Bureaucratic Behavior : A Review of the Theory and its Application to Serbian Public Administration

A professional and competent public administration is necessary for successful transition from a socialist, centrally planned economy to a functioning market democracy. The difficulty lies in building an organized and effective civil service sector. This paper gives a brief overview of formal bureaucratic reasoning, beginning with Niskanen’s theory, followed by the principal-agent model. The “Weberian state hypothesis,” which provides an alternative view of bureaucracies in less developed countries, is subsequently explored. Finally, the currant state of Serbian public administration is described, along with a discussion of challenges to be addressed in the future.


Introduction
The transition from a centrally planned socialist economy to a functioning market democracy requires significant institutional reform.This can entail creation of wholly new organizations in addition to revision of the existing political and economic institutions.Professional and legally bound public administration is necessary to achieve these ends.
Under Communist rule, the civil service sector of Central and Eastern European countries and the Soviet Union were highly politicized.State administration became only an implementation apparatus with no role in policymaking.Civil service agencies were under the full political governance of the Communist party.In Yugoslavia, however, this situation was somewhat different because its bureaucracy has long been impartial, even under Communist rule.Miloshevic's authoritarian regime undermined this tradition during the last decade (Verheijen and Rabrenovic, 2001).After the democratic changes of Serbia in October 2000, the new government was faced with a challenge to reform its civil service sector.Despite numerous initiatives to improve bureaucratic structures, the Serbian public administration system remained largely unchanged.This paper describes the features of a well-organized and effective bureaucracy, and illustrates Serbia's status relative to this ideal.Section two outlines formal reasoning regarding bureaucracy, beginning with Niskanen's theory followed by the principal-agent model.These two theories were chosen because of their long-held acceptance.Section three explains the "Weberian state hypothesis", an alternative view pertaining to the bureaucracies of less developed countries.Section four unites theory with empirical findings about Serbian public administration and explains challenges that must be addressed in the future.Niskanen (1971) was the first public choice theorist to model the behavior of bureaucracies.His theory rests on the following assumptions.First, the bureau has a virtual monopoly on true supply cost information.Second, the bureau knows the legislature's demand for the bureau's services.These assumptions allow agencies to make all-or-nothing offers (concerning both the budget and output) to the legislature.Consequently, the budget of a bureau would always be above the point where marginal public benefits (from the activities performed by the bureau) equal marginal costs.In other words, it is assumed that the bureau's principal objective is to secure an ever-larger budget.At the same time, the agency is determined to produce an output above the social optimal level.

Theoretical Framework
As demonstrated by Mueller (2003, p.363-364) this concept may be expressed mathematically.The budget of the bureau is a function of the final output of the bureau's services.This function could be regarded as a public benefit which, as the first and second derivations show, increases with increasing output, but at a diminishing rate.If P is the agency's budget, then: P=P(q), P'>0, P''<0 (1) On the other hand, the agency has a cost function T that increases at an increasing rate: The fact that agencies can make take-it-or-leave-it proposals to the legislature results in bureaucrats having exclusive knowledge pertaining to this cost function.The legislature is consequently handicapped in its ability to determine whether the output of the agency reflects the socially optimal level.Therefore, the bureau can maximize its budget with only one constraint, to cover the output's total costs, that is: max P(q) s.t.P(q) = T(q) Z=P(q)+λ(P(q)-T(q)) P'(q)= λ/1+λ T'(q) From the sponsor's perspective, the optimal condition requires that P'(q)=T'(q), but because the Lagrangian multiplier is a positive value, P'< T'.
As Mueller asserts, Niskanen's theory was a pioneering effort to model bureaucratic behavior, but new theories later emerged with different assumptions and hence, conclusions.Of particular interest is the principal-agent model.It is necessary before proceeding to clarify some simplifications on which Niskanen's analysis rests.
First, as noted by Bendor and Moe (1985, p.756) Niskanen assumes that decision makers in both the agency and the legislature base their choices on complex mathematics.These calculations would be beyond the capacity of most persons.Second, Niskanen's analysis rests on the assumption of perfect information on the part of the agency.There is only information asymmetry between the bureau and the legislature.In other words, agencies hold all the cards.
Niskanen developed his theory on neoclassical grounds, so it is understandable that he uses this premise.Additionally, formal modeling requires the adoption of simplifications like these.The quest for a more realistic paradigm of bureaucratic behavior leads us to principal-agent models.
The principal-agent framework recognizes the hierarchical relationship between bureaus and the legislature, with the legislature as a principal and the bureau as its agent (Moe 1984, p.770).Politicians, acting as principals, are more powerful than Niskanen asserts, since in reality they issue orders, control the agenda, create the incentive structure, etc.Moreover, these models accept the notion of information asymmetry.This results in greater attention being paid to the incentive scheme and monitoring instruments.These schemes are used to mitigate the asymmetry and hence confront the problems of hidden action and hidden information (Moe 1984, p.766).In other words, for problems of political control of bureaucracy, principal-agent models are better suited than Niskanen's theory.
These models are used to explain the relationship between upper level bureaucrats and their subordinates.Namely, the existence of the residual (profit) motivates the manager to control his team members in private firms.In the same manner, principal-agent theory implies that the slack, the difference between the true costs of a bureau's services and what the bureau actually spends, can be used as an instrument of internal control (Moe 1984, p.763).
Incentives are of great importance in principal-agent models because they serve as a monitoring and control mechanism.However, risk aversion on the part of subordinates can make incentives expensive and/or inefficient.Similarly, Holstrom (1982) argues that asymmetric information would cripple im-plementation of any incentive scheme that could reconcile organizational efficiency with the self-interest of individuals.
Finally, in low-income countries there is a lack of political will on the part of the principals to create and implement strategies that will induce better performance of agents.Moreover, corrupt governments are not motivated to engage in their serious monitoring, which makes principal-agent approach less applicable (Rauch and Evans 2000, p. 51).The principal-agent approach could be implemented in these countries but only after significant structural reform of their bureaucracies.The Weberian approach to bureaucratic reform (Rauch and Evans;1999, 2000) attempts to define the administrative structures likely to induce economic growth.The next section is dedicated to this line of reasoning.

Weberian State Hypothesis 1
As mentioned earlier, Niskanen's work was a pioneering effort within neoclassical political economy.It aimed to describe the functioning of the bureaucracy and its impact on the growth of government.Following this man's arguments, many later theorists viewed bureaucrats and politicians as rational agents engaged in budget maximization and thus inefficient spending.The allocation of financial resources controlled by them, according to this view, is vulnerable to corruption and clientelism, rather than motivation to serve the public interest.Few theorists within this branch of literature view politicians and bureaucrats as altruistic, honest individuals driven by good intentions and harboring high values and principles.
The question of which state structures are necessary in a country promoting economic growth was long subservient to interest directed at the proper model to describe bureaucratic behavior.Despite this, economists such as Knack and Keefer, (1994) Rodrik, (1999) and Kaufmann (2004) began focusing their attention on basic principles of good governance.Knack and Keefer (1994) measured the quality of governance using indicators such as government repudiation of contracts, risk of expropriation, corruption, rule of law, and bureaucratic quality.
By the mid-1990s there was a growing consensus, among economists studying market transitions, that a suitable government structure is a precondition for successful reform.Formulation and implementation of both broad and specific macroeconomic policies necessitate a professional and effective public administration.In order to fulfill these complex tasks it was necessary to define basic principles on which good public administration should rest.
Max Weber, one of the influential founders of classical administrative thought, viewed the bureaucracy, independent from societal pressures, as a full-time agency devoted to administrative tasks.He believed that an administration's duties need specific definitions, and that such policies are vital in the construction of an advanced market economy (Ahrens 2002, p.142).Building on Weber's insights about public administration, Evans and Rauch (1999, p.752) define the Weberian-type bureaucracy as featuring the following institutional characteristics: 1. meritocratic recruitment of personnel based on their education and competitive examinations, 2. predictable prospects of long-term career rewards, 3. strictly defined, non-overlapping jurisdiction of each public administration within the state apparatus.In their empirical testing of characteristics of bureaucracy most related to good economic performance, Evans and Rauch choose meritocratic recruitment and predictable and rewarding career ladders.These factors were selected based on the ease by which they could be measured across countries.Evans and Rauch argue that meritocratic recruitment results in minimal personnel competence.Moreover, these employees typically eschew the agency's norms and values.The latter is of great importance for the creation of esprit de corps among bureaucrats and their sense of loyalty to the organization.In other words, meritocratic recruitment and promotion contribute to the development of the corporate identity among bureaucrats.This in turn increases the cost of corrupt behavior.Making this activity less attractive for public officials would reduce the implicit tax on the private sector, which is nearly 20% of entrepreneurial revenue in transitional countries (World Bank, 2004).Furthermore, since meritocratic procedures can attract more talented individuals and increase their integrity and level of professionalism.Therefore, competent public officers are better able to help private entrepreneurs to overcome problems of coordination and collective action.This is especially true when they are entering new markets.
Corporate coherence and minimization of corrupt activity is also facilitated by predictable long-term career rewards.These also contribute to a bureaucracy's propensity to invest in infrastructure rather than spend resources on personal promotion.

Serbian Public Administration: Current State of Affairs and Challenges to be Addressed in the Future
This paper has analyzed dominant theories of bureaucratic behavior.It began with paradigms that for a long time represented the mainstream view of public administration.Niskanen's theory, developed on neoclassical grounds, viewed civil servants as public officials whose rational political behavior may translate into economically irrational actions.Furthermore, as Prokopijevic (2000, p. 71-78) asserts, if we accept the methodological individualism on which public choice theory rests, it is hard to accept the notion that the principal objective of the bureaucrats is to serve the public interest.Bureaucrats, like other political participants (politicians, voters), are interested in the maximization of their own utility.In pursuit of material well-being, bureaucrats try to maximize resources under their control and allocate them according to their own demands (Ahrens 2002, p. 43).Niskanen's theory has been criticized on several grounds, mostly because it asserts that bureaucracy holds all the cards, whereas in reality legislature is the principal and therefore has greater authority.This is recognized in the principal-agent model.The main contribution of this model to bureaucratic theory has been the notion of legislature-agency information asymmetry.The model pays special attention to the incentive scheme and monitoring instruments which mitigate this asymmetry.By doing this, problems of hidden action and hidden information are neutralized.
In young democracies there are few checks and balances to limit the corrupt behavior of their politicians.Moreover, lack of political will on the part of the principals (politicians) to create and implement strategies that will induce better performance of agents, and to engage in their serious monitoring, make the principal-agent approach less applicable in these countries.In fact, the principal-agent model could be implemented but only after significant bureaucratic reform in these countries.The Weberian approach could serve as a guiding principle for reform and explain on what grounds this approach rests.
This section connects theory with empirical findings regarding Serbian civil services.The following is an outline of the primary flaws within Serbian public administration.
First, there are some characteristics of Serbian bureaucracy that markedly correspond to Niskanen's theory.Namely, within Serbian civil service the minister and his employees make one team, and they seek as many resources as possible so as to improve the position of their Ministry within the government.To secure this position they seek ever-larger budgets, more employees than necessary, better facilities, etc.This behavior usually leads to an inefficient allocation of resources (Sevic 2001, p.318).
Second, Serbian agencies rarely share the same institutional features as Weberian bureaucracies.For example, meritocratic recruitment of personnel is not a guiding principle of Serbian administration.Ideally, civil service is apolitical and professional; in reality close connections and/or membership in the dominant parties can help enrolment and advancement within the administration.At the same time, little job security exists.After a ruling party loses an election, the careers of current employees are placed at risk.Sweeping reforms and tremendous turnover commonly occur when the reigns of government change hands (Sevic 2001, p.313).The World Bank reached a similar conclusion.This institution's research (World Bank 2004, p.17) shows that within Serbian administration current senior management positions are mainly politicized and do not allow the development of a senior civil service.It is the Secretary General and Assistant Ministers that are usually held by political appointees, and they are the first to be replaced after the new minister's arrival.The World Bank warns that this kind of politicization has hampered effective management of ministries, with an adverse effect on both policymaking and implementation processes.
Finally, Serbian civil service fails to meet the third characteristic of Weberian bureaucracy because jurisdictions of public administrations are not strictly defined and indeed often overlap (Kavran and Rakic 2001, p.5).Other important features of Serbian civil service include widespread problems of motivation, a shortage of administrative equipment and infrastructure, weak mechanisms for inter-ministerial coordination, a rule-oriented, risk-averse administrative culture, and strong centralization of decision-making authority (Eriksen 2005, p.6).At the same time, according to Eriksen's research, Serbian administrative reform is overly legalistic and he believes that excessive legalism makes civil service rigid and hard to change.
The remuneration system present in Serbian public administration is another feature needing reformed.The contemporary salary system does not provide sufficient motivation for young qualified staffers to remain in the administration.At the same time, this system is costly, as Serbia maintains a high public sector wage bill in relation to GDP. The World Bank (2004, p.18) proposes "the reform of incentive structures should make remuneration packages more attractive and more performance-based, while containing the fiscal cost of the system".The World Bank (2004, p.10) also declares that "Serbia has made limited progress in reforming its public administration system, which could endanger formulation and implementation of future economic and social policies.Key conditions for effective reform are the adoption of a clear administrative reform strategy and action plan, and the creation of a reform management structure with clearly assigned responsibilities and accountability.Only after completing these two conditions, Serbia would be prepared to build merit-based and de-politicized civil service."