by trade

unions and intellectual elites.

Following independence, many LDCs adopted central state planning to further

industrialisation and modernity, often at the behest of external financiers and

development experts. Such state capitalism resonated with the socialist ideals espoused

by many post-independence leaders, who created large-scale bureaucratically rational

organisations, often state-owned. Accounting was central, as it can provide a rational

calculus to inform planning decisions, and rule-based hierarchical controls to co-ordinate

accountability from enterprises to state central planners, and ultimately the Minister in

Parliament.

However, in practice, such controls withered in significance (though not presence) being

superseded by politicised state capitalism. Multi-party unions within enterprises became

conduits of inter-party struggles and their leaders combined with middle class educated

politicians. Rather than the body politic establishing broad structures of planning, control

and strategic direction and granting discretion to operational managers to execute them,

they used their bureaucratic powers to intervene into operations to secure patronage and

political favours. Commercial rationality exercised by managers and government officials

diminished in the face of political rationalities. This had ramifications for accounting

systems: they did not whither but often became dualistic. Formal bureaucratic systems

remained but were subservient to political interventions. Moreover, given the MACSs’

operational unreality and the dominance of political criteria in decisions, they often

became irrelevant for meaningful operations management and accountability, leading

7

managers to devise local unofficial MACSs based on operational and political realities.

Formal accounting, including MACSs, became loosely or de-coupled from operations,

maintained ceremonially to gain legitimacy from the general populace and external

financial institutions.

Politicised state capitalism brought a cocktail of poor governance, economic grievances,

turbulent industrial relations, and ethnic tensions that fuelled popular protest. Policy

failures resulted in loss making state enterprises and repeated fiscal crises (Uddin and

Hopper, 2001). External donors advocated policies of market capitalism based on

privatisations of state organisations, more open markets, and abandonment of state

central planning to remedy this. Proponents presumed that private ownership would

diminish political interference, increase profitability and tax revenues, promote economic

growth, and improve accountability (World Bank, 1993, 1995a, 1996; Cook and

Kirkpatrick, 1995). The belief was that competition would induce contemporary Western

financial and managerial practices and better governance, and lead inefficient enterprises

to go bankrupt or be taken over by efficient ones (Adam et al., 1992; Hanke, 1986; Rees,

1985; Furubotn and Pejovich, 1972). Politicians were in a quandary: to go down this

avenue would impair the benefits of office; not doing so would diminish external sources

of finance upon which they relied. They shifted, albeit sometimes reluctantly, to market

capitalism policies emphasising market exchanges, private ownership of enterprises,

external labour markets, and free trade and export zones (sometimes accompanied by

weakened trade unions, less industry-wide collective bargaining, and lower employee

protection). The presumption was that this would lessen state power and political

interventions, with the state adopting an economic supply role to attract multinational and

international capital, facilitate stronger capital markets, provide an effective

infrastructure, and enforce greater financial regulation. Democratic, transparent

governance and ‘new public sector management’ within state organisations were

encouraged. Accounting is central here, especially MACSs that enhance market-based

planning, controls and rewards; and effective financial reporting and auditing to lubricate

the functioning of capital markets.

8

Burawoy predicted a gloomier outcome, arguing that relatively powerless workers would

suffer a re-imposition of coercive regimes of production to attract capital from rich

countries (ibid, 1985 p. 262). Uddin and Hopper, (2001, 2003) confirmed this, with respect

to privatisations in Bangladesh. Imposing market capitalism in societies with a small

capitalist class, inclined to familial transactions, patronage, and irregular means, can

facilitate crony capitalism. Capital comes to dominate labour, privatisations transfer assets

to political elites, large income differentials are exacerbated, and private ownership

becomes more concentrated in well-connected often family-based elites and multinational

capital. Moreover, when democratic parties have charismatic leaders from socio-

economic elites, and there is considerable factionalism based on regions, religion, and

ethnicity, regulatory capture and weak regulation often occurs. This, and the need for

politicians to mediate ‘modern’ market-based policies with traditional religious and

cultural expectations of constituents, can perpetuate decisions following political rather

than commercial considerations. Thus, market capitalism may evolve into politicised

market capitalism, with enterprises’ accounts becoming private rather than transparent,

meaningful sources of information for capital markets and other constituencies; budgets

being top-down, pressured and physical rather than means of delegated responsibility;

and weak compliance with external regulation – i.e. coercive despotic control returns.

3. RESEARCH METHODS

The research reported here examines whether the cultural political framework associating

MACS changes with transformations in regimes of control, derived from studies in

Bangladesh and Sri Lanka, was valid in a Ghanaian enterprise, AGC. The study adopted

a case study approach. It focussed on MACSs but also explored industrial relations,

incentive schemes, production and marketing strategies, and organisation structures. The

initial contact with AGC was with its Financial Controller in March 1999. The Obuasi

Mine’s Financial Controller and the Group Management Accountant at head office

helped identify suitable interviewees. The study had four stages. The first, conducted at

the head office in Accra, identified, inter alia, head office influence over the mines and

corporate policies. The Group’s management accountant, human resource manager,

9

assistant company secretary, and the treasury manager were interviewed for

approximately two hours each, in an informal and semi-structured manner.

The second stage was in the Obuasi Mine, the original and oldest mine of the group. It

was the largest division with 80% of the group’s employees and most of its production.

Its importance made it appropriate for the second research phase. Here the investigator

interacted with employees at diverse levels. Formal interviews of approximately two

hours were conducted with the Financial Controller; General Managers for Mining,

Mining Exploration, Human Resources; Underground Management Accountant, Surface

Mining and Processing Management Accountant, Purchasing and Supplies Manager, 3

underground managers, 4 mine captains, Maintenance Manager, and the maintenance

administrator. In addition, informal discussions were held with 5 miners, 2 drivers, and 2

administrative assistants. An underground mine visit with the general manager for mining

and a Canadian expatriate manager responsible for inspection, facilitated observations of

the underground production process and thence an understanding of its operation be

gleaned. The visit to the different underground mine levels lasted five hours and the

investigator could engage miners in conversation and even operate equipment. Several

visits to the processing plants, the maintenance unit, and the surface mine were

undertaken, and some management meetings observed.

Planned visits to three other mines were abandoned after discussions with their financial

controllers: they were small compared to Obuasi, had a similar management structure,

and were considered less significant. Constraints of time and funds limited the field study

to two months, but subsequently the investigator returned to the Obuasi mine in early

2003 for further discussions and follow up interviews with people interviewed earlier,

and then to head office to explore emergent issues.

The third research stage involved interviews of three hours each with officials from the

Bank of Ghana and the Ministry of Finance to discuss the government’s economic

recovery programme, and with World Bank consultants closely involved in Ghana’s

10

economic recovery programme to discuss World Bank and IMF policies and assistance,

accountability and transparency issues, and privatisations in Ghana.

The fourth and final stage of the study involved searching through AGC archives at the

Guildhall Library in London and the library of the University of Birmingham. The

archives contain information about the structure, operations and financial performance of

the company, including its memorandum and articles of association, board and annual

general meetings minute books and confidential inwards and outwards mines

correspondence.

The data was initially analysed by preparing tables listing issues frequently raised in the

interviews and the informal discussions (Potter and Wetherell, 1995; Yin, 2008).

Evidence gathered from documents (such as the archives) and the observations at the

research site were subsequently matched with these themes. The next stage was to

establish links between the themes and our political economy theoretical framework. This

enabled us to iterate how MACS operated under the different regimes.

4. MACS AND REGIMES OF CONTROL

4.1. Colonial Despotism

Ancient Ghana derived power and wealth from gold. The introduction of camel and trade

routes increased the quantity of goods transported across the Sahara, sometimes as far as

Europe. Modern day Ghana constitutes much of the region often known as the Gold

Coast, which has a long history of settlement and wealth. Successive wars brought

vacillating but organised states amongst diverse traditional tribes but an Ashanti

confederacy governed much of the region by the nineteenth century.

The first European traders were the Portuguese in 1471. The Dutch arrived in 1595 and

drove the Portuguese away by 1642. The British arrived shortly after and started trading

gold mined by the indigenous people. The Dutch were unsuccessful in expelling the

11

British and after hostilities a peace treaty was signed both parties in 1667. By 1750 three

European countries traded gold on the Gold Coast – the British, the Danes and the Dutch,

but the latter two left in 1850 and 1872 respectively leaving the British with control. The

British colonised coastal towns with minimal resistance but the Ashanti people fiercely

opposed them in the hinterland. Following several wars, the Ashanti were defeated and

the Ashanti Kingdom came within the British Empire in 1900.

Early Europeans observed women and children panning for gold along riverbanks and

coastal shorelines and perceived indigenous mining techniques as crude. However, initial

European attempts to impose Western mining methods and controls failed (Ayensu,

1997). Three Ghanaian entrepreneurs established the Ashanti mine after buying mining

concessions from Ashanti Chiefs.4 The British defeat of the Ashanti and the exile of their

king paved the way for a London Stock Exchange listing (Ayensu, 1997) but it failed to

attract European capital (Dumett, 1998). They continued labour intensive mining,

employing by 1894 200 miners working four shafts. However, insufficient resources for

large-scale mechanised mining induced the owners to sell their interest to a British firm,

Ashanti Goldfields Corporation (AGC), which immediately monopolised mining in the

area by persuading the chiefs to transfer the concession to them, and the Colonial

Administration to ratify this (Dumett, 1998). The subsequent expansion of operations

owed much to major investments in railway and harbours and the defeat and annexation

of the Ashanti kingdom by the British (Dumett, 1998).

4.1.1. MACS during Colonial Despotism

The Company did not employ formal controls, including MACSs, to control labour in the

mines. Industrial relations were not a problem, as the colonial authorities prohibited trade

unions and ignored harsh treatment of African miners and their deplorable living

conditions (Silver, 1978). The colonial state intervened little, enabling local colonial

company states to emerge. As long as companies produced returns for the colonial state,

4 The Ashantis, a major tribe in Ghana, were its most adventurous warriors and they strongly opposed British colonial rule.

12

the state tolerated despotic company controls that maximised surpluses and regulated

labour processes, as in Zambia (Burawoy, 1985).

The limited documentary evidence available on management controls during colonialism5

indicates that coercion, not consent, prevailed in AGC, "...floggings, beatings, cuffs about

the head and body, not to mention verbal abuse, were administered as a matter of course

for "slow work", tardiness, malingering, or failure to carry out assigned tasks" (Dumett,

1998, p.180). However, miners initially offered little resistance to the harsh capitalist

MOP, partly because they had no organised labour unions, which were illegal (Crisp,

1984). Moreover, the colonial administration offered no regulation to protect miners.

However, during later colonisation, the emergence of independence movements created

the platform for organised resistance by miners.

Although formal MACSs were not used during early colonialism, accounting was not

absent. Dummett (1988, p.509) studied correspondence between the London headquarters

and the mines from 1904-1949, and concluded that:

Considering the value of its properties and the volume of its annual gold output, the AGC did not have a particularly elaborate or specialized administrative system. In its first two decades the corporation’s central administration of Ashanti consisted of a five-man board of directors … supported by a small London office staff. The twin functions of managing director and consulting or chief engineer were combined in one man, who divided his time between the London office and Obuasi. There was little in the way of middle level management. The chief engineer’s orders were carried out at the mines by two assistant managers who in turn controlled the shift bosses and foreman, who supervised three separate underground work crews, totalling 1,596 people by 1911. Neither forward centralised planning nor accounting procedures were very effective during the first ten years. Detailed knowledge of the local geology, complex methods for chemical treatment of ore and experience in how to superintend a large indigenous work force in a West African country were rudimentary. The records reveal a kind of rough-hewn management exercised by the directors, most of whom lacked any experience in mining, and the London office staff, which constantly harped about increasing profits and reducing costs, but relied on local engineers for conduct of month-by-month operations.

No formal cost accounting system or cost accountants are apparent then but

correspondence between AGC mine managers and head office in London reveal that

mining engineers constantly assessed and monitored costs (Dumett, 1988). Moreover, 5 AGC was owned by foreign owners and listed on the London Stock Exchange after independence until 1972. As Burawoy (1985) argued, independence did not necessarily end colonialism and company states.

13

accounting reports enabled head office in London to control remittances and compute

dividends, which had consequences, for example in the early 1900s, shareholders

recommended a major management restructuring following low dividends (AGC, 1897-

1934).

Neither MACSs nor indeed any bureaucratic controls regulated operations, including the

control of labour. In Ghana, like Zambia, village life was tribal and devoted primarily to

subsistence agriculture. From inception, the mine predominantly recruited from rural

areas. Miners maintained their tribal and village culture and moving to a capitalist MOP

created problems. Miners perceived the tight physical targets and working conditions as

inhumane - they had no influence on production targets, methods, or working conditions -

and their indigenous cultural values received scant attention. Most miners’ families

remained in their villages, which compounded problems. As a manager noted, “Most

miners felt very lonely at the mine because that was the first time they had left their

family. In fact, for most this was the first time they had travelled outside the community.

It was a long time before they were able to come and take the family to join them in the

mine’’.

Despite no recognised trade unions, miners started to resist by refusing to work for low

pay or sign six and nine month contracts tying them to one company, or perform tasks

like going underground (Crisp, 1984). Miners increasingly turned to collective action and

foiling management: reduced wages and the introduction of full-day Saturday shifts

brought a strike in 1905. To counter this, the mine’s British management started to

forcefully recruit miners from the predominantly Moslem North (Ayensu, 1997).

Previously miners had come mainly from the Christian South but now, especially

underground jobs, were filled by Northerners partly because they were cheaper and

Southerners filled supervisory positions, which precipitated ethnic conflicts that still

persist. Despite Financial Times reports of coercive recruitment in 1922, the colonial

authority colluded. For example, in 1922 the colonial administration in Accra sent a

message to political officers in the North that stated, “Strict orders have arrived from His

Excellency the Governor that 1,000 labourers are to reach the mines by Christmas at

14

least … Do your utmost to get them down on time” (Crisp, 1984, p.47). However, miners’

resistance continued (Crisp, 1984) and, as in Zambia,6 workers' strikes in Ghanaian

mines, including AGC, ended company state supremacy and forced the state to enter

industrial relations.

Initially, the state used tribal structures (Robotham, 1989). From 1938, employees had to

bring labour disputes with management initially to their tribal headman. Subsequently,

the colonial state encouraged miners to form official trade unions, which became

permissible from 1941. This brought more formalised labour controls7, e.g. the ID system

to track maintenance boys (Crisp, 1984). The trade union became an intermediary

between mining companies and workers, and helped contain resistance within a

rudimentary bureaucratic industrial relations structure, as in Zambia (Burawoy, 1985).

This diminished arbitrary ‘hire and fire’ powers and coercive controls of managers but

restricted trade union officials’ activities and shifted power to higher echelons.

Nevertheless, the presence and the potential power of trade unions deterred managers from

restoring colonial production relations. However, the stability gained from recognising

trade unions soon degenerated and miners constantly confronted management and the

government (Crisp, 1984). Silver attributed “the increasing militancy of rank and file

mineworkers, as expressed both by strikes and by workplace-level struggles; … [to] the

widening gap between these workers and their union; the co-optation of the union by

mining capital and the state, thus creating a capital-state-union alliance against the

workers; and the increasing necessity for the state to intervene, often in its most

repressive form, to put down the insistent revolt of the mineworkers” (ibid, 1978, p. 67).

The Ashanti employees’ union struck in 1945 and 1946 over abolished deferred

payments; wages; working hours; more jobs; better machinery; and lower rents and better

housing. However, the union lost control of strikes in the late 1940s and mid-1950s as,

“workers made the running and the union officials became the reluctant articulators of

their grievances” (Crisp, 1984, p. xvi-xvii). The strikes followed ethnic or occupational

6 Workers' strikes of 1935 and 1940 eroded company state supremacy, forcing the state into industrial relations (Burawoy, 1985). 7 This differed from mining companies in Southern Africa where no unionization was permitted.

15

divisions (Robotham, 1989). The state often intervened by getting the police to crush the

revolt or using tribal headmen to organise unskilled labour against skilled labour. Capital

shortages during World War II affected production resulting in some mine closures

(Crisp, 1984) and some black miners replaced white headmen who had returned home to

enlist. Though black headmen were paid less than their white counterparts, this era saw

the enskilling of black miners and the birth of the ‘black bossman’ (Robotham, 1989).

The all-mines strike action of 1947 stopped management trying to reverse this when

white people returned from the war.

However, management began to exercise new controls. In 1950, South African consultants

began implementing Scientific Management. Robotham (1989, p.43) comments; “an

entire reorganisation of the managing system now took place starting with the

establishment of Planning Department. Time and motion studies soon followed and

standard complements for each job were devised and rigidly adhered to". Systematic

work routines were established, e.g. standardised stoping and workplace conditions;

planned equipment, material usage, and maintenance; and productivity bonuses. AGC

used Taylorism continuously, like Zambian mines (Burawoy, 1985).

It is not clear whether MACS changes accompanied this. Some material8 refers fleetingly

to budgets during the 1930s and later, but searches of company records at the Guildhall

and University of Birmingham libraries could not corroborate this. Moreover, Dumett,

(2000, p. 587) argues that whilst rudimentary management functions evolved

continuously from the 1920s it was not until the 1960s (when Lonrho took over AGC)

that ‘the modern organisational structure of gold mining emerges’. The tentative

suggestion is that crude MACSs for management purposes existed, maintained by

engineers in the field and head office officials.

So, in summary, accounting during colonialism was important, for mine engineers at

AGC and managers in London constantly needed to swap cost calculations. However,

8 Anglo Gold Ashanti’s history at http://www.ashantigold.com/NR/rdonlyres/BDB5B570-C236-4C5B-9F53-1B5ED83F7E0C/0/Ashanti_CompanyHistory.pdf, December 2009.

16

possibly partly due to the inherent uncertainty of mine operations and the simple

organisation structure of the mine, no formal MACS to compute costs or for control was

apparent, as the management structure was simple and slight, and labour was controlled

by despotic means, often exploiting ethnic divisions. Nevertheless, financial accounting

enabled mine managers and head office to communicate, and London officials to exert

some control at a distance. For much of the time AGC operated as a local company state

with the colonial state playing little part. However, growing labour militancy forced the

state to lay the basis of an internal state to govern industrial relations. However, this did

not eradicate despotic controls or racial employment policies, and towards the end of

colonialism miners’ unions colluded with independence movements.9

4.2. State Capitalism and Politicised State Capitalism

The 1940s to 1950s saw an upsurge in the nationalist movement agitating for

independence. The Trade Union Congress (TUC), which the Ghana mineworkers’ union

(GMWU) had joined, was instrumental in this struggle (Crisp, 1984). The British granted

independence in 1957. The fight to overthrow colonialism wrought strong links between

politicians and trade union leaders that continued when they assumed leadership of the

new state. They shaped state capitalism and eventually politicised its ostensibly rational

bureaucratic controls.

Dr. Nkrumah’s government assumed power. It was associated with grandiose socialist

planning, direct political interventions, and nationalisation of foreign owned companies.

The state made massive investments in state owned manufacturing industries to create an

infrastructure, mainly in major cities, for industrialisation. Most Ghanaian state enterprises

operating today are legacies of this policy: the public sector share of total formal sector

employment rose from 55.3% in 1960 to 70.2% in 1965 (Boateng, 1997).

9 Independence struggles brought conflicts in and around the mines but AGC’s employees were often protective of European managers affected.

17

The 1958 Industrial Relations Act legally recognised trade unions and made collective

bargaining binding on employers and employees. Trade unions had to register with the

government to gain political rights, which linked political activities to the official

machinery. The TUC gained a monopoly of trade union activities and membership was

compulsory for civil servants. Trade union activities deemed contrary to the

government’s interest were illegal and banned. Dr. Nkrumah used his powers to intervene

in large state organisations, granting employment to political loyalists and annexing

popular movements (including the TUC). For example, he made the loyal General

President of the GMWU a director of the newly created Ghana State Mining Company in

1961, and in 1963, Chairman and Managing Director. He pursued government interests

by hiring and rewarding employees loyal to Nkrumah and dismissing those not (Silver,

1978).

Regulation of industrial relations, trade union recognition, and the protection of miners

from arbitrary hire and fire changed controls at the point of production at AGC, for

example, floggings, beatings, cuffs about the head and body, not to mention verbal abuse

were eradicated. The arbitrary power of colonial bosses (usually white) to control miners’

life outside work diminished. The internal state (formal collective bargaining) was

strengthened during state capitalism, as Burawoy (1985) noted in Zambian mines, and

Uddin and Hopper (2001) found in Bangladesh. Miners enjoyed a relatively stable

environment during this period and their living standards rose constantly (Silver, 1978).

The next two governments - the military National Liberation Council (NLC) of General

Ankrah and the civilian Progress Party (PP) government of Dr. Busia - held power from

1966 to 1972. Their economic policies favoured private sector participation (Boateng,

1997). By 1966 all Ghana’s gold mines except AGC had been nationalised within the

State Mining Corporation. This did little to revive production, which progressively

declined, partly because of ‘excessive state control’ (Hilson, 2002, p.22). AGC was not

nationalised despite its strategic importance, possibly because of the relationship between

its European owners and Nkrumah’s government (Ayensu, 1997). However, after the

1966 coup, the new military government sought new investors for AGC, after an

18

argument with its European owners led AGC’s Chairman to threaten to flood the mine. In

1968 it was acquired by the controversial10 Lonrho company, headed by ‘Tiny’ Rowlands

- a perceived champion of independence movements in British colonies (Ayensu, 1997).

The government gained 20% of its equity with an option to purchase 20% more at a

nominal price.

4.2.1. MACS during Politicised State Capitalism

As Dumett (1988) notes, Lonhro’s ownership brought contemporary organisation

structures and controls. It progressively adopted management and financial control

procedures recommended by Peat Marwick, Mitchell and Co (UK Competition

Commission report, 1979). AGC, one of 21 management regions within Lonrho, had

several profit centres. A Finance Committee chaired by the finance director reviewed

their performance using quarterly and summary monthly financial reports. Subject to

meeting budget, reporting and control requirements, AGC could determine its internal

controls and exercise considerable operational autonomy.

Accounting controls were integral for the mine’s management. Managers from the pre-

1972 period recalled a centralised organisation structure where head office in London

made most planning and control decisions. A senior manager observed, “The head office

in London designed all controls including budgets. Budget figures were agreed in

London and sent to Ghana for implementation”. Similarly, financial statements were

prepared in London - managers in Ghana had little input. A senior manager commented:

“We were not directly involved in decision-making. Someone sits in London and decides

on what you have to do miles away in Ghana. Emphasis was heavily placed on

production and therefore what concerned them was for us to meet production targets”.

Financial matters were the prerogative of managers in London where the accounting

department resided: they prepared financial reports and set production targets. Physical

10 The British Prime Minister, Ted Heath, described it in the House of Commons as “an unpleasant and unacceptable face of capitalism”. The London Stock Exchange de-listed the company. In May 1973 the Secretary for Trade and Industry appointed inspectors to investigate the affairs of Lonrho. Their report was highly critical but the Director of Public Prosecutions decided further action was not justified.

19

budgets, with output the main consideration, were emphasised, alongside cost reduction

driven by direct controls over labour at the point of production (Robotham, 1989).

However, hostilities between management and miners from the end of Nkrumah’s

administration escalated during the NLC and PP governments. When, in the early 1970s,

government and management rejected miners’ demands11 for the same severance pay as

their counterparts in a nationalised mine, and for the first time, removal of certain

managers, the miners went on the rampage. The company’s attempts to control them

failed: as Silver observed (1978, p.73), "Capital attempted to resist these labour

encroachments, but being repeatedly unable to carry out its labour control function, was

increasingly forced to turn to the state for overt assistance, in the form first of police, and

finally, of the military". The military quelled the strike, killing four miners and injuring

others.

In 1972, Col. Acheampong overthrew the PP government and assumed power, supported

by the GMWU. The new military government refused to comply with the IMF and World

Bank policies and established a quasi-socialist policy (Economist Intelligence Unit,

2002)12 based on central planning. In 1972, the government acquired a 55% interest in

AGC by decree, and made it a Ghanaian company with its head office in Accra. Whilst

the government used its controlling interest to influence decisions and appoint the

deputy-managing director (a Ghanaian mining expert who acted as the government’s

representative on AGC’s board), Lonhro’s expertise and substantial investment enabled it

to appoint the Managing Director. The government ordered AGC’s miners to rejoin the

GMWU and within 2½ years miners’ wages increased by 300% (Silver, 1978). The

government helped indigenous Ghanaians become managers in AGC for the first time:

11 In the early 1970s, workers of AGC temporarily split from the GMWU and formed their own union in the belief that it could negotiate higher wages with AGC which was more profitable than the state mines (Silver, 1978) Moreover, the Busia government’s abolition of the TUC rendered miners’ strikes less effective. 12 The Acheampong regime initially refused to repay loans taken by the PP regime. When the IMF, World Bank, and other aid agencies took retaliatory measures (Petchenkine, 1993), it had to honour its commitment (Payer, 1982, p.47).

20

several young Ghanaians received further education overseas to prepare them to assume

some expatriate positions (Ayensu, 1997).

However, political turbulence and economic problems affected internal controls. Gold

production peaked at 533,000 ounces in 1972 but output declined rapidly over the

following decade to 232,000 ounces. Several sites were exhausted and high taxes, a lack

of foreign exchange, a tightly controlled economy, and an overvalued currency made

buying spare parts or investing in new plant difficult for AGC. The government meddled

in decisions, especially management appointments. The head office move to Accra

brought management control closer to the mine but head office was still divorced from

operations, being 200 kilometres away. Head office centralisation frustrated operations,

for example, delays in getting urgent replacements for broken equipment.

Budgeting was production oriented as the market could absorb any output. Each mine

manager ascertained ore reserves: head office then derived each mine’s production target.

A manager described the system as “crazy”. Another noted, “Then somebody sits in

Accra and says produce one million ounces of gold, at a particular grade and cost

without actually consulting people working at the mine. … If we did not achieve this, the

market is disappointed”. Mine managers had little control over financial decisions and

ambitious production targets: budgets bore little relation to past operations. One

commented that: “When they come out with the figures, you get so frightened because

they are usually high. Having done this it is great but you have ‘killed’ somebody.”

Robotham, (1989, p.67) observed that; “The structure of authority in the mine was

extremely authoritarian, the general principle governing operations being that the

essence of management is control.” and “The mines resorted to ineffectual cost-cutting,

exercises which were often counterproductive. Equipment was run down and the whole

programme of maintenance and tight management control established in the fifties was

allowed to fall into abeyance” (ibid, p.52). Physical production targets set at head office

dominated and rendered cost planning and control marginal within the mine. The over-

reliance on labour-intensive production and cutting labour to save costs proved disastrous

as production fell and targets went unmet.

21

The military government violently destroyed the internal state at AGC. When miners

organised a twenty-five minute sit-down strike in 1972, the Minister for Labour ordered

miners to turnover strike organisers, and threatened to charge future organisers with

economic crimes and trial by military court (Silver, 1978). Miners perceived the

government’s appointment of a conservative opponent of the miners’ strike as General

Secretary of the GMWU with suspicion. Miner’s resistance was subdued but worsening

economic conditions in the mid-1970s intensified conflicts. Strikes occurred in 1977 over

wage increases, payment of bonuses and demands for removal of some managers.

Management granted a substantial wage increase but used covert actions using

informants to garner information about imminent strikes. Consequently, when the

government heard about a planned strike in December 1977 they quickly dispatched

soldiers to the town.

Although this strike never occurred, management realised the difficulty of controlling

miners with the GMWU weakened and the internal state destroyed. Silver (1978)

summarised the dilemma of management and the military government thus; “Capital

needs a union strong enough to be able to control labour, but not so strong as to threaten

capital. The military wants the mineworkers sufficiently happy that they do not join the

professionals and students in opposing the regime, but fears the inflationary

consequences of paying the wages which alone would keep them happy”.

Mounting company debts led management to tighten controls over labour to reduce costs

and militancy (Silver, 1978) but this failed because miners resisted: “68% of the time no

work was done … largely due to the labour arriving late for work and closing early”

(Silver, 1978, p.80). Theft of gold, commonplace for a long time and institutionalised into

miners’ culture, was rampant - estimated at 25% of output. The appointment of ex army

and police personnel to head security and using informants proved unsuccessful as

security personnel collaborated with miners to perpetuate thefts. Management then

employed 12 to 15 year old boys to wash gold naked to prevent them pocketing it but

theft continued in a mill that washed pure gold. Management argued that thefts were

22

immoral but some miners claimed it was a reaction to inhumane working conditions -

working 7 days a week - and low pay (Silver, 1978). Further measures to halt thefts such

as using aggressive supervisors provoked violent responses: miners sabotaged mining

equipment, mixed waste with ore, fiddled supervisor’s production figures, and even used

physical violence. A European supervisor who tried to halt a theft was severely beaten

and lost an eye. Consequently, supervisors often ignored thefts (Silver, 1978).

Officially, management controls were as prior to ‘nationalisation’ but in practice they

were marked by corruption, ethnicity and personalised relations. Supervisors had

problems controlling subordinates with close relationships with higher management

(Robotham, 1989). Miners complained they had to bribe officials such as the Shift Bosses

for promotion, the medical officer for sick leave, and the welfare officer for

accommodation. Robotham (1989, p.71) comments: “the crucial difference was that

during the colonial period, rigid control and depersonalisation (at least between Seniors

who were Europeans and workers who were Africans) was maintained by the entire

system of colonial domination in the mines and in the wider society. With the dismantling

of that system a new approach had not been developed and there was little real

accountability in the ‘Ghana time’ system.

European supervisors found exerting tight labour controls used in other African mines

difficult in Ghanaian mines (including AGC) (Silver, 1978). Miners often resisted and

consequently, “a ‘technical’ management problem having to do with efficiency and

technique, is in reality a struggle over control of the workplace” (Silver 1978, p.82).

Previously management sacked strike leaders instantly but increasing militancy meant

this might provoke violent clashes. Silver (1978, p.82) illustrates the severity of the

problem: “one expatriate manager … reported that he no longer sends the department

vehicles to the Transport Department to be serviced, since the Transport Department has

become a ‘black power shop’, effectively taken over by workers who defy management’s

control, and laugh at their fulminations”. Management attempts to infiltrate the miners

and buy some off failed, partly because it weakened labour unions and thence their

control of members.

23

To summarise, management controls in AGC during state capitalism became less harsh

as managers needed the consent and cooperation of black miners to maintain production,

especially when managers had little influence externally. Initially a strengthened internal

state emerged and secured some control but controls became politicised. The post 1972

military government destroyed the internal state rather than using it to coordinate the

interests of capital and labour - and interventions based on ethnicity became important.

Accounting’s role was marginal in a highly politicised context, especially after

‘nationalisation’, when control resided primarily in the political arena of trade unions,

ethnic divisions, military governments and their respective agents. The domain of the state

grew during Acheampong’s regime, employing 86.4% of those formally employed by

1978. However, the economy entered dire straights: export earnings fell by 52% and

domestic savings and investment by 12% (World Bank Report, 1995; Ho Won, 1993);

cocoa sales - a major export earner – declined and its contribution to GDP fell by 11%;

manufacturing declined by 6%; and mining production declined from 710,000 oz to 285.3

oz. (see Appendix 1). Between 1970 and 1983, inflation rose from 3.7% to 123% and GDP

declined on average 2% annually (World Bank, 1995b quoted in Hilson, 2002). Political

violence, economic problems, and pressure from international aid agencies, especially the

IMF and World Bank, forced the government to institute policies based on market

capitalism (Ray, 1986; Pelow and Chazan, 1986; Gyima-Boadi, 1993).

INSERT TABLE 2 NEAR HERE

4.3. Market Capitalism and the Politicised Market Capitalism

Flight Lieutenant Jerry Rawling’s Provisional National Defence Council (NDC) military

government assumed power in 1982. Rawlings ruled Ghana for nearly 20 years, first as

military leader and then as civilian president. His initial socialist and centralised economic

policies faltered during 1982-83 when Ghana experienced a severe drought and export

revenues fell significantly. The administration turned to international aid agencies for

assistance. The IMF and World Bank responded but any financial aid and technical

24

expertise carried conditions, such as removing price and exchange rate controls, and

government subsidies; and introducing structural adjustment policies based on market

capitalism.

Reforms began in 1983 with the Economic Recovery Plan. The first phase focused on

eliminating price distortions, ensuring a balanced economy, reversing agricultural sector

decline, curbing inflation, enhancing exports, and improving social infrastructure and living

standards of the average Ghanaian. The second phase relaxed and liberalised price controls

and stimulated direct foreign investment. Government investment plans provided tax

incentives for foreigners that invested directly, or bought shares in some state enterprises.

Privatisation was encouraged, especially in key sectors like mining and banking. The third

phase focused on minimising and eliminating government interventions into markets,

encouraging domestic savings and foreign investment, privatisation, and improving the

balance of payments. Significant financial sector decisions included abolishing interest rate

controls, making the banking sector more secure, and reducing government’s interest in

major state banks. The recovery programme had favourable outcomes: by 1992, GDP

growth averaged 5% per annum and gross national savings rose to 6.6% of GDP by 1994.

After the reforms US$5 billion flowed into mining. The IMF and World Bank believed

mining could boost export revenues and AGC got a loan of £159 million in 1985 from a

consortium of banks led by the International Finance Corporation (IFC)13 for

improvement and mechanisation programmes. A manager noted its high interest but as it

“was the only source of funding available at the time, the company had no other choice”.

Foreign exchange earnings from mining increased from US$108 million in 1985 to

US$710 million in 1999.14 Predictions that miners would oppose IMF/World Bank

intervention did not materialise15 (Robotham, 1989, p.59). New mining and investment

13 An affiliate of World Bank that supports viable companies in developing countries. 14 http://www.unctad.org/infocomm/diversification/cape/pdf/barning.pdf 15 Subsequently the government was accused of bowing to international capital. For example, a Canadian Council for International Co-operation report (2003) notes how Ghana’s weak regulatory environment makes mining attractive to foreign investors and enables the government and mining companies to ignore pollution, land degradation and health issues. AGC shifted from underground to cheaper but more environmentally damaging open mining. Resistance by some local people and NGOs has largely been unsuccessful because of government support for mining companies.

25

laws in 1986 let AGC retain 45% of export earnings, which released funds to repay loans

and invest, for instance in sixty scrapers. In 1995, AGC received another IFC loan to

open up new workings at Ashanti. Gold production increased, especially from surface

mining, rising sevenfold from 1980 to 1999, despite a decline from 1980 to 1986 to

287,124 oz. (see Appendix 2). In 1993, output was 1261,424 oz. and in 1998 exceeded 2

million oz., accounting for 38% of Ghana’s exports. AGC’s mines accounted for over

50% of production (Hilson, 2002) (see Appendix 2). Increased gold prices and the

expansion programme improved financial performance. Operating profits increased from

1991 to 1995, and dividends were paid until the late 1990s (see Appendix 3).

In 1994, AGC floated its shares on the London and the newly created Ghana Stock

Exchanges. The government became a minority (20%) shareholder16 by reducing its

shareholding by 25%. Overseas investors became majority shareholders but the

government could still appoint AGC’s Chairman and veto major decisions like potential

mergers.17 In 1996 AGC’s shares were listed on the New York Stock Exchange - its first

African company. In 1998 Lonrho was demerged: AGC went into Lonhro plc,

subsequently renamed Lonmin plc in 1999.18

4.3.1. MACS during Politicised Market Capitalism: a New Despotism?

Government interference in AGC’s activities continued during the initial Economic

Recovery Programme. For example, in the early 1990s the mine’s European General

Manager was deported and replaced by the Deputy Managing Director, a Ghanaian.

16 Since 2001 AGC has sought government relinquishment of its 20% ‘golden shares’. The debate has economic and political overtones. Ellison (2002) argues that: “For a country like Ghana that has a rather weak regulatory regime, the golden share provides a tough oversight on the predatory behavior of international business”, whereas others (including the management of AGC) believe it has negative impacts including stopping a takeover by a foreign investor. 18 In April 2004, the South African mining giant - AngloGold took over AGC for USD1.55b with the government of Ghana’s backing. The merged company, now called AngloGold Ashanti Limited, is listed on the New York, Johannesburg, Ghanaian, London and Australian stock exchanges, and the Paris and Brussels bourses. Its head office was transferred to South Africa though the new group maintains an office in Ghana.

26

Furthermore, the government appointed union leaders and members of workers’ defence

committees, and informants’ alerts of union activities and strikes often brought military

intervention. Politicisation, nepotism, ethnicity and covert economic relationships often

permeated labour controls but vestiges of the internal state remained.

Stealing gold persisted. A manager recalled how, “Some miners went to the extent of

swallowing the gold and getting it out when they go to the toilet. Sometimes we had to

take suspected miners to the hospital. They would then be forced to go the toilet and the

gold retrieved.” Security increased in the late 1980s but introducing metal detectors

precipitated a strike. A miner observed that, “Before this time, no one really cares about

his salary because once you go underground you are sure you can come out with some

gold to sell. There was a ready market for it, so we made so much money. Then,

management introduced these metal detectors. We demonstrated that we do not want

‘pim pim’” (the sound of the detectors). In hindsight, some interviewees accused

employees and management of complacency in the belief that the company’s wealth was

everlasting. A manager recalled how, “Controls were there on paper but were very often

relaxed as money was no problem”. However, industrial relations and pay became more

prominent as controls tightened and diminished miners’ illegal supplementation of

income.

Trade union activities, and the influence of the GMWU declined following the sale of

Government shares in 1994. Workers recognised their dependence upon international

capital and the union became less confrontational. Management tried to facilitate

negotiation, transparency, and cordial relations by co-opting the unions, granting them two

representatives on the board. Nevertheless, management used covert actions such as

infiltrating the union. An industrial relations unit in each mine monitored miners’

behaviour and discipline, and reported imminent strikes. Miners viewed industrial

relations officers with suspicion, seeing them as spies and controllers. The miners’ union

fight for improved conditions after flotation had little success - the final offer of salaries

indexed to the US dollar fell short of their demands. Miners believed that union leaders

had ‘sold them out’, and they voted to remove them. A miner at this meeting recollected

27

that, “It was a hostile environment. Some of the miners threatened that if the existing

leadership is re-elected they will beat them up and burn things in the mine. We all knew

that top management is supporting these leaders and we made sure they were booted out

of office. Everybody thinks they have betrayed us and about 90% voted for a new

leadership.” The mistrust between miners and their leaders, and redundancies in 1998 &

1999 in AGC and mining companies elsewhere weakened miners’ power and collective

bargaining at AGC.

As Appendix 2 reveals, AGC reported increased profits after the flotation. However, due

to the shift from underground to surface mining and poor cost control operating costs rose

from USD241 to USD336 per ounce from 1994 to 1997. Nevertheless, gold prices were

high enough to generate profits and in 1996 AGC’s dividend peaked at 65%. However,

gold prices fell drastically (from US$359 in January 1997 to US$283 by December 1997)

when several European Central Banks (including the Bank of England) reduced their gold

reserves. At the end of 1997, operating costs per ounce exceeded the average gold price,

with obvious negative effects on profits (see Appendix 2). This called for drastic

measures, and the financial orientation and budgeting changes post flotation became

more pronounced. Falling gold prices exposed the lax financial controls and poor

financial decisions: margin calls on a hedging programme ‘backfired’ bringing significant

losses.19 AGC’s net profit declined from 1996, and 1999 to 2001 saw massive losses. The

company’s shares that had traded at USD16 on the New York Stock Exchange fell

sharply and international stock markets suspended trading of them for several months in

late 1999. Large lay-offs of miners helped reduce financial losses (see Appendix 4). AGC

could not fund creditors’ demands for payment, which precipitated attempts to takeover

the company. The government, holding 20% of AGC shares used its veto powers to

19 Following falling gold prices, AGC’s management gambled and hedged a substantial proportion of its gold reserves in anticipation that gold prices would fall. By 1999, 38 % of its gold reserves were hedged at $300 per ounce. However, gold prices suddenly started rising and on October 2 1999, hit a two-year high of $337/oz. The 17 banks with which AGC conducted its derivatives business made margin calls for USD570m. AGC’s hedge fund created a liability which could not be met - the gold sold was buried underground. Consequently AGC breached other loans and conditions, including revolving credit facilities. AGCS’s CEO admitted that: “I am prepared to concede that we were reckless. We took a bet on the price of gold. We thought it would go down and we took a position.” http://www.ghanaweb.com/GhanaHomePage/features/artikel.php?ID=21933

28

scupper a bid of USD665m by Lonmin, which held 32% of the company’s shares.

Moreover, Goldman Sachs, the largest hedge fund creditor of AGC, opposed a quick sell-

off. In February 2000, AGC arranged interim finance of USD100m with Barclays Bank

until a $326m refinancing package was finalised.20 This renewed a US$270 m. revolving

credit facility and retained hedge protection without potential cash calls for three years.

[INSERT FIGURE 2 NEAR HERE]

The 1994 ownership change brought a new organisation structure, partially depicted in

Figure Two. The chief executive, a Ghanaian, was supported by expatriate finance and

operating managers. AGC was organised into seven autonomous operating mines, each

with its own board of directors, with two representatives of senior staff and three from the

union (miners and junior staff), reporting to the chief executive at head office in Accra.

Newly appointed financial controllers reported to a new head office position - Chief

Operating Officer. Mines gained more autonomy: although the executive board still set

output targets, each mine now controlled the means of achieving them.

The economic crises brought further delegation and decentralisation of internal controls.

Management accountants responsible for the main operations (underground mining,

surface mining, and processing) moved from the Finance Department to their respective

operating departments. Previously, as an underground miner commented, “We hardly go

to the finance office. We do not feel comfortable going there since it looks like a place for

the big men.” Workers came to recognise the management accountants as they interacted

with them daily. A miner commented: “The management accountant is closer to us hence

we call him by his first name. We think he is part of us and understands our problem

more than those in the Finance Department. On the other hand, we call the chief finance

officer by his last name because he is far above us. He is on a different level and we do

not necessarily speak the same language. He deals with top management while the

management accountant deals with us”.

20 http://www.businessinafrica.net/leisure/books/175440.htm

29

Relocating the management accountants improved their understanding of the mining

process and the language of miners, which they believed improved budgetary control.

The management accountant for processing commented, “I am not a technical person but

I go to the plants every day just to see what they are doing. My mere presence there tells

them we care about what they are doing.” Accounting and accountants became

significant for miners, engineers and other personnel. Mining engineers had to master

‘costing’, now the dominant business language: even mine captains (the lowest

supervisory level) became accountable through monthly performance reports revealing

the units, total cost, and cost per unit of their operation. The cost accountants now

interpreted performance reports for them and compared their results with those of other

mine captains. A management accountant observed how, “Most managers are now

accepting responsibility for budgeting for their operations and top management is

emphasising cost control”. Previously, budget proposals only required limited

justification – now cost control was a major managerial preoccupation. AGC’s weekly

newsletter (July 27, 1999) exclaimed, ‘’Fighting cost at all cost is a duty of all employees

… Fight it till the last drop’’.

Cost consciousness rose amongst senior ranks. Mine managers started to demand reasons

for variances, which increased the power of the finance department. A manager remarked

that, “The Chief Finance Officer is now very influential in every decision. Even the

Managing Director comes to consult him almost every day.” A management accountant

remarked, “Now at every meeting, they want the accountants to be there. If you are not

there they start panicking. The gold price has given us power and made us important.” A

senior account manager observed how, “In the past the mine captains and managers did

not pay much attention to cost and other accounting reports. Now they run to the

accountants for explanation for figures.” This brought cost cutting initiatives to reduce

bureaucracy, streamline operations, and remove or reform activities perceived as non-

value adding such as procurement. An annual report commented that, “Improvement in

cost at the mine was achieved through closure of the high cost surface operations as well

as effective cost control measures” (2002, p.13).

30

Others were not as enthusiastic. The new financial controls created confusion and

conflicts between staff departments and miners used to physical production targets

dominating operations. Budgetary involvement and effective use of budgets was hindered

by lower level, often indigenous, employees’ lack of accounting knowledge. A

management accountant commented how, “Whites historically dominated the company at

the top and senior management level, with Ghanaians mostly in positions of very junior

employees. Most of these locals had little or no formal education and now occupy some

positions in middle and senior management. Most therefore lack managerial knowledge

and skills, especially in planning and control, and this is reflected strongly in the

preparation of budget proposals”. Insufficient accounting training increased the pressure

on local managers, produced confusion and misunderstanding, and eventually led many

to treat the budgeting process ceremonially.

Moreover, many lower level employee and miners believed that tight controls fell only on

them, gold mines had ample resources, and top managers were plundering gold mines.

This impeded securing realistic financial budgets for, as a manager commented, “Their

budget proposals are ‘wish lists’ based on the premise that the Company can afford

anything they demand”. However, cost reduction strategies brought redundancies

amongst miners and lower level employees (AGC’s employees declined from 12,850 in

1998 to 9,841 by 2001 – see Appendix 4), as in other privatised mines like Tarkwa,

Prestea and Nsuta (Akabza and Darimani, 2001). This aggravated workers’ relationship

with management (especially top management), whose attribution of frequent budget cuts

to market conditions were not universally accepted. A manager remarked how, “The

budget department and top management justify budget cuts [by blaming] low budgeting

skill levels, negative employee attitudes towards budgets, rapidly increasing costs, and

the declining gold prices. Most managers believe they are not given the opportunity to

defend their budget proposals and those in support areas think their activities are not

considered important enough to warrant detailed scrutiny … by top management and the

budget department, hence some cuts are clearly unjustifiable.” However, fears of job

insecurity, a weakened union, and ethnic divisions going back to the colonial regime

muted miners’ resistance. Controls remained highly disciplinary, especially for unskilled

31

Northern black workers, and their differential application across ethnic groups engendered

conflicts amongst workers, which weakened any remaining solidarity.

To summarise, management controls and controls at the point of production underwent

major transformations following the flotation. Politicised controls and interventions

diminished but did not disappear, especially before the financial crisis. Increased foreign

private ownership brought major managerial changes: more delegated financial control

enabled senior management to control at a distance, consistent with expectations in the

market-based structural adjustment programme imposed by international agencies.

Accounting, especially costs, came to dominate the language and preoccupations of

managers, especially European ones trained and habituated in such a discourse. Initially

this had commercial success in a previously rapidly deteriorating sector and the cost

orientation increased following the financial crises engendered by falling gold prices and

financial engineering errors by senior managers.

The new accounting regime’s financial discipline and espoused logic of the necessity of

cost savings and increased productivity justified intensifying miners’ work and lay offs.

Trade unions, weakened by management backed by the state, left workers bereft of ways

to express their resistance collectively. The internal state became relatively inactive with

only a passive co-opted trade union presence remaining. Collective bargaining diminished:

consent came via the external labour market and fear of redundancies. Ethnic divisions

further weakened resistance. Nevertheless, unlike managers and some supervisors,

accounting targets did not directly motivate miners but accounting influenced their

language and actions. Arguably, from the miners’ perspective, market policies exposed

them to a new form of despotism.

5. CONCLUSION

Figure 1, framed the basic research question, namely why and how management controls

in a Ghanaian gold mine transformed over a century. Table 1 framed the social, economic

32

and political factors believed to impact on regimes of control at various stages of

development. The empirical evidence from Ghana was consistent with this framework.

Over time, colonialism and foreign capital transformed indigenous, non-capitalist gold

production into a large-scale industrialised capitalist mode. The European owners’

management controls reflecting Western models and culture were alien to miners

recruited primarily from diverse rural traditional communities. White ‘Europeans’

dominated management, although indigenisation gradually occurred under post-

independence governments. Throughout, ethnic differences, sometimes deliberately

fostered within management controls, created tensions. Resistance caused the colonial

company state to whither and an internal state that governed industrial relations to

emerge.

AGC was never fully nationalised within a programme of state capitalism after

independence, although other gold mines were. Independence brought greater

indigenisation of supervisory and managerial positions, and brought AGC under

Ghanaian company law. In 1972, the government assumed a large stake and exercised

considerable control. As elsewhere, political interventions for party advantage and

patronage within a turbulent and factional political arena overrode legal-rational

accountability and controls, e.g. the military government destroyed the internal state.

The dominance of political considerations eventually precipitated fiscal crises. Ghana’s

economy deteriorated rapidly making it one of the most heavily indebted and lowest per

capita-income countries in the world. The government had to turn to international

financial institutions like the World Bank, IMF and commercial banks. Ghana’s adoption

of a structural adjustment programme based on market capitalism became a condition of

aid. Under IMF and World Bank auspices, Ghana implemented economic reforms,

including some divestment of government shares in AGC - a quasi-privatisation.

Nevertheless, whether market reforms improved enterprise efficiency is debatable -

AGC’s operational performance arguably deteriorated but high gold prices masked this

and enabled profitability to increase. However, when gold prices sharply declined and

33

hedging strategies failed, AGC made acute losses. Although it subsequently became

profitable, no dividends were paid and it was sold to a South African mining company in

2004, with control operating outside Ghana. Throughout, development was at the behest

of foreign capital.

Accounting and control systems varied under different regimes of control (Uddin and

Hopper, 2001; Hopper et al., 2009). Under colonialism, accounting was largely for

stewardship - supplying economic performance reports for head office in London to make

financing, dividend, and planning and control decisions. Minimal formal accounting

occurred at the enterprise, though engineers constantly estimated and discussed costs with

head office. A simple hierarchy of European managers emerged at the mine. Control was

personal, physical and direct, with coercive and physical controls that exploited ethnic

divisions amongst indigenous miners. This largely continued after independence due to

the continuing private ownership of AGC.21 However, an internal state emerged to

control labour relations during late colonialism and beyond; operations management

became more formal and systematised with the application of Scientific Management;

and Lonrho’s ownership brought divisional profit centre management, and improved

financial reporting to head office. When the state assumed more control, they forcefully

diminished the internal state and intervened into operational controls for political

advantage, culminating in deteriorating productivity and finances. After a ‘privatisation’

whereby the state relinquished most of its shares in AGC, financial controls remained

centralised and focussed on head office reporting and production orientations

predominated management. Miners, fearing for their jobs in the face of layoffs were less

militant and management co-opted their unions within a weakened internal state. It was

not until the financial crisis that major MACS changes occurred, namely decentralised

management accounting, tight budgetary control, and cost-oriented management.

21 Other mines were nationalised and made subject to bureaucratic-rational structures of control and accountability within state central planning. These became politicised as noted elsewhere.

34

As Dumett (1985) suggests,22 the evolution of controls at AGC is broadly consistent with

Chandler’s (1975) thesis of corporate development, whereby organisational growth relies

on the adoption of budgets and accounting performance measurements culminating in

divisional structures. And from inception, cost calculations were a primary concern of

managers at AGC. Initially these derived from engineers’ experiential judgements but

gradually more formal, conventional cost estimation systems supplemented these. Hence

the evolution of MACSs at AGC can be seen as a consequence of increasing size and

complexity in the face of market forces. However, reliance on such Western centric,

market-based explanations is partial. It masks ideological issues, promotes

managerialism, legitimates past and present practices as inevitable, and diverts attention

from disenfranchisement of employees and civil society.23 Of course, markets are

important: as the case notes, MACSs were most powerful when gold prices and hence

profits were low. Indeed, from inception, profit for private foreign owners predicated

controls, except for an interlude with a harsh, often venal, and ineffective military

government that partially ‘nationalised’ AGC. Today markets and private ownership are

cornerstones of policies determined by international financial institutions. Being poor,

Ghana relies on foreign capital and exports of a few key commodities with volatile prices

that quickly have socio-economic impacts.

But markets and thence controls did not evolve naturally. Throughout, despotic controls

upon employees, relied on state interventions, which in the last resort violently quelled

resistance. Thus, physical direct controls that exploited indigenous ethnic differences

administered by a hierarchy based on racial divides became possible during colonialism.

Within operations, MACSs were unimportant for control, i.e. they were substitutable.

Poor education and training, their powerlessness, and divisions of labour promoting

deskilling and skilling based on colonial beliefs in racial superiority disenfranchised

labour. Resistance brought more formalised controls, especially an internal state and

22 Dumett made these comments in the context of arguing that African business history is over-concentrated on early colonialism, national developments, and labour relations rather than corporate investigations. Similarly, Chandler explicitly stated that his thesis neglects labour history which he recognised as important to corporate development but beyond his immediate research aims. 23 Civil society and social and environmental issues and reporting are not studied here. This does not mean that they are not important.

35

Taylorism, but these waxed and waned with as the power of labour changed.

Independence and state capitalism brought the promise of reforms. Sometimes miners’

and political leaders’ interests overlapped resulting in a modicum of reform, other times

national governments repressed miners with violence, removed internal states, exercised

rapaciousness, and exploited the industry for political ends regardless of commercial

consequences. Restoration of private ownership and market forces brought budgets and

segmental reporting but these only involved senior, largely expatriate managers. Not until

declining gold prices wrought a crisis in profitability did lower level employees become

directly involved in MACS processes. However, the legacy of ethnic divisions, harsh

controls, distrust of official explanations, and uneven exposure to redundancies frustrated

their adoption. Viewed from the miners’ perspective, market reforms have brought little

improvement: controls may have changed but remain essentially despotic: neo-

colonialism has arguably replaced colonialism. Overall, the paper argues that operation of

the MACS in AGC was shaped by the different regimes of control.

Bibliography

Adam, C., Cavendish, W., and Mistry, P. S. (1992), Adjusting Privatisation – vase studies from LDCs. London: Ian Randle Publishers Ltd.

AGC (Ashanti Goldfields Corporation) (1897-1934). Board and annual general meetings minute books, 1897-1934, (Archives of the Guildhall, Library London: Manuscript Number Ms 14164: Control Number: y9034035)

Akabzaa, T and Darimani, A (2001), “Impact of mining sector investment in Ghana: A study of the Tarkwa mining region”, A Report prepared for SAPRI, (http://www.saprin.org/ghana/research/gha_mining.pdf)

Ayensu, E. S. (1997), The African Legacy of the World’s Most Precious Metal: Ashanti Gold, Marshall Edition Developments ltd, London. Boateng, K. (1997), "Institutional determinants of labour market performance in Ghana",

research paper, March, Centre for Economic Policy Analysis, Accra. Burawoy, M. (1979), Manufacturing Consent, University of Chicago Press Burawoy, M. (1984), The Contours of Production Politics. In Bergquist, C (Ed.) Labor in

the Capitalist World Economy, Beverly Hills, Sage. Burawoy, (1985), The Politics of Production, London: Verso. Chandler, Jr., A. D. (1975), The Visible Hand: The Managerial Revolution in American

Business. Cambridge, MA: Harvard Belknap, xvi + 608 pp. Cook, P. and Kirkpatrick, C. (Eds.) (1995), Privatization Policy and Performance:

International Perspectives. Prentice Hall: Harvester Wheatsheaf. Crisp, J. (1984), The story of African working class: Ghanaian miners struggles, 1870-

36

1980, Zed Books, London. Dumett R. E (1985), “Africa’s Strategic Minerals during the Second World War" Journal

of African History, Vol. 26, pp. 381-408. Dumett R. E (1988), “Sources for Mining Company History in Africa: The History and

Records of the Ashanti Goldfields Corporation (Ghana) Ltd." The Business History Review, Vol. 62 No. 3 pp. 502-515.

Dumett, R. E. (1998), El Dorado in West Africa: The Gold Mining Frontier, African Labor, and Colonial Capitalism in Gold Coast, 1875-1900, Ohio University Press

Dumett, R.E. (2000), Gold: Mining Industry of Ghana, 1800 to present, in Encyclopedia of African History: Volume 1 A–G, K. Shillington (ed.),Mobipocket, pp. 586-7

(The) Economist Intelligence Unit (2002), Country Profile 2002, The Economist Intelligence Unit, London.

Ellison, K (2002), “In the Matter of the Golden Share”, January 25, Profile on Africa (http://profileafrica.com)

Fortune Magazine (2001), Ghana Overview: political ‘model’ tackles economy, December, No.26, pp.s9 Furubotn, E. G., and Pejovich, S. (1972), “Property Rights and Economic Theory: A

Survey of Recent Literature”, Journal of Economic Literature, Vol.10, No.4, pp. 1134-1162.

Gyima-Boadi, E. (1993), The Search for Economic Development and Democracy in Ghana: From Limann to Rawlings, in Ghana Under P.N.D.C Rule, Codesria Book Series.

Hanke, S. H (1986), “The Privatization Option: An Analysis”, Economic Impact, Vol.3, No.55, pp. 14-20. Hilson, G (2002), Harvesting mineral riches: 1000 years of gold mining in Ghana Resources Policy, Vol. 28, No. 1–2, pp. 13–26. Hopper, T., Tsamenyi, M., Uddin, S. and Wickramasinghe, D. (2009) "Management

accounting in less developed countries: what is known and needs knowing." Accounting Auditing and Accountability Journal, Vol. 22, No. 3, pp. 469-514

Ho-Won, J. (1993), The Impact of the World Bank and the IMF on the Political Economy in Ghana, Unpublished A Ph.D. Dissertation, The Ohio State University

Olowo-Okere, E., and Tomkins, C. (1998), “Understanding the evolution of government financial control systems’’, Accounting, Auditing and Accountability Journal, Vol.11, No.3, pp. 309-331.

Payer, C. (1982), World Bank: A Critical Analysis, Monthly Review Press, New York, NY.

Pelow D. and Chazan N. (1986), “Ghana Coping with Uncertainty, Westview Press, London.

Petchenkine, Y. (1993), Ghana In Search of stability, 1957-1992, Praeger, London. Potter, J., Wetherell, M., 1995. Discourse analysis. In: Smith, J.A., Harre, R.,

Langenhove, L. (Eds.), Rethinking Methods in Psychology. Sage, London. Rahaman, A. S. and Lawrence, S. (2001a) "Public sector accounting and financial

management in developing country organisational contexts: a three dimensional view." Accounting Forum, Vol. 25, No. 2, pp. 189-210

37

Rahaman, A. S., & Lawrence, S. (2001b), “A negotiated order perspective on public sector accounting and financial control”, Accounting, Auditing and Accountability Journal, Vol.14, No. 2, pp.147-165.

Ray, D. I. (1986), Politics, Economics and Society, Frances Pinter (Publishers), London Rees, R. (1985), “The Theory of Principal and Agents”, Bulletin of Economic Research,

Vol.37, No.1, pp3-26 Robotham, D. (1989), Militants or proletarians? The economic culture of underground

gold miners in southern Ghana, 1906-1976, Cambridge African Monographs, Cambridge African Studies Centre.

Silver, J. (1978), “Class Struggles in Ghana’s Mining Industry”, Review of African Political Economy, vol.5, no.12, pp.67-86

Spearpoint, F. (1937), “The African native and the Rhodesian copper mines”, Journal of the Royal African Society, Vol.36, No. 1, pp. 1-56.

The Canadian Council for International Cooperation (2003), “Crossroads at Cancun: What Direction for Development? Focus on: Saying No to the New Singapore Issues”, www.ccic.ca

Thompson, W S (1993), Foreword to Ghana In Search of Stability, 1957-1992, Praeger London. Tsamenyi, M., and Mills, J. (2003), “A Survey of Perceived Environmental Uncertainty,

Organizational Culture, Budget Participation and Managerial Performance.” Journal of Transnational Management, Vol. 8, No 1 & 2, pp.17-52.

Uddin, S., and Hopper, T. (2001), “A Bangladesh soap opera: privatization, accounting, and regimes of control in a less developed country”, Accounting, Organisations and Society, Vol.26, pp. 643-672.

Uddin, S., and Hopper, T. (2003) “Accounting for privatisation in Bangladesh: testing world bank claims”, Critical Perspectives on Accounting, Vol.14, No. 7 , pp.739–774.

Uddin, S. and Tsamenyi, M. (2005) "Public sector reforms and the public interest." Accounting, Auditing and Accountability Journal, 18(5): 648-674.

UK Competition Commission Report (1979), Lonrho Limited and Scottish and Universal Investments Limited and House of Fraser Limited: A report on the proposed merger of Lonrho and Scottish and Universal Investments and on the resulting merger situation between Lonrho and House of Fraser. Competition Commission, HC 261, London.

Wickramasinghe, D. and Hopper, T. (2005) "A cultural political economy of management accounting controls: a case a study of textile mill in a traditional Sinhalese village." Critical perspectives on accounting, Vol. 16, pp. 473-503.

Wickramasinghe, D., Hopper, T. and Chandana, R. (2004) "Japanese cost management meets Sri Lankan politics: disappearance and reappearance of bureaucratic management controls in a privatised utility." Accounting Auditing and Accountability Journal, Vol. 17, No.1, pp. 85-120.

World Bank, (1993), Bangladesh Implementing Structural Reform (World Bank, Report No-11569-BD, March 24).

World Bank (1995a), Bangladesh: Privatization and Adjustment (World Bank). World Bank (1995b), Staff Appraisal Report, Republic of Ghana, Mining Sector

Development and Environmental Project. World Bank Report No. 13881-GH, Industry and Energy Operations, West Central Africa Department, Africa Region, World Bank, Africa.

38

World Bank Report (1995), Trends in Developing Economies, The World Bank, Washington, D.C.

World Bank Report (1996), Annual Economic Update: Recent economic developments and medium-term reform agenda, World Bank Country Development I, Washington, D.C, Oxford University Press.

Yin, R. K. (2008) Case study research design and methods, (Applied Social Research Methods) London: Sage Publication, 4th edition.

39

Figure 1

Transition of Regimes of Control in Less Developed Ex-colonial Countries

Pre-colonialism (traditional indigenous mgt a/c?)

Colonial Despotism Govt a/c A/c for foreign cos. Physical coercion

Independence (revolution) State Capitalism Mgt a/c for central planning & control

Politicised State Capitalism Loosely/decoupled mgt a/c Political control

Market Capitalism Structural adjustment programmes Mgt a/c for control & efficiency

Politicised Market Capitalism? Privatisations – varied results

Profit criteria not congruent with development aims New avenues of politic control, e.g. regulation

Reproduced from Hopper et al. (2009)

40

Financial A/c

Figure 2

Partial Organizational Structure

Group Board of Directors

CEO

Group Human Group Operations Group Financial Resource Mgr Director Controller

Mine Board of Directors

Mine Managing Director

GM GM GM GM GM Engineering Processing Mining Human Resource Finance

Senior manager Senior manager Senior manager Senior manager Management North South Central Technical A/c

Underground Underground Mgt Accountant Mgt Accountant Manager -1 Manager –2 Underground Processing/Surface Mine Captains Mine Captains Cost Accountants Cost Accountants

41

Table 1

Regimes of control in ex-colonial LDCs: contextual factors and MASs

Mode of Production

Culture

Ethnicity & race:

State, Regulation & Law

Politics

TU & labour markets

International Finance & capital market

MASs

Colonial Despotism (Actual Regime)

Non-capitalist Agricultural & domestic production. Small capitalist merchant/ & owner class. Colonial capitalist enterprises in primary sector

Mainly traditional, ethnocentric. Closed & stable communities.

Divide & rule tactics based on ethnicity

Company states. Minimal state regulation

Imperialism TUs illegal & weak. Weak labour markets. Nascent unionism & state regulation of industrial relations.

Colonial capital. Otherwise minimal capital. No capital market.

Coercive control based on racial & ethnic differences involving physical violence Accounting for HQ regulations & control

State Capitalism (Ideal Regime)

Industrialisation Capitalist accumulation by SOEs. Fair distribution Continuation of small merchants & traditional agricultural production. Closed economy

Growth of modernistic, urban cultures incorporating rational progress, science & education, meritocracy, individualism & nuclear family.

Nation- alism emphasised not ethnicity

Bureaucratic state central planning. Legal-rational authority. Intervention & welfare oriented. Strong regulation & account-ability to Parliament.

Economic development based on industrialisation.

TUs recognised. Growth of collective bargaining on industry basis

State banking, Central bank regulation Emerging but weak capital market Deficit financing for development.

Bureaucratic rational-legal accounting Enterprise budgeting within national central state planning Creation of formal wage bargaining & internal labour markets

Politicised State Capitalism

(Actual Regime)

State extraction of surplus. Hegemony of political criteria in commercial & production decisions. Power with political elite linked to trade unions. Distribution follows power & patronage

Cultural fragmentation & diversity. More open & less stable sub-cultures. Increased urbanisation alongside strong traditional cultures

Divisions heightened. Ethnicity partly basis of party & political organisation

Legal-rational structures of regulation maintained but captured or ignored by politicians. State patronage, often for party advantage. Weak enforcement

Factional & volatile. Often charismatic/ dynastic leaders of parties rather than ideological. Sometimes non-democratic. Production & state politics often converge

Powerful political party unions. Multi-unionism. Top down leadership. Leaders from political elite. TU membership & power in public enterprises.

Weak politicised, & poorly regulated capital markets, Bank failures Fiscal crises of state lead to aid dependency & reliance on IMF/WB. External financing often for Cold War reasons

Accounting for external legitimacy. Ritual ceremonial practices only MAS irrelevant for internal controls Decisions for day-to-day activities captured by political players

Market Capitalism (Ideal Regime)

Market-based exchange relations & distribution Private ownership of enterprises. New public sector management

Greater individualism & individual economic self-betterment. Consumerism & materialistic choice

Considered irrelevant.

Reduced state power, supply side economic role. Oriented to attract multinational & international capital. Stronger capital market & regulation, especially of

Democratic & transparent government

Strong external labour markets. Weakened Trade Unions. Decline in industry-wide collective bargaining. Lowered employee

Globalised capital Export zones Stronger capital markets Greater financial regulation & enforcement.

Market based controls Contemporary Western best practice Tight production targets Economic performance measurement

42

utilities protection. Lessened political intervention

External reporting for capital markets

Politicised Market Capitalism (Actual Regime)

Domination of capital over labour Wider income differentials Fractions of capital, ownership diffuse/financial institutions/multinationals/local families. Crony capitalism.

Mediation of ‘modern’ market cultures with traditional & political

Often the basis of political & social decisions

Regulatory capture by political-economic elites Weak enforcement Decisions politicised

Democratic parties based on charismatic leaders from socio-economic elites Faction-alism based on regions , religion & ethnicity

Segmented labour markets between core & periphery, Trade unions co-opted into political parties. Lower labour protection a power.

External financial agents especially IMF/WB strong influences on policy. Family/crony capitalism alongside more multinational capital. Politicised regulation & privatisation.

Private accounts, top-down physical budgets Return of coercive control but no physical violence Weak compliance of external regulations – financial & non-financial Toothless trade unions with low bargaining power Internal sub-contracting

Reproduced from Hopper et al. (2009)

43

Appendix 1 Ghana Gold Production Year Gold production (‘000 ounces) 1970 714.4 1971 693.8 1972 710 1973 731.7 1974 709.6 1975 583.1 1976 515.7 1977 531.1 1978 465.7 1979 387.7 1980 342.9 1981 338 1982 337.7 1983 285.3 1984 282.3 1985 299.6 1986 287.1 1987 328.9 1988 373.9 1989 429.5 1990 541.2 1991 847.6 1992 1,004.6 1993 1,261.9 1994 1,438.5 1995 1,715.9 1996 1,583.8 1997 1,7525 1998 2,371.1 1999 2,608.1 2000 2,457.2 2001 2,381.4 2002 2,235.6

(Source: Minerals Commission, Accra, Ghana)

Appendix 2

AGC Gold production (in ounces) 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Total 580,000 650,000 785,000 823,000 937,000 1,029,000 1,169,000 1,500,000 1,560,000 1,740,000Obuasi Mine 580,000 650,000 785,000 823,000 937,000 860,000 858,000 885,000 743,000 640,000Other Mines - - - - - 169,000 311,000 615,000 817,000 1,100,000

44

Appendix 3

Financial performance of AGC 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Turnover (US$ million)

210.6 238.6 275.1 319.2 405.6 458.7 531.3 600.3 582.1 582.2 554.4

Operating profit (US$ millions)

70.7 81.6 96.5 119.9 105.9 81.4 83.6 99.5 96.3 89.1 76.6

Operating profit %

34 34 35 38 26 18 16 17 17 15 14

EPS (US$) 0.69 0.72 1.08 1.33 1.17 0.64 0.54 0.68 0.59 0.27 0.56 Profit attributable to shareholders (US$ millions)

57.9 60.6 90.2 97.8 101.6 60 53.7 40.7 (183.9) (141.1) (62.7)

Dividend payout (%)

19 32 16 21 32 65 41 27 - - -

Current ratio 1.43:1 1.28:1 1.38:1 1.53:1 2.66:1 1.53:1 1.25:1 0.98:1 0.58:1 0.95:1 0.80:1 Gearing ratio (%) 53 36 54 29 34 88 118 115 148 166 104 Gold price per ounce (US$)

362.10 343.86 360 384.12 384.05 387.82 330.98 294.12 278.55 279.10 272.67

Operating cost per ounce (US$)

N/A N/A N/A 241 295 332 336 294 285 284 276

Tax to government (US4 million)

3.9 15.8 0.3 0.5 0.9 0.1 2.4 - 2.7 8.8 6.8

Royalties to government

N/A N/A N/A 9.5 11.5 10.5 10.6 12.6 12.2 13.7 13

Appendix 4

Employees in the mining sector and AGC

Year 1994 1995 1996 1997 1998 1999 2000 2001 2002 AGC employees

9,946 10,499 12,199 12,010 12,850 11,200 10,429 10,189 9941

Mining sector employees

21,268 22,515 21,017 20,336 21,252 17,848 16,524 16,344 14,299

AGC % of mining sector employees

47% 47% 58% 59% 60% 63% 63% 62% 70%