Political [Personnel] Economy – a Political Economy Perspective to Explain Different Forms of Human Resource Management Strategies

A political economy approach to explaining the existence of different human resource strategies is developed in this article – in short: a political personnel economy. The starting point is a critical analysis of the abstinence of politics and power and the resulting explanation deficiencies of traditional microeconomic approaches and of the transactions cost theory. The Marxian labour process theories also discussed in this article, while certainly “political”, primarily exhibit problems related to the theory of action as a foundation. A micro-analytical theory of action-based version of a political economy approach is therefore outlined and applied to the explanation of different human resource management strategies.


Objective and approach
What are the reasons behind the different human resource management strategieshuman resource management strategies understood as patterns of action for "dealing with" the human resources or the work force of an organization?Why do we on one hand have human resource management strategies, where the majority of employees have a long-term employment contract with the organization (up to a "life-long" employment guarantee), and where good working and wage conditions prevail, while on the other hand there are certain strategies that are characterized primarily by very short-term employment contracts and are frequently associated with poor working conditions and relatively low wages?Economic approaches to an explanation trace the origin of the different human resource management strategies back to the actors'primarily the employers' -aspirations for efficiency.Often neglected here are the differences in power between the employer and employee, but also other political aspects of the exchange between "capital" and "labour".In this article, I suggest a Political Economy Perspective as an explanatory approach (see also Nienhüser 2002).For this approach, I resort mainly to considerations of power and exchange theory that exhibit parallels to microeconomic approaches, in particular from the realm of institutional economics, but I also pull from thoughts from the neo-Marxian labour process concepts.
My starting point is the fact that human resource management has four problems to solve; problems that overlap to some extent with respect to content, but that are most often discussed separately in the literature on the subject: the production costs problem, the transaction costs problem, the transformation problem and the problem of appropriation and safeguarding the power structure: The first task of human resource management must be toward a solution as to how to procure labour-power at a "favourable" price.Secondly, the costs of setting up and adapting the operational labour-power, but mainly the costs of the exchange itself, must be kept at a low level.This overlaps partially with the third task, the transformation problem: "Purchased" labour-power must continuously be transformed into labour actually done.And the fourth task of human resource management is to help facilitate for the capitalists the smoothest possible appropriation of the values produced by the employees.Another factor of this equation is to reproduce the exchange system, i.e. ultimately the power structure and authority system: Functional ideas on the part of the actors regarding a "fair and just" exchange (thus the subject of equitable wages is an important personnel management topic) must be stabilized via the macro-exchange system, that is: via the economic and social system.Economic theories influenced by neo-classics consider only the first problem, while newer institutional economic approaches also take the transaction costs and transformation problem into account to some extent.As a rule, the economic theories completely ignore the aspects of appropriation, power and authority.Theories which have their origin in Marxism concentrate mainly on the third and fourth problems.They therefore, on one hand, consider the effects of one dimension of human resource management strategies -the forms of control -on the safeguarding of the capitalists' dominance over the workers, and on the other hand the effects of the productive forces (primarily technology) at given relations of production (forms of ownership) on the human resource management strategies or forms of con-trol.Traditional economic approaches exhibit a certain curtailed view of efficiency and ignorance of social conditions, in particular power relations.These types of models therefore have a limited power of explanation.I maintain that a fitting explanation of the differences in human resource management strategies requires that all three problems mentioned be included, and that the mechanisms behind these problems and their respective solutions be brought out.Such mechanisms include utility, exchange and power.I will accordingly resort to those theories which make statements about such mechanisms.My objective is on one hand to develop an alternative to microeconomic explanations, because they disregard power structures and processes.On the other hand, my approach is intended to represent an alternative to Marxian approaches, since they lack a theory of action-based micro-foundation.The overall issue is the development of a micro-analytical political economy approach based on theories of exchange and power.
I will elaborate the argument that economic approaches, including later versions, are non-political and cannot appropriately explain human resource management strategies.On the other hand, Marxian or Marxian-inspired political economy concepts are indeed political, but they also exhibit a series of difficulties.I therefore propose a micro-analytical, theory of action-based version of a political economy approach, the essential elements and outlines of which I shall sketch out and apply to the explanation of the emergence of various human resource management strategies.

Non-political theories -clarification of a concept and a problem
I maintain in the information below that theories and explanations that do not take politics and power into account are non-political and -far more importantly -cannot appropriately explain human resource management strategies.What do we mean by non-political?Perspectives based on theories of politics and power generally include the following common features: Politics is seen as "power in action" (Jennings 1994: 6, Pfeffer 1981): Actors use power, on one hand, to enforce their interests (even against those of others), and they attempt, on the other hand, to maintain or broaden their positions of power (see also Weber 1980: 822).Power can be based on the control of resources, but it can also lie in exerting influence over the perceptions, values and knowledge of the other actor.Power influences the exchange, i.e. in our case the exchange between capital and labour.Three characteristics emerge from these considerations, which can be used to assess theories or models of explanation regarding human resource management strategies as political or non-political.A theory or model of explanation is non-political if at least one of the following three conditions is met (similar to Bowles/Gintis 1990;1993): (1) A non-political explanation does not include the entire range of resources controlled by the actor or the resources relevant for an exchange.Significant sources of power and significant causes behind the outcome of action are thus ignored.This type of view is therefore not only non-political, but also leads to inappropriate explanations: Whoever controls resources that are needed by the other person -even if they are not the subject matter of the current act of exchange -has power.Power is used to influence the exchange in one's own best interests.The resource provisions available to actors must therefore be included in a comprehensive and systematic way.
(2) An explanation is thus non-political when it perceives the exchange of labour for wages to be the equivalent of a sales contract.Ideally, sales contracts come into being on a voluntary basis, without any pressure.All claims of both parties can be fully defined in the contract.After entering into the agreement, it is therefore not necessary to assert these claims using political means (power) in the exchange process itself.But the dynamics are actually quite different in the transaction of "labour for wages": Both parties generally have incomplete information at their disposal.The person who has demand for labour/the employer does not know the future behaviour of the employees; he/she cannot completely control whether the requested performance of work has actually been accomplished, nor does he/she know whether or not it might make more sense to impose different conditions for the work tasks or the requested volume of work.The person offering labour/the employee cannot be sure that the wages will actually be paid; he/she does not have a concrete idea of how the employer will act or how his/her own alternatives for action will change, etc.Because of this, the claims of the parties within the exchange process itself must continually be reformulated and enforced within the exchange.In other words: The exchange of labour for wages is not completed at the time the employment contract is entered into.It is actually a "contested exchange" (Bowles/Gintis 1990;1993), because the interests of the parties are antagonistic and both parties are attempting to also use (micro-) political means to realize their interests.
(3) An explanation is also non-political if it is based on fixed preferences and patterns of knowledge and perception that can no longer be changed within the exchange process.But in reality, these factors do change within the exchange procedure.They also change due to the fact that the influence exerted on preferences and patterns of knowledge and perception of the other contractual party is used as a means of power, in order to better enforce one's own interests (see also the bargaining theory of Walton/McKersie 1965).
In the following step, I will show that economic theories (and models of explanation based on them) must be viewed as non-political with respect to these criteria.An analytical distinction must be made between theories that make fundamental statements about basic laws, or at least make empirical generalisations, and models (not necessarily of a formal nature) or explanations, which I construe as the application of theories to concrete empirical facts (see also Nienhüser 1996).Below, I will concentrate primarily on models, on applications of theories in the form of explanations.

Are economic theories and models non-political and therefore less
capable of providing an explanation?

The microeconomic standard or textbook modelcomplete rationality and complete ignorance of the political
The microeconomic standard or textbook model is rarely applied in pure form in explanations of empirical human resource management phenomena.The assumptions of most modern microeconomic theories and models are also clearly removed from this starting model (see Sadowski 2002, for example).Nonetheless, or for this very reason, it is useful to call the standard model to mind again in order to present alternative explanations that are richer in contrast when presented against this background.
The standard model assumes a situation where there are two actors: The first one wants to sell "labour" (therefore called the supplier of labour or "the employee"), while the other (the person demanding labour or "the employer") is interested in the work performance and has the means to purchase this performance.Both actors, according to theoretical assumptions of behaviour, want to maximise their benefit and have at hand all of the information required for reaching a decision.Assuming a sufficiently large number of alternative suppliers and demanders of labour, the model predicts that the parties will reach an agreement on an equilibrium wage, which corresponds to the marginal revenue product of the performance on the demand side, and to the marginal rate of substitution between income and leisure time on the part of the supplier (see also Weise and others, for example, 1991: 293 ff).
It makes little sense to speak of human resource management strategies in this model.Human resource-related actions by firms are to a great extent meaningless in this model.The two sole parameters, which could be used to describe dealings with workers, are the wages, the performance, and the relationship formed from these two factors.
The design of the exchange underlying the standard model is non-political.The only resources that are controlled by the actors are the volume of work and the purchasing power.Even the resource of purchasing power is effective to a limited extent only.The actors are not in a position to alter the prices.They are price takers, who can only react to the market price by means of the demanded and/or supplied volume.Power within the meaning of influencing action plays a role only in a very limited sense.A baker could possibly be motivated to bake special bread rolls for a customer who has an especially high level of purchasing power, while a customer might be enticed into making a purchase by a very appetising display of rolls.The employment relation (the exchange of wages for performance of work) is perceived in the same manner.Just as the purchaser of bread rolls can seek out a baker and then disassociate himself from this baker again, so can an employer seek out an employee and subsequently disassociate himself from this employee (this applies in the same manner to the employee).All actors have the same power, namely no power at all (Bowles/Gintis 1990; see also Swedberg 1987).Power is irrelevant to an explanation of action and results.I further maintain that this signifies a conceptional narrowing and ignorance of reality: Employers do not merely have purchasing power, but also additional resources; they control (at a minimum) the means of production.As a general rule, employees could also have control over additional resources (e.g. through inheritance or assets acquired from winning the lottery).To what extent both sides have which resources at their disposal is an empirical question, which cannot be decided in advance theoretically as in the standard model.
The standard model is also non-political from a second standpoint, in that it implies that both parties are able to specify their aspirations in employment contracts.Thus, that which is easy to observe empirically is masked.For example, it is often very difficult to determine which actual work performance an employee has to render or precisely which actual performance has been precisely rendered and what exact compensation is to be paid for the performance in question.The fact that what results is a contested exchange, in which the assertion of claims cannot actually be completely exogenised, is masked by the theoretical assumption of complete information.
The standard model also fulfils the third condition of being non-political: The preferences are assumed to be constant.(Knowledge and patterns of perception also play no role whatsoever).The behaviour of the actors is geared toward realising those alternatives that best satisfy the preferences.The possibility that preferences can also change within the labour process is left out of the equation.From empirical job satisfaction research, however, (e.g.Bruggemann 1976), as well as within the scope of most of the social psychological theories, it is clear that preferences and thus aspirations adapt to the possibilities of their realisation (aspiration level adaptation).The fact that a player could still be able to influence the preferences of the other or systematically profit from situational-based changes in preference is also not conceivable within the model.

Institutional economics/transaction cost theorylimited rationality and limited ignorance of the political
Microeconomics has not stood still the case of the standard model sketched here.How is the political element presented in these further developments and how will the phenomenon of human resources management strategy be conceptualised in them?I would like to take a closer look mainly at the transaction cost theory and treat it as representative of modern economic approaches.
The transaction cost theory (see Williamson 1984;1985, in particular) renounces the acceptance of perfect rationality and acts instead on the assumption of bounded rationality.People do not have available complete information about present and future conditions, nor can they, as a rule, process this information.This results in uncertainty regarding alternative courses of action, their consequences and their probability of occurrence.The theory also presumes a tendency toward opportunistic behaviour.
The assumption is that people tend to utilise situations up to the point of deception.In connection with the assumption of uncertainty this means that people within exchange situations -during transactions -try to safeguard themselves against possible consequences of environmentally and behaviourally caused uncertainties.They invent institutional arrangements for this purpose, precautions for dealing with the exchange.The theory thus explains the existence of institutional arrangements regarding their function for reducing uncertainty and protecting against the opportunism of the other party of the exchange.The problems of uncertainty vary depending on the characteristics of the exchange object, the exchange parties and the exchange relation, and depending on these characteristics other arrangements are institutionalised.
Operational human resource management strategies, within the scope of the transaction cost theory, are institutional arrangements that serve to reduce the uncertainties of the exchange of work performance for remuneration or the negative consequences arising from such uncertainties (see below : Williamson 1984; see Williamson/Wachter/Harris 1975).In contrast to the standard model, it is not only the price for labour or the distress of work in the sense of the effort extended in rendering the actual work performance that is significant.Within the theory, these entail for one thing the so-called production costs.But in addition -and this is the innovation of the theory -transaction costs are also taken into account.Transaction costs are to be viewed as the "economic counterpart" of friction in mechanical systems (Williamson 1985).Included here are the costs for procuring information about the employee (and for the employee about the employer), the costs of possible ex-post-negotiations regarding the performance to be rendered and the appropriate wage-performance ratio or other conflicts between the employer and employee (Williamson 1984: 91), but also feelings of injustice among the parties of the exchange (Ouchi 1980: 130).Institutions -which do not come free of charge themselves -serve to reduce transaction costs.Since the actors are striving toward minimising costs, they prefer the alternative course of action for which the sum of production costs and transaction costs (and of costs for the institutional arrangement) are lowest.
Let us now examine the exchange of labour for wages and the calculation of the employer a bit more closely.The employer attempts first and foremost to secure income from firm-specific qualifications -i.e.investments in human capital -and secondly, to ensure control of productivity, so that the worker have no opportunity to draw wages without rendering the respective performance in return.Figure 1 depicts which institutional arrangements are efficient in conjunction with certain contextual conditions (Williamson 1984: 91).These arrangements form the key element of more complex human resource management strategies in each case.The expressions of the two variables "Level of difficulty of metering the productivity of an employee" and "Degree of firm-specific qualifications" therefore determine the situations in which the four forms of exchange regulation are efficient and therefore empirically expected.
(1) Internal spot market.In the case of easy metering of productivity and non-specific qualifications, no particular form of control or exchange regarding the employment relationship is required.The employees can be terminated without incurring costs and can themselves give notice of termination.An example of jobs for which -according to Williamson (1984: 91) -this form would be suitable, are for example those which are exercised by migrant workers in (California) agriculture.
(2) Primitive team.In the case of difficult metering and non-specific qualifications, a "primitive team" is transaction-cost efficient.Membership in such teams, according to Williamson (1984: 91) can be changed without incurring costs, since no investments in specific human capital are lost.Control of the work performance cannot be effected by way of remuneration on an individual basis, since the performance of the individual cannot be metered and/or is not attributable.The unloading of cargo can be used as an example of this type of situation.This would therefore be a type of job for which group remuneration would be suitable.
(3) Obligational market.In a situation of easily metered productivity but highly firm-specific qualifications both parties, employer and employee, have an interest in a long-term exchange relation.Williamson advises the employer (1984: 91), for example, to establish a non-transferable pension scheme for these employees in order to avoid the migration of human capital.(4) Relational teams or clans.In the case of difficult metering of productivity and highly specific qualifications, Williamson feels (1984: 92) that so-called "relational teams" are suitable."Relational teams" are characterised by comprehensive socialisation of employees, with the objective of replacing the difficult outside control or external control with self-control or inside control (Ouchi 1980).Added to this are long-term employment guarantees to prevent the migration of specific human capital, etc. Examples of jobs for which such an exchange system would be efficient could include consulting activities by bank employees when granting loans or in the real estate business.Specific skills are acquired when these activities are practised, and the employees build up personal relationships, which help them better handle the tasks at hand.The theory thus attempts to provide answers to the question as to how human resource management strategies can be developed on the basis of theoretical concepts and why there are differences in the strategies: Human resource management strategies are institutional arrangements for regulating the transactions relevant to personnel management; differences can be seen depending on the need for specific human capital and the extent of control problems.
The transaction cost theory -at least in the version and method of application propagated by Williamson -is "political" in a very limited sense (see also Dow 1987;1993;Marginson 1993, below).First of all: One essential assumption of the theory is that the claims of the parties cannot be completely specified or that there are differences in the possibilities for specifying them.The claims must therefore be enforced within the work process, within the exchange itself, and cannot be completely developed from without.In this item, the transaction cost theory is therefore political.Secondly, it also includes more resources than merely the purchasing power on the demand side and the volume of labour on the supply side.Transaction cost theorists point out that the employees can better enforce their interests when they have highly specific, firmrelated knowledge that has been acquired over a long period of training.The construct of power is nevertheless not systematically (or systematically not?) included in the theory.This non-consideration is surprising insofar as transaction cost theorists continually address the means of the employers, which one would declare as control of resources (and thus as a basis of power).Ouchi (1980), in particular, therefore emphasises that the employers use the socialization of the employees as a specific means to align the values and preferences of the members of the organisation toward the goals of the firm.It is evidently assumed that the employers have the means to exert specific influence on the composition of the group composition, on rituals, propaganda, etc.This means that the employers not only have control over far more resources than mere purchasing power.It also means that they can influence the preferences of the employees, even though changes in preference are not modelled.Williamson has been arguing for a number of years against the claim that power or differences in power must be included systematically as explanatory factors.He states that power is a "dif-fuse" concept (see also Williamson 1995, for example: 235).Assumptions about other resources for influencing action and also about the changeability of preferences within the work process are therefore not systematically incorporated into the core of the theory.I therefore refer to the transaction cost theory, despite its potential, as not completely non-political, but non-political to a limited extent.
One could at this point argue that it is not a disadvantage for a theory to be nonpolitical, because including assumptions about power and politics does not provide any additional input toward an explanation.It is correct that a simpler model should take preference over a more complex model, provided it has at least the same explanatory power.In this respect, I do not present the argument that assumptions about power and politics per se must be included.They must be taken into account only if they provide additional input toward an explanation.I should like to briefly show that this is precisely what happens, however, that differences in power make a difference in the appearance of different human resource management strategies.Williamson (1984) explains the existence of certain organisational forms of the exchange and the utilisation of work as the result of the freely-made decision on the part of the employee as well, although power plays an important role here.Williamson thus assumes that it is not only the owners of the orange plantations in California who have an interest in entering into very short-term "day labourer" employment contracts with the orange pickers.He assumes that the workers also prefer this type of day labourer employment contract, because they can then market their non-specific human capital at any time under more profitable conditions: "Neither workers nor firms have an efficiency interest in maintaining the association" (Williamson 1985: 245).It is assumed on the part of both actors that they wanted this arrangement.This is a false conclusion, however.Williamson infers the interests (inadmissibly) from the results of the action and does not take into account the initial distribution of resources.This type of "explanation" does not provide any valid information.With the help of a thought experiment, this assertion can be illustrated as follows: Let's assume that the day labourer would like to have an employment relation with the plantation owner that is similar to that of a civil servant so that only the employee can dissolve the employment relation at any time, while the employer remains obligated to engage the employee as long as the latter desires.This would be efficient for the employee.Why does this type of agreement not come into being, instead of the one that is in the interest of the employer?Because the plantation owner plainly and simply owns the plantation.The employee has virtually no alternatives for the utilisation of his labour, while the plantation owner can sell both the plantation and his labour; what's more, the employer can generally select from among several employees, while the employee has far fewer choices to make between alternative employers.In short: The employer has a power advantage in the exchange relationship with the employee, the existence of which must be included to provide a sensible explanation based on reality.
Let us summarise: The transaction cost theory could easily be "politicised"; considerations of power can be easily integrated into the core model of the exchange as well as for an explanation of human resource management strategies.And in no way do all transaction cost theorists have such a strong aversion against power as Williamson (see also North 1990, for example).

Marxian-oriented approaches to the political economy of human resource management -an alternative?
There is a large number of academic studies on political economy which deal with human resource strategies in a wider sense.These studies stem primarily from authors with a (neo) Marxian background.The works of Braverman (1974) and those belonging to the so-called labour process debate strongly refer to topics which are relevant in the discussion of HRM strategies.All these theories take account of either economic or political structures and processes and are thus of relevance to the political economy perspective which I favour and intend to expand on in this study.

Harry Braverman and his labour process theory
Braverman starts from the assumption that (capitalist) labour organisations and the type of employment relations provide capitalists with a solution to their problems.He posits two fundamental problems facing employers, which they attempt to resolve with the aid of human resource management strategies.Firstly, Braverman assumes that employers do not only have to solve the problem of surplus value production (for details on the theory of surplus value see Marx 1980: 165, 532), but also have to deal with the problem of surplus value appropriation.Surplus value is produced by the workers.The capitalist wants to be able to skim off and use this surplus value for his or her own purposes in its entirety and without transaction costs, i.e. without the opposition of the workers.Therefore, the capitalist uses certain instruments to avoid opposition.In the same way as Williamson, Braverman secondly assumes that by buying the product of "labour-power" employers merely have labour capacity (the potential) at their disposal which still has to be transformed into concrete work performance.This is also known as a transformation problem.
In order to solve both problems (the uncertainty of surplus value appropriation and the necessity to transform labour-power into work performance) employers use control instruments (cf.similar Neuberger 1995: 227).The concept of control plays a central role and should be understood in the wider sense, encompassing not only supervision but also management and regulation.In the end, control stands for the management of labour, of the organisation of labour, of the employment and/or contractual relations and of the workers.Braverman concentrates primarily on the organisation of labour.In his opinion, the Taylorist organisation of labour and the resulting dequalification of the workers fulfil the control function better than other forms of labour organisation.This also means that the forms of organisation and types of employment relations that are economically efficient in the narrower sense (i.e., under whose regime the highest surplus value is produced) are used less often than those where the appropriated surplus value is highest (see also Marglin 1974).
Above all, Braverman explores the reasons for the emergence of the Taylorist organisation of labour.His explanation and reconstruction of the historical development can be described as follows: The development of capitalism is characterised by the expansion of markets, heightened competition, increased employment of technology, and firm growth.While this development on the one hand brought about a more efficient production of surplus value, the organisation of labour and the form of control, on the other hand, came under pressure to adapt.An increase in growth of firms re-sulted in an escalation of problems relating to transformation and appropriation, which could no longer be solved through personal supervision by individual capitalists.According to Braverman, the capitalists therefore installed managers to act as their agents and to take over control of the work and/or the workers on behalf of the owners.Moreover, the form of the employment relations changed.In the past, the workforce had been predominantly employed on a "self-employed" basis.In other words, the workers were entrepreneurs in their own right who only offered their own labour-power in the market.These "self-employed" workers were then integrated into the organisation by means of employment contracts, i.e. they were subjugated to stricter hierarchical control while less emphasis was placed on the price and/or market systems as a means of control.At the same time, the Taylorisation and mechanisation of the labour process led to a dequalification of the workers.This resulted in an overall loss of power on the part of the worker and a gain in power on the part of the employer.Generally speaking, Taylorisation fostered the dominance of capital over the labour process.This image of capitalist "progress" has been criticised and developed further by many authors, in particular those involved in the labour process debate.

Further development in the labour process debate
Many critics (see for example Friedman 1977; Edwards 1979; and, for an outline Thompson 1983;Hildebrandt / Seltz 1987;Neuberger 1995: 212ff.)object primarily to Braverman's labour process theory on the grounds that his notion of the control of workers by the management is all too simplistic.Braverman's portrayal, they say, gave the impression that Taylorist control over the workers had been perfectly achieved.However, this is hardly the case.There has always been resistance (e.g.acts of sabotage or strikes) on the part of the workers.In addition, Braverman disregards the fact of partial co-operation between workers and management (Burawoy 1979).Finally, empirical changes in the qualification of the workers can also no longer be reconciled with Braverman's thesis of dequalification (cf.for example Kern/Schumann 1984).
These points of criticism were taken up during the development of more diverse typologies of control, which take into account both the resistance and the possibly negative effects excessive "Taylorist" control may have for the management.These forms of control can be interpreted as human resource management strategies or examples for the treatment of workers.There is often reference in literature to a distinction originating from Edwards (1979) between simple, technical, and bureaucratic control.(1) Simple control is based on personal supervision and sanctioning of the workers and exertion of influence by their immediate superiors.(2) Technical control is exercised through a technical arrangement of the labour process, for instance in the form of assembly-line work.(3) In the area of bureaucratic control, behaviour is channelled by means of (mostly written) rules and regulations.In the course of the labour process debate, it became clear that it would make sense to add a fourth, more indirect form of control to these three forms of direct control.The result is (4) the strategy of "responsible autonomy" (Friedman 1977).The management transfers responsibility to the workers for a specific area, enlarges their task autonomy, increases loyalty to the organisation through social security benefits etc. Normative involvement is established in the place of pressure or material incentives only, so that the workers themselves want to do what they are supposed to do.Marxist theory rates this strategy as a particularly "sophisticated" solution to the ever-increasing problem of transformation and appropriation.
The efficacy of a specific control strategy is dependent on the extent to which control is necessary.This last inevitably varies according to both technical production conditions and the political organisation of the workers themselves.Moreover, control strategies by no means solve the aforementioned control problems completely, on the contrary, they often result in new conflicts.

Assessment
What conclusions can be drawn about the construct of human resource management strategy from the Marxian perspective and the labour process debate?Essentially, it is a question of the regulation of the exchange between capital and labour.In contrast to transaction cost theory, however, it is now a question of a contested exchange coupled with the exercise of power and the safeguarding of dominance.The most significant influence over the type of exchange regulation is technology and the current state of development of the forces of production.As a result, the need for control and the respective type of control are subject to change.With the increasing complexity of market adaptation as well as transformation and control problems, human resource management strategies become differentiated through the application of "more sophisticated" instruments which use less direct control, are more individualised, and rely more on the self-motivation of the workers.
This perspective can certainly be described as political.Firstly, it considers the fact of an extensive amount of resources and the distribution of control over the same, with the labour process perspective primarily focused on the control of the means of production.Secondly, the assumption that the employment contract is undetermined is characteristic of this approach.Thirdly, it is not only assumed that labour (the labour process) moulds people and influences their thoughts and preferences but also that in a society in which the means of production are controlled by capitalists ideologies develop which fulfil a power-stabilising function for this group.
Nevertheless, from my point of view there are some fundamental theoretical and methodological problems (for more details see Nienhüser 2002).These should be remedied by my version of a political economy explanation.(1) The hypotheses of rationality are too far-reaching (similar as in Türk 1995:31ff.).It is assumed that the management correctly recognises which human resource management practices or strategies are suitable for developing and maintaining power and therefore adopts these strategies.However, it would make more sense to start with limited rationality, as is the case with transaction cost theory (Simon 1979;Taylor 1975;March/Olsen 1976;Nutt 1984).It is doubtable whether the management is always in a position to assess the situation correctly and to select the expedient means.As a solution I propose (as also set out by transaction cost theory) working with diminished assumptions of "rationality".This, however, introduces the problem that substantially more consideration must be given to the specific, variable preferences and knowledge of the actors in order to understand their decisions and actions.(2) The political economy perspective of Marxian origin furthermore tempts us to lose sight of short-term changes and the differences that exist within capitalism, for example, among one firm and another.The merits of theories in the Marxist tradition consist in the accurate assumption that human resource management strategies are not produced in a social vacuum, but are rather closely related to fundamental, long-term changes in production and property relations.This view, however, raises a problem of its own: the Marxist theorists dealt with here predominantly fall back on general "laws" which belong to a social level.Such "laws" are mainly based on economic driving forces, which in turn originate from the aspiration for utilisation and increase of capital against the background of advancing technical development and competition.As a result, any advance in the productive forces (technology in a very wide sense) is hampered at a certain point of its development, so to speak, by relations of production and in particular by property relations.Therefore, relations of production "must" continue to change until capitalism as restrictive social system is finally overcome.Such assumptions on mechanisms of social change relate to highly aggregated units, whole societies, and their development.It is therefore difficult to explain the differences between the firms in any one society at a particular point in time.It is also doubtful whether such hypotheses meet the requirements of a central postulate of methodological individualism, i.e. that social phenomena can be explained by means of assumptions about individuals' actions and assumptions about structural contexts (cf. Elster 1985).My solution -as previously suggested for other problems -is to choose a more distinct micro-orientation, to exploit assumptions about individual behaviour, and to take into account perceptions, the interests of the actors, the perceived alternatives for action and the power relations (the controlled resources and their alternatives).This means considering social structure.Social structure shapes interests, limit and facilitate actions, and determines a framework for power relations in a given organisation.

Exchange theory-based fundamental principles of a micro-analytical political economy of human resources management
The core of (political) exchange theory connects various assumptions about expected utility, intentionally rational behaviour, , exchange and power (see above all Emerson 1962;Coleman 1991;Esser 1993;Pfeffer/Salancik 2003, see also Nienhüser 1996;1998).The repeated exchange between "capital" and "labour" within a context that is defined by regulating institutions influencing the exchange and by the (unequal) distribution of power is the focal point of a political economy model based on this core theory and aimed at explaining human resource management strategies.Exchange reproduces and shapes institutions and power relations.Therefore, the first major element of a political economy theory is the construct of an exchange system.The elements of a social system such as this are actors and resources (or events), over which they exercise control and in which they take an active interest (Coleman 1991: 34).The fact that actors do not normally fully control the resources or events in which they are interested results in the generation of transactions between actors and the development of a social system resulting from the process of exchange (Coleman 1991: 35f.).The actors of a social system (in this case initially minimal, consisting of two actors) make exchanges until social equilibrium is achieved.Equilibrium signifies neither equal distribution nor pareto optimality.Even in the case of a voluntary ex-change it is possible that this pareto-optimal condition either cannot be achieved at all, or only in multiple form (Coleman 1991:48).Coleman (1991: 169) points out specifically that the exchange must be seen against the background of initial distribution of the control over resources amongst the actors.This initial distribution is essential in terms of power and the probability that the actor in question will be able to successfully implement their preferred action strategy (when necessary against the will of other actors).And on what does the actor's power depend?A particularly clear answer to this question that is of use for an action-based political economy theory is given by Emerson in his power-dependence theory (1962; 1972a; b; 1976; see also Cook 1987;Nienhüser 2003).The core components of the theory are summarised in the following sentences.The first one takes the form of a definition: (1) The higher the dependency of actor A on actor B, the less power A is able to exercise over B. The second and third one take the form of hypothesis: (2) The dependency of actor A on actor B is the larger, (i) the more value A places on specific objects controlled by B, and (ii) the less opportunity A has to gain hold of these objects outside the scope of the A-B relationship.The same thing applies in B's case.(3) The more A is dependent on B, the more likely B will be in a position to enforce his or her demands, thus overcoming A's potential resistance to these (Emerson 1962: 32).Emerson (1962: 33) draws attention to the parallels between the above and economic theories: the importance (here, Emerson speaks of "motivational investment") of an object for B is to be seen as demand, the availability of the resource outside the A-B relation as a supply of resources.However, Emerson's approach is applicable to economic and social problems on a far wider scale.Furthermore, his power theory differs from the economic models outlined above as a result of his broadly applicable concept of resources.It is also not based on the idea of a fully informed and rationally acting homo oeconomicus.On the contrary, it simply assumes a kind of minimal rationality in human action, namely that people attempt to realise their subjective interests.(Emerson 1972a;b).
Emerson is concerned, amongst other things, with answering the question of how the (unequal) distribution of power influences human actions and social structures.From the assumption that people attempt to put into effect their interests follows that actors wish to reduce their dependency (thus increasing their power).To this end, they perform balancing operations, as Emerson (1962) calls it.These operations describe the theorybased possibilities of how to bring about changes in power distribution, be it through the actions of the exchanging parties, or be it by changes occurring outside of the exchange and independently of the exchanging parties.There are four possibilities of balancing operations: (1) Actor B reduces his striving for the resources controlled by actor A. This can be achieved through B's subjective emotional devaluation of the resources in question ("this job and the associated income are not so important to me after all"), or if he or she chooses to suppress his or her desires and relinquishes the resources and their benefits.(2) B uncovers alternative sources leading to the same resources, thus becoming less dependent on A. It may also be the case that alternative resources arising for B independently of his or her endeavours, for example, new jobs for employees for which B is eligible or other sources of income (e.g. the much-hoped for lottery winnings).(3) B's dependency on A is further reduced if the resources at B's disposal become more important to A, or if B succeeds in making this appear to A to be the case.(4) B's dependency on A is reduced if alternative resource sources leading to the same resources, which A aims to gain from B, run dry, or if B succeeds in causing this to happen.These operations not only describe possible courses of action for the various actors in question (and it is here that Emerson's term "operation" is to some extent misleading), but also events occurring independently of their actions, which are capable of changing the balance of power."Operations" 1 and 2 in particular lead to the possibility that the resources could become renewable or increase.It could, for example, become necessary for an organisation to use new forms of technology, thus making employees experienced in the field of the aforementioned technological advances more powerful and less dependent.
The central variables of the core model which can be used for a political economy analysis are therefore clear: the interests (based on values and expectations) of the (individual or collective) actors, the actors' control over certain events and/or resources, the amount of alternative resource sources available to the actors at any given time as well as the exchange relationships (transactions).Power is a function of interests, control and alternative resource sources; power imbalances lead to balancing operations, which result in new balances of power (and do not necessarily result in equal distributions or even distributions that are felt to be just).Social structure is altered by balancing operations.Nevertheless, not all operations are possible in every situation, for example the creation of alternative resource sources can be limited.

Outline application of theory to explain human resource management strategies
I have outlined above the major cornerstones of a core model involving the economic and "political" power-related statements.However, this model is highly abstract and refers to any exchange whatsoever.The political personnel-economy perspective now uses the theoretical ideas described in the core model, filling them with content so to speak.It regards the relationship between employer and employee as a concept of contested exchange where disputes over the realisation of interests -in which power is used -lead to the creation of institutions that retroactively regulate the exchange.Therefore, the disassociation from a "pure" economic perspective is also clear.
(1) The political personnel-economy perspective assumes that the power of the economic exchange is not exogenous but rather lies within the exchange itself, i.e. it is also used in the labour process and in employment relations to enforce demands made on workers.The "labour contract" is undetermined (Braverman 1974;Bowles/Gintis 1990).This zone of uncertainty offers room for the assertion of interests with the help of power.Power, or rather more dominance in the form of institutionally developed power, is based on the control of resources and creates the potential to change the alternative costs of the other party to one's own advantage (cf.Weise et al. 1991).(2) Seen from the perspective of political personnel-economy, power does not constitute itself solely through purchasing power; power that is firmly established in society, manifested in its institutions and organisations, is also essential for the functioning and therefore the understanding of an economy.Seen from this perspective, control over financial resources and means of production, over the labour process, and over state legislation, essentially is the basis of power.(3) A third assumption of political personnel-economics is that preferences and patterns of perception are shaped by society and develop above all within the employment relation and the labour process.Personnel or HRM strategies serve to ensure the targets of efficiency and appropriation of the product of labour.It is therefore likely that it is not always the most efficient human resource management strategies in the narrow economic sense that arise, particularly when less efficient forms of the appropriation target are more helpful.
I will now try to apply these theoretical elements, defined in greater detail in a further step, to the explanation of human resource management strategies.The facts to be explained will then be conceptualised with the help of the theory.
In this light, human resources management strategies are forms of utilising and reproducing labour-power, transformation of labour-power into work performance and the appropriation-efficient and dominance-ensuring regulations of the exchange relationship between capital and labour.Depending on the strategy used, these human resource management problems that have to be solved with the help of the strategies differ in terms of scope and form.For ease of understanding, I distinguish between just two human resource management strategies, which I refer to as the short-term external and the long-term internal strategies (abbreviated as SE-HRMS and LI-HRMS) (similar as in Lutz 1987).These strategies represent end points of a continuum and roughly correspond with the forms of regulation referred to by Williamson as "spot market" and "relational team".For purposes of clear depiction, the intermediary forms have been excluded.The characteristics of the human resource management strategies and their causes are portrayed in Figure 2.
Short-term external human resource management strategy (SE-HRMS).This strategy is characterised by short-term employment relationships, i.e. by short-term contracts and a higher proportion of periphery workers (e.g.subcontracted workers, agency and temporary workers), by recruitment on the external market, by a working situation that is unfavourable for the employee, and by low wages.What are the common economic and political causes for such a strategy?An influential factor of major importance is the complexity of the work tasks.It could be said that this is one aspect of the productive forces.If the level of complexity is low, the demands that the employer places on human capital level, human capital specificity and social capital (in the form of co-operation among the workers, and between the workers and their superiors) are relatively low.All that is required is unspecific, low qualifications; firm-specific knowledge and experience play a small part in simple work tasks, as do loyalty and the transfer of values and knowledge.On the part of the employees, this means that they control relatively few resources; the employer can easily replace employees with others or with machines for example.This results in a high level of power inferiority on the part of the employees.Employers have no interest in a long-term employment relationship; their costs are lowest when they fulfil their demand for labour as necessary via the external market.With simple work tasks, the problem of having to transform labour-power into work performance can be solved by direct supervision and performance-orientated pay.The appropriation and dominance problem is solved by the existence of a, however variable, "reserve army" (any employee can be substituted, even though the availability of labour varies, e.g. for demographic reasons).Moreover, the dequalification of labour that could be observed in the context of Taylorisation still serves as a means of exerting dominance in the area of simple work tasks and relatively low demands on knowledge.Low proportion of periphery workers Antagonistic conflict of interests: strategy corresponds with the interests of the employer not the employee; the employees for their part do not have enough power to assert their interests Antagonistic co-operation: strategy corresponds with the interests of the employer and partially with the interests of the employees

Political-economic causes (typical combination of conditions)
Low complexity of work tasks; low demands on human capital level, human capital specificity, self-control and social capital (employees control few resources, employer has extensive substitution options); extreme power inferiority on the part of the employees.
High complexity of the labour tasks, high demands on human capital level, human capital specificity, self-control and social capital (employees control many resources, employer has a large interest in the resources and has few substitution options); lower level of power inferiority on the part of the employees.(Employer's production and transaction costs are lower with a market orientated solution; employer's transformation problem is solved via external control; appropriation and dominance problems are solved by a "reserve army" and dequalification).
(Employer's production and transaction costs are lower with an organisation-orientated solution; employer's transformation problem is solved via self-control; appropriation and dominance problems are solved via socialisation, involvement through "responsible autonomy" (combination of material and immaterial incentives)).

Low trade union density of employees
High trade union density of employees High density of membership in employers' associations Low density of membership in employers' associations Coalition between the employer and the "state" Coalition between employees and the "state" No alternative resource sources for employees Alternative resource sources for employees An antagonistic conflict of interests exists with this short-term externally orientated human resource management strategy.The SE-HRMS corresponds with the interests of the employers but not with those of the employees.Due to the fact that the resources controlled by the employees have little value for employers, employees do not have enough power to assert their interests.This type of human resource management strategy is strengthened further by a low trade union density among employees.When the trade union density is higher and the workers therefore pool their labour supply, thus reducing the substitution options of the employer, they are more likely to succeed in asserting their interests with a view to job security and favourable working and payment conditions.This goes in the direction of an LI-HRMS.A higher density of membership in employers' associations also supports the SE-HRMS as employees have less possibilities of switching to employers with potentially better offers.The power of the employers is all the more stronger and the power of the employees all the more weaker when the employers succeed in asserting their interests on the level of the state labour legislation and social welfare legislation, for example with regards to employee co-determination, dismissal possibilities or state social welfare benefits.For example, if state social welfare benefits were to be reduced, from the political economy perspective outlined here, this would mean that the employees would have fewer resource sources available, their dependence on the employers would increase and the employers would be able to better assert their interests.Coalition building therefore provides a source of power that, regardless of scarcity, human capital specificity, and other direct labour process-related resources, has an influence on power relations and therefore on the employers' choice between a long-term internal and a short-term external HRM strategy.It is not only the membership of employees in trade unions that counts as coalition building, but also "alliances" with political parties going beyond the workers' unions.In phases of a social democratic government and corresponding legislation, we can therefore expect the interests of the employee to become more prominent and a long-term internal human resource management strategy to be established.

Long-term internal human resource management strategy (LI-HRMS
).This strategy is characterised by long-term employment relationships, i.e. by longer terms of contract up to "life long" employment guarantees.A high proportion of core workers is dominant here, mainly recruited via the internal labour market through internal promotion.The remuneration and working conditions in this case are rather more favourable for the workforce.The complexity of the work tasks also represents an important influence factor with this strategy.With a higher level of complexity, the demands that the employer places on human capital level, human capital specificity and social capital are relatively high.Scarce qualifications that can only be reproduced at increased costs are required.Contrary to short-term external strategies, firm-specific know-how, loyalty and the transfer of values and knowledge at shop floor level play a significant role in the successful completion of work tasks.Consequently, a short-term adaptation of the existing workforce via the external market is problematic.To preserve a high level of specific qualifications and social capital, a long-term orientated use and maintenance of human resources is required.In this way, the employees control a relatively large amount of resources (in comparison to the SE-HRMS situation).It is only with increased costs that the employer can substitute one employee by another or by machines.The power inferiority on the part of the employees is therefore less marked.The employers best solve their costs problem in this situation by internally covering their needs for human assets and performance.In the case of complex tasks, the transformation problem needs to be solved in a different way to the simple task scenario.Direct monitoring and performance-orientated remuneration are of little use here.For example, due to the nature of jobs remuneration cannot simply be linked with quantity performance.The higher need for internal control is maintained by long-term exchange relationships in the form of relatively secure jobs and the promise of promotion.This contributes to solving the appropriation and dominance problem in the same way as the socialisation and involvement of workers through strategy elements which correspond to the "responsible autonomy" form of control.We are dealing with a combination of normative, calculative-instrumental involvement here (Etzioni 1961), whereas with the SE-HRMS the calculative involvement and "the force of circumstances" are functional.In the case of the LI-HRMS, the basic conflict between capital and labour is not at all resolved; however, the conflicts of interests are objectively and subjectively much less pronounced than with short-term external human resource management strategies.This can go as far as a partial reconciliation of interests.A high level of trade union density contributes towards strengthening the LI-HRMS.A decrease in the trade union density can, for the employees, lead to a deterioration in the pay-performance-relation.Limits, however, exist through the interest of the employers in the co-operation and self-motivation of workers.Furthermore, the longterm commitment does not change when the trade union density sinks.A low density of membership in employers' associations has a strengthening effect on the strategy.When membership density is low, the supply of working and remuneration conditions varies more.This can bring about incentives for the migration of employees and require efforts on the part of the employers to keep employees in the firm.The effects of the degree of organisation will, however, be lower in comparison to the SE-HRMS.The stronger the coalition between the employers and the "state" (in particular the government but also political parties), the greater the pressure on employers to implement the LI-HRMS.With such a coalition, stronger dismissal protection legislation, co-determination laws etc. become more probable, leading towards a long-term internal strategy.The state also plays an important part for employees (and employers) as an alternative source of income.If, for example, social welfare benefits were easily accessible and substantial, this would make the (potential) labour force less dependent on earned income.Employers are therefore highly interested in the so-called "wagegap-rule": social welfare benefits or similar sources of income should lie significantly under the lowest level of earned income.In contrast, critics of the system in the 1980's demanded a basic income provided by the state on the grounds that this would result in pressure on the employers to improve the working and remuneration conditions for the workforce.

Concluding remarks
I have focussed on the combination of ideas based on utility and power theories for a political economy approach to explain human resource management strategies.The political personnel economy perspective sees human resource management as an element within a social and political area of conflict constituted by the interests of "capital" and "labour".Furthermore, it functions as the exercise of power and the control of the management over the workforce.It is material interests and conflicts of interests against the background of given power and market conditions that push ahead changes in human resource management strategies and explain differences.The explanatory model is based on simple assumptions and the relevant explanatory sketch works with stylised facts.Both the theory and the explanatory model can be expanded upon.For example, I distinguish between just two human resource management strategies.This typology could certainly be elaborated in a more complex manner (governed by theory) and could take into account mixed forms.Changes in preference could also be considered more carefully.If, for example, an ideology of individualism were to be more strongly established, the perceived importance of long-term employment and secure income would also alter; self-employment appears as an additional alternative to dependent employment etc.This in turn would have effects on the power conditions and therefore on human resource management strategies.An empirical test is indeed desirable but has yet to be established.