Peters (Ch.3) - The Recruitment of Public Administrators

C h a p t e r 3

The recruitment of publicadministrators

Merit versus patronage 86Representative bureaucracy 89Public versus private employment 91Methods of recruitment 92Methods of recruitment 110Summary 124Notes 127

8 5

Before anyone can make much progress toward administering a public program,the political system must enlist and train a group of public administrators. Thisfact is common sense, but it simply points to the importance of recruitment in thestudy of public administration. In order to be capable of saying what a publicorganization will do (and how it will do those things), we must first have someconception of who will perform its tasks and for what purposes – public or per-sonal – the personnel will act. Unlike earlier assumptions concerning organi-zational management, such as Weber’s ideal-type conceptualization of thebureaucrat or the Taylor scientific-management school, individuals who occupypositions in public organizations are not interchangeable parts.1 This is widelyunderstood for partisan political leaders – presidents, prime ministers, etc. – butthe same ideological and personality characteristics generally assumed to affectpolitical leadership often are not assumed to influence bureaucrats and theirbehavior in office.

Public servants, as well as any other political elite, bring to their jobs a hostof values, predispositions, and operating routines that will greatly affect the qualityof their performance in the bureaucratic setting, as well as the type of decisionsthey will make.2 Some of those predispositions are a function of their social, ethnicand economic backgrounds. Other decisional premises arise out of the academicand professional preparation of the civil servants. Still other influences will comefrom the work experiences, both within and outside the public sector, of theseemployees. It is therefore very important to understand how governments selecttheir employees, and who within the societies seek to work for government andwhy they do.

Again we must emphasize that public administrators, even those at relat-ively low levels in the organizational hierarchy, are indeed public decision makers.The proverbial story of the judge having burned toast for breakfast and then sen-tencing the defendant to death may be as true, albeit in less extreme situations, ofthousands of administrators deciding on thousands of demands for governmentservices from clients each day.

This chapter examines the way in which governments select administrators,and thereby one of the ways in which they narrow the range of possible outcomesof the policy-making process. As well as being a question about predicting behav-ior and improving management, recruitment is also a question about democracy.One standard of good government is that it, and its decisions, should berepresentative of the public that is being served. Therefore, we will also be inter-ested in the extent to which the public service in a range of countries reflects thepublic and the consequences that this representativeness or lack of it may have onservices rendered.

Several somewhat conflicting themes have dominated the discussion of adminis-trative recruitment. The first of these has been the search for efficiency throughmerit recruitment. One of the defining characteristics of Weber’s model ofbureaucracy was that civil servants should be selected on the basis of achieve-

T H E P O L I T I C S O F B U R E A U C R A C Y

8 6

Merit versus patronage

ment criteria and merit, rather than ascriptive criteria such as caste, race, class, orlanguage. This has been referred to as selection by ‘‘neutral competence.’’3 Thecriteria used to select personnel, therefore, may be either their training and exper-tise for certain identified positions in the bureaucracy, or their general compe-tence and intellectual abilities. In either case, the assumption underlying meritselection of personnel is that bureaucracy must be able to recruit the best possiblepersonnel, and merit recruitment is the logical means of filling the available posi-tions with the best qualified personnel.

It should be remembered, however, that in some instances ascriptive cri-teria may be important for achieving the purposes of a public organization. Anorganization that must provide services to members of minority communities can,everything else being equal, do a better job by using a large number of minorityemployees in contact with the community than by using members of the dominantcommunity. The concept of ‘‘merit’’ in public office is more complex than it issometimes assumed to be, and race, creed, color and gender of public employeesmay be important in determining how well government really will deliver its ser-vices.4 As ethnicity, gender and other ascriptive criteria continue to increase inimportance as political cleavages within both industrialized and developing coun-tries, issues of representativeness along those dimensions are likely to be evenmore crucial in the future.

In developmental terms, a second impetus for adoption of merit recruitmentwas the desire to remove the appointment of administrative positions from politi-cal patronage and to require merit qualifications.5 Thus, in addition to removingthe inequalities and possible inefficiencies of ascriptive recruitment, the meritreforms of civil service were intended to remove the inefficiencies and favoritismof political appointment. As desirable as the idea of employing the best personpossible for each job in the public service may be from the perspective of enforc-ing the achievement norms of a modern society, and perhaps of achievingcontemporary goals of social equality in a developing society, some important inef-ficiencies may result from merit recruitment.

The potential inefficiencies of merit recruitment may be especially notice-able when contrasted with the alternative: political appointment. Merit recruit-ment appears to imply the more mechanistic conception of the administrator orbureaucrat as the value-free administrator of programs who will administer publicpolicies regardless of their intentions or impacts on society. It is assumed that suf-ficient legal and technical criteria will guide their choices and that personalcommitment to the program being administered, or rejection of it, will have littleinfluence on behavior. This conception of the administrator simply does notconform to the realities. Individuals selected by a spoils system were at least moredisposed toward the programs of the political party in power than the supposedlyneutral appointees of a merit system who may, in fact, be hostile to that program.A committed bureaucracy requires that political appointees be selected for somecombination of political disposition and administrative talent, however, and not fortheir political predisposition alone. Thus, ‘‘responsive competence’’ may be moreimportant than ‘‘neutral competence’’ in assuring that the tasks of government areaccomplished well.

This difference between political and merit appointment is, of course, one of

T H E R E C R U I T M E N T O F P U B L I C A D M I N I S T R A T O R S

8 7

degree. Virtually all political systems have some level at which appointments arequite clearly political – frequently referred to as ‘‘policy-making’’ positions – andthey also have jobs for which appointment is made on a relatively routine basis onsome sort of merit system. The question, then, is how far up the ladder of theadministrative hierarchy merit appointment, or at least not overtly politicalappointment, is intended to go; and conversely, what are the limits of politicalappointment?

The differences between two major administrative systems – the UnitedStates and the United Kingdom – illustrate the range that may exist in appoint-ment. A president in the United States can appoint approximately three thousandpeople to office, and four or even five echelons of political appointees may standbetween a career civil servant and the cabinet secretary.6 In the United Kingdomeach ministry will have only a few political appointments other than the ministeror secretary of state in charge – the largest number now is the Treasury with sixappointments – but even then, the major interface between political and adminis-trative leaders occurs between the minister and a single career civil servant, thepermanent secretary.7 The reforms of the British civil service have, however,opened up a number of positions to appointments, including the chief executivesof agencies and also positions in a growing number of ‘‘quangos.’’8 These systemsare obviously different, but each seems to function effectively within its owncontext and own political culture.

What accounts for the differences among countries in the extent to whichpolitical appointments are employed as a means of attempting to ensure com-pliance within an organization? One argument is that bureaucratic structuressimply have evolved through history and no one has seen any real reason to alterthem.9 A more rationalist hypothesis would be that the more fragmented thedecision-making structures of a government, the more likely it is to provide itsministers with a number of political appointees to provide some integration ofpolitical intentions and actual administration. The relative integration of the politi-cal elite and policy system in the United Kingdom, contrasted with that of theUnited States, could be taken to argue that the United Kingdom simply does notneed so many political appointees on top to create compliance within themachinery of government.

We have been using the United States as the example of a country with alarge number of political appointees, but several continental European countriescould be used equally well. In many of those European cases, however, the politi-cal appointees, rather than being inserted directly into the structure of the depart-ments or ministries, are linked directly to the minister and his or her privateoffice.10 This system removes some of the direct authority that the appointeesmight have over the department, but it provides the minister with both policyadvice and a group of people whom he can use anywhere within the ministry. Inother systems, the minister may be able to exercise some choice over the civil ser-vants who will serve him or her, dismissing some and calling others to office. Tosome extent this system combines the expertise of the career civil servant withthe commitment of the political appointee.

In addition to using political appointment through cabinets and other policyadvisors, there is an increase in more overt forms of politicization.11 For example,


8 8

there is an increasing number of appointments available in quasi-governmentalorganizations and the newly-created boards and agencies that permit politicalleaders to attempt to control these organizations through appointments.12 In addi-tion, the deinstitutionalization of many civil service systems provides additionalopportunities for making political appointments to positions that previously hadbeen reserved for career civil servants.

A second dominant theme in a discussion of recruitment into public administra-tive positions is that of equality of opportunity and representativeness of the publicbureaucracy.13 Since Kingsley coined the term ‘‘representative bureaucracy,’’there has been concern over the extent to which the bureaucracy does, or should,represent the characteristics of the population in whose name it administerspolicy. Thus, just as some scholars have emphasized the necessity of merit in therecruitment of public administrators, others have stressed the importance of pro-ducing a set of administrators whose social and economic characteristics aresimilar to those of the people with whom they will be working. The arguments aretwofold. The first is that narrow recruitment from any social stratum will tend tobias programs and policies. This is an especially important factor in social pro-grams, because there is a higher probability that these personnel will be workingmembers of minority communities and may tend to impose dominant groupvalues. Studies of teachers, social workers, the police and other types of publicemployees indicate a tendency to reward those clients who correspond toaccepted values in the dominant culture and to punish those who do not.14 Thisdifference in value structures may not only impair the personal interaction ofclient and administrator, but will also tend to prevent a number of qualified indi-viduals from receiving services.

The second argument in favor of greater representativeness is that theability of the public bureaucracy to hire personnel should be used as a positivemeans to alter the social and economic structure of the society. Thus, hiringminority community members can serve not only to attack any prejudices withinthe society but also to provide a means of economic advancement for members ofthe minority community. In the United States this has taken the form of ‘‘affirma-tive action’’ programs in which employers (especially government) are pledged tomake positive efforts to hire women and members of racial minorities.15 In Indiathe government is similarly required to attempt to hire members of castes andtribes that traditionally have been discriminated against in Indian society.16 Inethnically plural societies in which the differences among the ethnic communitiesare not necessarily those of dominance and submission, or in which the ethniccleavages are intensely politicized, the argument for representative recruitmentcan be altered to say that it can be used to preserve the social structure and therights of each of the ethnic communities in administering policy.17

In both versions of this argument, however, exists the underlying premisethat bureaucracies should be representative not simply because it is democraticfor them to be so, but because the pattern of recruitment will have a fundamental


8 9

Representative bureaucracy

effect on social structure and social stratification across time. Thus, governmentmay have a special responsibility for fostering greater equality of employmentand, over time, creating greater social equality.18 This recruitment pattern may beadopted to enhance social equality per se but may even be important for improv-ing the efficiency and effectiveness of service delivery by the public sector. Thisrepresentativeness is especially important as a means of creating a bureaucracythat will be similar to the clients with whom they come into contact on the job.

The political debate over representative bureaucracy has not ended, andcontinues at the present. In the spring of 2000 the Labour government in Britainexpressed a variety of concerns about the alleged elitism of leading British univer-sities and recruitment to the civil service, demanding that the universities and thecivil service look more like the country as a whole. As a consequence the govern-ment advocated additional means of enhancing the representativeness of thepublic service.19 The issues being raised were not at all dissimilar to those raisedby Kingsley in 1944, emphasizing both the efficiency and democratic elements ofthe representative’s argument.

Before anyone goes too far with the idea of representative bureaucracy, however,several important caveats must be advanced. The first is that research onrepresentative bureaucracy has consistently found an over-representation ofmiddle-class (broadly interpreted) backgrounds among civil servants. This is to beexpected. The civil service is in itself a middle-class occupation, and the sons anddaughters of the middle class tend to have a much higher probability of attainingmiddle-class occupations than do the sons and daughters of the working class.This tendency toward middle-class recruitment is due in part to the nature of edu-cational recruitment which, even in modern ‘‘welfare states,’’ tends to substantiallyover-represent the middle class, and due in part to the nature of the motivationsand incentives inculcated in middle-class households.20 In either case, the numberof middle-class offspring in the civil service is not a particularly damning findingfor the nature of the bureaucracy but, rather, reflects general patterns of socialstratification and mobility in society. Interestingly, this pattern of recruitment per-sists in societies that have sought to eliminate class barriers in public life, such asChina and the former Soviet Union.21 Such evidence as we have indicates that indi-viduals occupying positions in the upper echelons of these civil service systemstend to come from families of fathers who also held ‘‘middle-class’’ occupations.The civil service in particular tends greatly to over-represent individuals comingfrom civil service and political families.

A second caveat is that the advocacy of representative bureaucracy assumesthat the social class of parents will tend to determine the behavior of their off-spring. This is an underlying assumption in a large amount of elite research, butthe empirical research attempting to link background with behavior providesquite disappointing results.22 Putnam described the assumption that social back-ground influenced behavior in public office as ‘‘plausible, but ambiguous andunsubstantiated.’’23 Socialization is a lifelong process, and as administrators from


9 0

Caveats

working-class backgrounds attain middle-class status, they tend to adopt thevalues of that class rather than their class of origin.24 Further, although membersof minority ethnic groups may not be able to change their status quite so readily,those who are in posts (especially senior posts) in the public bureaucracy willtend to support the values of the dominant community at a higher rate than othermembers of the minority group.25 This phenomenon produces something of aparadox. Regardless of the degree of representativeness in the recruitment of civilservants, there will tend to be relatively great homogeneity of social and politicalvalues. We must remember, however, that the composition of the civil servicemay be as important for symbolic and political purposes as it is for ‘‘real,’’decision-making purposes.

Finally, we must understand that there may not necessarily be as broad agap between merit recruitment and programs of ‘‘affirmative action’’ as theremight appear to be at first glance.26 To some degree, the possession of relevantascriptive criteria may be an important qualification for the efficient administrationof public programs, especially at the client-contact level of the bureaucracy. Lan-guage, race, or class differences may prevent the adequate administration ofpublic programs because clients may perceive these differences as a measure ofthe program’s commitment to their needs. To prevent those differences frombecoming too significant in administration, some attention to ethnic or gender bal-ancing of personnel must be given. Thus, to some degree, defining the personbest able to carry out a job can depend upon their demographic characteristicsjust as it can depend upon formal education and the possession of certain skills.

A final general question about the recruitment and retention of civil servants is therelationship between the advantages of public and private employment in terms ofsalaries, benefits and working conditions. Most citizens want an efficient and well-qualified civil service, but many do not want that civil service to compete exces-sively with the private sector for the best personnel. The attractiveness of publicjobs is especially great in times of recession because of the relative security of agovernment job. The argument against making civil service employment tooattractive is that, through taxation, employers are actually coerced into supportingtheir competition in the labor market; further, the public sector lacks any effectivemeans of pricing most of its products. Therefore, governments can, to a point,drive up the price of labor to an unreasonable level from the viewpoint of actualproductivity of personnel if they were employed in the market economy. This dis-economy may become especially evident when public personnel are allowed tounionize. Thus, we come down to a rather simple question of the relative demandfor public and private goods and the consequent willingness to pay for each typeof goods. The evidence would appear to argue that there is considerably lessdemand for publicly produced goods – especially as they take on the character-istics of public goods – compared with the demand for private goods.27 Citizensappear to value a new car or a new TV more than they value a new park. Thislower demand for public goods may accentuate the diseconomies of public


9 1

Public versus private employment

employment. There may be no ready solution for this problem, but it is one thatmust be considered when we discuss the extent and type of recruitment into thepublic bureaucracy.

If we leave aside the economic arguments for hiring ‘‘the best and brightest’’versus an adequate civil service (sometimes at best), there are political andadministrative questions that must be addressed.28 One is that even if the publicdoes not express an overwhelming demand for publicly provided services, thoseservices are still important; we have already argued that, in fact, they are increas-ingly crucial for contemporary societies. Therefore, there may be a need to hirehigher-quality personnel than a simple economic model of the public sector mightdictate. Further, again there is a symbolic factor to be considered, and few citizenswould want much of their fate determined – whether that fate included a pension,air traffic control or nuclear destruction – by someone who is just ‘‘good enough’’for the job. The public sector does provide sensitive and important services, andhaving skilled and capable personnel is important.

In addition to the rather broad questions concerning recruitment outlined above,several issues deal with the more specific methods of recruitment and assess-ments of qualifications for positions. These questions are, of course, greatly sim-plified if political patronage or other sorts of non-achievement criteria are used,for then only simple appointment by the appropriate political official is required.

The first question is the type of training required for a position and, associated withit, the type of testing employed. Here we are interested primarily in the recruit-ment of the upper echelons of the bureaucracy rather than the clerical positions forwhich relatively uniform skill requirements can be established. In general, recruit-ment to policy-making positions requires some sort of post-secondary education,with the major question becoming the degree of specialization of that education.This brings us to the now standard argument between the advocates of generalistsversus specialists in public bureaucracy.29 The generalist school, as typified bypractice in the United Kingdom, selects individuals for the top roles in the civilservice largely on the basis of general intellectual abilities and performance in post-secondary education. The Northcote–Trevelyan Report of 1854 not only called forthe establishment of a merit-based civil service in the United Kingdom but alsonoted that training in the classics was perhaps the best preparation for a futureadministrator.30 The assumption was, and largely still is, that general intelligence(rather than specialized education or training) is all that is required to master thetask of sifting information and preparing advice for ministers.

The tradition of the ‘‘talented amateur’’ has persisted in Britain despite theincreasing technological content of government work, and despite attacks on thepolicy by the Fulton Report (1968) and other public and private investigations of


9 2

Methods of recruitment

Education and training

the civil service.31 As is shown in Table 3.1, arts and humanities graduates decreasedsomewhat as a percentage of all entrants to the senior civil service, but that percent-age remained well over half of the total in 1987. Certainly those entering the highercivil service in Britain are extremely talented intellectually and will be able to mastermuch of the required material while on the job, but they begin their careers at asevere disadvantage when dealing with technical (including economic) questions.

Recruitment to the civil service in the majority of continental Europeancountries represents a different form of generalist education. In the majority ofthose countries the role of the administrator is quite similar to that of the jurist,and a law degree is a requirement for almost all senior positions. For example, inone study 66 percent of the senior civil service (Beamte) in the Federal Republicof Germany have legal degrees.32 Other studies have shown somewhat similarlevels of recruitment of lawyers in other continental European civil services.33

Even then, however, the degree of reliance on lawyers may differ by ministry. Inone study of the Austrian civil service several ministries (Interior Justice) hadmore than 80 percent lawyers among upper echelon employees, while others(Health, Agriculture) had less than ten percent.34 As with British civil servants,some training in technical matters will take place on the job, although it is muchmore common for German and other continental civil servants to take advanceddegrees in more technical subjects, especially economics. In addition, in Germanythere is a national civil service academy to provide training and education forthose working within the career structure so that they can more readily acquirethe needed technical education.

The United States and France represent two different forms of specialisttraining for civil service careers. In the United States people are recruited to manypositions in the civil service on the basis of having some particular educationalqualifications prior to entry. For example, if an individual is seeking a job with theDepartment of Agriculture, he or she is usually expected to have training in


9 3

Table 3.1 University concentrations of recruits to the higher civil service in the UnitedKingdom (percentage)

Arts and Social Natural sciences/ Otherhumanities science applied science

1961–67a 62 24 11 31975b 54 28 17 –1985 56 27 17 –1989 56 32 12 –

Sources: A. H. Halsey and I. M. Crewe, “Social Survey of the Civil Service,” The Civil Service(London: HMSO, 1968), vol. 3, pt. 1, p. 93; House of Commons, Expenditure Committee, 11thReport, 1976–7, The Civil Service (London: HMSO, July 25, 1977); Civil Service Commission,Annual Report (London: HMSO, 1985); Civil Service Commission, Annual Report (London:HMSO, 1989)

Notesa Direct entrants to the administrative class; percentage of total concentrations mentionedb Recruits for administrative trainee positions

agronomy, agricultural engineering, or some other relevant specialty. As a con-sequence of that pattern of recruitment, well over one-third of higher civil servantsin the US federal government have formal scientific or professional qualificationsof some type.35 In fairness, legal degrees are often a generalist qualification forworking in government rather than being used just for law-related positions.

In France, the specialized training needed for one to become a top civilservant is provided by government itself. The principal source of entry into thehigher civil service is the Ecole Nationale d’Administration (ENA), which pro-vides instruction in finance, management and law. The engineers and technicalstaff needed by government in its economic roles are trained in the severalgrandes ecoles, such as the Ecole Polytechnique. Lower-level administrators aretrained in the Instituts Regionaux d’Administration (IRA). In total, the Frenchgovernment runs approximately eighty different types of school for the training ofcivil servants.36 In this system the government can prepare its future employees inexactly the manner it wants, either as administrators (ENA and IRA) or as tech-nical and scientific staff.

Less developed countries are in a more difficult position when decidingbetween generalist and specialist recruitment strategies. In the first place, whencountries have recently gained independence, they are frequently left with a civilservice trained by the former colonial power and thereby trained according to thetraditions of that European country.37 Further, most underdeveloped countriesgenerally lack technical talent and must opt for a more generalist stance in recruit-ment into new positions. Some countries have attempted to replace an indigenoustechnical force with one drawn from Western countries – usually the former colo-nial power – but the demands of national pride and the need for jobs for their ownpeople frequently require that jobs be given to individuals from within the nation,even if they are not always as skilled as potential hires from abroad. At the sametime that the administrative system may be somewhat deficient in specializedtalent from an absolute point of view, it may have a relative monopoly on suchtalent within the country. A principal characteristic of many developing countriesis that the political system is forced into the position of becoming the major direc-tive force in social and economic reform. As one commentator put it:

While there is no uniform pattern the experience of many newly independ-ent countries shows a growing emphasis on centralized planning, direction,and implementation of development programs. Thus, the government reliesmore on the bureaucracy than the private sector to carry out the task ofnation and state building, economic growth, and social reforms – activitieswhich are preeminent in the consciousness of the rulers and the ruled.38

Reliance on the private sector has been increasing in most developing countries,in part because of external pressures, but compared to most developed countriesthere is less implementation capacity on which to rely.

There is, however, no single pattern of recruitment for governments indeveloping countries. For example, the countries of South and Central Americathat have been independent for a number of years have had time to develop theirown distinctive patterns of recruitment. In some instances these recruitment


9 4

patterns are generalist and highly politicized, while in others they may be moretechnical and merit based.39 Most of the African countries and many Asian coun-tries, e.g. India and Pakistan, follow patterns inherited from their former colonialmasters, often without the modernization of managerial and organizational tech-niques implemented over the past decades. Pressures from donors, e.g. the WorldBank and the United Nations, are pushing all these countries toward recruitmentand management practices more like those found in the developed ‘‘First World.’’

The Western model of development – speaking broadly, as the Western eco-nomic and social systems evolved by several significantly different paths – wasblessed with a long time span and the absence of developmental pressures frommass publics and organized segments of the society.40 The developing countriestoday are faced with producing change within the context of widely disseminatedinformation about the glories of development and consumerism. Their leadersface demands for increased production of consumer goods at the same time thatthey know the need for investment in capital projects. Those investments shouldbear greater productive benefits in the long run but require a short-term retreatfrom a consumer-oriented economy toward a more state-directed economy. Theyface additional pressures from abroad to abandon substantial elements of stateintervention in the economy in order to ‘‘let the market work.’’ Given these prob-lems and countervailing pressures, it is apparent that these societies have a press-ing need for specialized administrators capable of proposing some solutions tothese problems, and a large number of skilled personnel to actually manage eco-nomic enterprises that may be run directly by the state.

The centrality of administration and administrative functions as describedabove for the undeveloped countries, if taken to the logical extreme, might be a relat-ively accurate description of nature of bureaucracy in the former Soviet Union and, toa lesser extent, other formerly communist countries.41 These characterizationsremain largely true of countries such as China and Cuba.42 As the state became notonly an economic planner, regulator and adviser, but also the chief entrepreneur, theneed for specialized talent tends to increase in government. Thus the average Sovietadministrator, even if not administering a highly technical project, tended to havescientific or social science training. For example, by the 1960s, over 80 percent of allpolitburo members and regional elites had technical training and this pattern per-sisted through the end of the Soviet Union.43 Interestingly, the more generalisttalents of ideological argument and broad knowledge of the intended purposes of theSoviet state appeared to be devalued by this set of upper-level administrators, whodiffered little in this respect from public administrators in other societies.44 They hadthe same, if not greater, demands for production that faced other managers, and theyare often confronted with highly technical problems that only someone with a tech-nical background may fully understand. Even more important is the fact that politicalleaders as well as the administrative elites tend to have that technical education intheir backgrounds, so the civil service did not have the degree of monopoly overskills and information it might in other countries.45

The above discussion has dealt with the recruitment of the higher civilservice. At some point in the civil service hierarchy almost all recruitment is doneon the basis of narrow, job-specific criteria. Typists, for example, are hired becauseof their ability to type. In terms of sheer numbers, therefore, most civil service


9 5

recruitment is on the basis of specialized criteria and the ability to perform a spe-cific job. The changing nature of work in modern societies, however, is altering theneed for many of the specialized skills previously needed in large numbers – manymanagers now do their own typing on word processors. Increasingly the emphasisin public employment will be on conceptual rather than mechanical skills.

Related to the type of training a prospective civil servant is expected to have is thequestion of the means through which the applicant and the position are expectedto find each other. Again, there are two principal answers: centralized personnelorganizations, or recruitment by each individual agency seeking employees. Cen-tralized placement is practiced by the United States, the United Kingdom, France,Belgium, Italy and the majority of Third World countries. The last set of countrieshas used centralized placement in large part as a function of inherited systems ofadministration. In the centralized pattern of recruitment there is a central civilservice organization of some sort that is responsible for advertising new positions,testing applicants and selecting some smaller set of applicants for final selectionby the agency seeking the employee.

The usual procedure for centralized placement is that a line agency notifiesthe personnel organization of the existence of a vacant position, a competitiveexamination is held, and then the agency seeking the person is sent a list of threeor more names from which to select the new employee. The selection may bemade on the basis of personal interviews or simply by taking the individual withthe highest score on the examination or by any other rational or irrational criteria.This means of recruitment obviously meets the requirements of merit recruit-ment. Competitive tests are used to fill the position; these tests are centrallyadministered to prevent bias, and the hiring organization accepts only thosedeemed qualified on the basis of the examination. In practice, there may be waysaround the merit system, especially for those who have professional qualifica-tions, such as physicians, lawyers, librarians and the like. They may be judgedqualified simply on the basis of their degrees and certificates and require nofurther examination. In addition, recruitment to the senior civil service may bedone not so much on the basis of particular needs in particular departments, buton the basis of creating a cadre of administrative elites that will have lifetimecareers somewhere within government.

The second means of hiring and recruitment is used primarily by the north-ern European countries, Spain, and many Latin American countries. In thesesystems there is no central personnel organization; rather, each agency isresponsible for hiring its own personnel. The most common procedure is for thehiring agency to publish a notice of a vacancy and accept applications fromprospective employees. These applications are generally judged on the basis ofappropriate minimum qualifications for the job – especially legal training in theScandinavian countries (Sweden, Denmark, Norway), Germany, and Austria.After the individual is deemed minimally qualified, selection may be made on thebasis of less achievement-based criteria.


9 6

Job placement

This system of decentralized recruitment obviously allows considerable lati-tude for the use of partisan and ascriptive criteria in hiring public officials, whocan become tenured in office and virtually impossible for subsequent regimes toremove. Charges of partisanship in the civil service are indeed made in thesesystems, even in Sweden and Denmark with their long histories of civil serviceindependence and prestige. For example, there were some concerns expressed asto whether a bureaucracy recruited during the 35 years of Social Democraticgovernment could serve the bourgeois coalition elected in Sweden in 1976. Asimilar question emerged after the 1991 elections which produced another right-of-center government with an apparently more radical program of market-basedreforms.46 It is a simple matter to hire partisans when there are no formal restric-tions to prevent it, and the parties in power would be extremely foolish if they didnot try to provide employment for their own supporters. They can also employadministrators likely to be favorably disposed toward the programs they will beadministering. As with many administrative practices, this is not a simple case ofrecruitment by merit or by patronage, but rather something of an intermediatemeans of recruitment that combines some features of both ideal-type methods.There is the potential for substantial patronage, but these opportunities arerestrained in practice by the norms, procedures and pride of the administrators. Ineach country in Europe in which the recruitment of administrators by agency ispracticed – with the possible exception of Spain – the civil service is a sufficientlyinstitutionalized and respected profession that few practitioners would seek todemean it by an excessive or blatant use of the personnel powers they find them-selves possessing.

The nature of recruitment by agency is made more complex when federal-ism is introduced as another variable. In Germany and Switzerland recruitmentto public administrative positions is done not only by the individual agencies, butalso by separate and in some cases highly independent subnational politicalunits, which are, in turn, responsible for the administration of nationalprograms.47 This system is further complicated by the Swiss bureaucracy’sneed to preserve some balance among regional, linguistic and religious sub-populations within the civil service.48 In general, the use of subnational bodies toperform the recruitment function may provide even greater possibilities for theuse of non-merit criteria in recruitment. Nevertheless, in Germany and Switzer-land we find again that the norms of the bureaucratic system are sufficientlyingrained so that merit criteria are strenuously enforced. Those who are hiredhave the necessary qualifications for the position – legal training and prior legalexperience – must pass an examination and then they are made to undergo somesort of post-entry training in the work of administration before they are grantedpermanent positions as administrators.49 There is an attempt – in practice arather thorough one – to employ people who are formally qualified according tothe requirements of the law. Thus, in this case as in others, although the rigidityof bureaucracies is often an impediment to innovation, it can also serve as animportant protective device for the society in preventing illegal or immoralactions on the part of government.


9 7

A third question concerning recruitment is the extent to which the public serviceis a distinct career, one for which the individual may prepare specially and that isregarded as a separate career hierarchy from the rest of the economy. It is inter-esting to note that movement back and forth between public and private employ-ment – especially in policy-making positions – has been used as an indicator oftwo rather different relationships between society and the political systems. Onthe one hand, such movement is frequently taken to indicate a healthy congru-ence between the value structures of polity and society, a means of ensuring therepresentativeness of the bureaucratic structures, and even a means throughwhich ‘‘typical’’ citizens can exert some influence on public policy. If the publicbureaucracy is a more open career, then it is more likely that they will act in waysthat the public would find legitimate and acceptable.

On the other hand, such movement among sectors can also be taken to indi-cate the colonization of the society by bureaucrats or, conversely, the colonizationof the public service by representatives of certain vested interests in the society.The former of these negative perceptions is best illustrated by the concern of theFrench over the pantouflage, or ‘‘parachuting,’’ of upper echelon civil servants intoimportant and lucrative positions in the private economy.50 This is taken as an indi-cation of the attempt on the part of fonctionnaires and technocrats to manage thewhole of economy and society and not just the governmental apparatus. It alsomeans that a great deal of executive talent developed at public expense isexported to the private sector free of charge. A somewhat similar pattern is foundin Japan, where civil servants who have finished one career in government (espe-cially in MITI – the Ministry of Trade and Industry – and other economics mini-stries) are then hired into powerful and prestigious positions in the privatesector.51 This type of movement may help ensure that government in Japan con-tinues to be supportive of the private sector.

The second, and negative, conception of the lateral movement between thepublic and private sectors is observed in the United States. The sentiment is oftenvoiced that too much of American government is being run by administrators cur-rently on leave from major corporations, major unions and other significant inter-ests in the society.52 Consequently, there is a belief that much of government ismanaged for the benefit of those interests rather than for the benefit of the publicat large. Of course, in the United States there is also a good deal of movementfrom government careers into management positions in the private industry, espe-cially from the military and civilian defense positions into defense contractingfirms. The Ethics in Government Act passed during the Carter administration has,however, limited the rapidity with which such movements can be made legally.53

Other countries have also sought to regulate the movement of civil servants intopositions in the private sector that are potential sources of a conflict of interest.54

The degree of concern over, or distrust of, lateral movements betweenpublic and private sectors would appear to be a function of several normative con-cerns of the society, especially as they relate to the administrative roles of govern-ment. One concern is the perception of the values, job and norms of the publicbureaucracy as being distinct from those of the private sector. In the French case,


9 8

Career distinctiveness

the fonctionnaires are perceived as a special group within society, by themselvesand by the general population. This perception contains some positive and somenegative elements, but the most common is that pantouflage is a means throughwhich they may seek to impose their conception of society onto the society. Thisis especially true of placing former employees of the public sector into key eco-nomic positions to ensure close coordination of the economy with the policies ofgovernment.55

In other societies that seek as much as possible to distinguish bureaucraticcareers from private careers, it is rather clear that employment as a public admini-strator is supposed to carry with it a rather distinct set of values and decisionalpremises. For example, in Germany and Sweden the public administrator hasbeen traditionally conceived of in a modified legal role.56 It is assumed that he orshe will act much as would a judge in impartially administering programs pro bonopublico and in accordance with the letter of the law. This may be too much toexpect from a mere human, but this separation and idealization of administratorand career patterns has been useful in justifying decisions made by administratorsin societies that rely heavily on administration in the conduct of public business.The tradition of the United Kingdom, Canada and other ‘‘Anglo-American Demo-cracies’’ has also been of a separation of public and private careers but without ajudicial interpretation of the role. Rather, the role of the public servant is con-ceived to be just that – someone who will devote a working lifetime to the serviceof the public.57

A second and related normative concern is the extent to which the societyfears bureaucracy and therefore seeks to prevent the development of a large andinflexible bureaucratic structure atop society. There are a number of means ofcontrolling the development of such a bureaucracy, and lateral entry at the upperechelons is certainly one of them.58 This is perhaps the logical extension of theidea of the amateur in administration, but it is one way in which general socialvalues can be injected into the conduct of government and administration, recog-nizing all the while that this will likely reduce the efficiency of organizationsalready attacked as being inefficient.

Civil service systems differ markedly in the extent to which their membershave experience outside government. At one end of the dimension is the UnitedKingdom, where civil servants rarely have experience outside government andtend to remain in government for their entire working lifetime. For example, inone study of careers in the British civil service, it was found that only 29 percentof senior civil servants had any working experience outside central government.Of that 29 percent, eight percent had experience in some other type of govern-ment and 12 percent had experience in teaching. Only seven percent had workedin private sector firms.59 That particular study is over a decade old, and therehave been some changes in the recruitments of senior managers in the Britishpublic sector. The continuing separation of policy and implementation, withimplementation functions being vested in agencies often headed by managershired from outside the civil service, has opened the system somewhat.60 To theextent that there have been changes, however, they have been to allow indi-viduals from the private sector to come into public positions, rather than allowingthose in the civil service to work elsewhere for short periods of time. The


9 9

Swedish civil service also had been relatively distinct from the private economy,but managerialist pressures have helped to open the system, at least in principle.In practice, very few senior managers have chosen to leave the private sector totake similar positions (Directors General) in the public sector.61 A study inSwitzerland showed that in most cantonal civil services there has been reducedmovement between the public and private sectors, but that for the federal civilservice younger civil servants are somewhat more likely to have had someexperience outside government.62

There are three patterns that assume less separation of the civil servicefrom the rest of the economy and society. One is the American revolving door, orthe ‘‘government of strangers’’ described by Heclo.63 In this system there is agreat deal of movement back and forth between the public and private sector, withmost people staying in government only a few years. They may, however, returnto government at a later date. The majority of these ‘‘in and outers’’ would be inpolitical appointments made by the President and his cabinet, but they would bein positions normally occupied by civil servants in other countries. Although theseappointees occupy the major positions in government, there is still a largepermanent civil service that can keep government functioning during the transi-tion from one regime to the next.64 The French government has used a variant ofthis approach by making political appointments to top posts in the administrationin order to maximize both political control of the bureaucracy and careeropportunities for politicians.65

A second pattern is that in which individuals leave government at acertain stage in their careers and go to work for the private sector. In theUnited States this has been a move primarily of retired military officers who goto work for defense manufacturers, but it is more common in France and Japanfor civilian administrators. It has been considered a normal part of the career,for example, for top civil servants from MITI to take lucrative posts in theprivate sector after retiring early from government.66 Civil servants of otherministries, to a lesser extent, may find similarly lucrative places to land afterleaving government. The search for a position after retirement is facilitatedwhen there are a number of public or quasi-public corporations into which theretiree can be placed.67

Finally, there are interesting cases in which being a civil servant is not con-sidered incompatible with a political career. This pattern is most evident inGermany, where the individual, once granted the status of Beamte, retains thatstatus almost without exception. The civil servant can leave government servicetemporarily while pursuing a political career and then return to service or beretired early with a full pension. In a typical lower house of the parliament (Bun-destag) one-half or more of the members will be civil servants on leave for a politi-cal career.68 Likewise in France the members of the Grands Corps remain sowhether they are actually serving in that capacity or not, and the National Assem-bly and the ministries are well-stocked with members of the Corps. Of course, inChina, the former Soviet Union, and other communist and single-party countries,the connection between administrative and political careers was very close, andthat closeness is necessary as the hegemonic party provides an all-purpose elitefor the management of the country.


1 0 0

We have already mentioned the question of incentives when discussing one of themore general aspects of recruitment in public bureaucracies. We now discussmore of the methods available to public administration to recruit and maintaintheir personnel. In a general overview of organizations and membership in organi-zations, Clark and Wilson developed a classification of the types of incentives thatan organization can offer its members; the three types of incentives mentionedwere material, purposive and solidary.69 Material incentives are factors such aspay, benefits and direct financial rewards. Purposive incentives are related to theability of the individual within the organization to have some influence over theshape of public policy adopted and implemented by government, or simply to getsomething done on the job. Finally, solidary incentives derive from the socialaspects of employment and group membership, which in the case of publicemployment may, in some countries, involve belonging to one of the more presti-gious organizations in the society.

Each of these incentives has, almost as a mirror image, a possible disincen-tive for working in the public sector. Some people may be attracted by the salariesof government jobs, but in most countries the top jobs in government are paidsubstantially less than jobs with similar responsibilities in the private sector.70

Likewise, many people are attracted by the opportunity to achieve policy goalsthrough working in government, but others are frustrated by the slowness withwhich government often appears to move and the barriers to action. Finally,public employees have become the targets of increasingly adverse publicity andattacks on their skills and integrity – often from their own political masters – thathave deterred more people from joining the government service.71 Thus, we canfind out why people say they choose not to join the public service, and why theyleave, as well as looking at why they say they join.

Any organization will potentially provide some of each of these three typesof incentives to employees. However, there are cross-national differences in theextent to which each of the three is perceived as an effective means of motivationby current and potential administrators. Some evidence about these differencescan be gained through survey data, although such data are available for only alimited number of systems, largely from Western nations. As is shown in Table3.2, there are some differences in response patterns even in this relatively homo-geneous set of countries. In the first place, it is interesting to note that purposiveincentives, which might have been thought to be the most significant means ofinfluencing people to join the bureaucracy, are not that important. In the cases forwhich we have data, one of the other incentives, most often solidary, is mentionedby a larger percentage of the respondents. The only exception to that generaliza-tion is Japan, where responses from new entrants into the civil service demon-strate a very high level of purposive incentives. This reflects in part the elitenature of Japanese public bureaucracy and its central position in governingJapan.72

It might be thought that the relative undervaluing of purposive incentiveswas a function of the subgroup within the civil service about which we have mostinformation. This group is composed largely of administrators near the top of the


1 0 1

Incentives and motivation

Tabl

e 3.

2In

cent

ives

in re

crui

tmen

t and

rete

ntio

n of

civ

il se

rvan

ts (p

erce

ntag

e)

Ince

ntiv

esU

nite

dU

nite

dFr

ance

cN

ewSp

ain

eIta

lyf

Turk

eyg

Indi

ah

Switz

erla

ndi

Aus

tralia

jJa

pank

King

dom

aSt

ates

b(1

969)

Zeal

and

d(1

967)

(197

8)(1

965)

(196

8)(1

987)

(197

5)(1

990)

(196

7)(1

985)

(196

6)Jo

inSt

ay in

Mat

eria

l20

3724

3621

4825

3820

3430

7Pu

rpos

ive

1923

3227

3911

126

5213

2749

Solid

ary

5640

3421

3036

5629

2226

156

Oth

er6

–10

1610

66

266

––

38To

tal

101

100

100

100

100

101

9910

010

073

7210

0

Sour

ces:

aBr

ian

Cha

pman

, “Pr

ofile

of a

Pro

fess

ion:

The

Adm

inis

trativ

e C

lass

of t

he C

ivil

Serv

ice,

” in

The

Civ

il Se

rvic

e(F

ulto

n Re

port)

(Lon

don:

HM

SO),

v. 3

, pt.

2,p.

12

bU

S G

ener

al A

ccou

ntin

g O

ffice

, Rea

sons

Why

Car

eer M

embe

rs L

eft i

n Fi

scal

Yea

r 198

5(W

ashi

ngto

n, D

C: U

S G

over

nmen

t Acc

ount

ing

Offi

ce, 1

987)

GA

O/G

GD

–87–

106F

Sc

Ezra

Sul

eim

an, P

oliti

cs, P

ower

and

Bur

eauc

racy

in F

ranc

e(P

rince

ton:

Prin

ceto

n U

nive

rsity

Pre

ss, 1

974)

, p. 1

20d

R. L

. Gre

en, M

. R. P

alm

er a

nd T

. J. S

ange

r, “W

hy T

hey

Leav

e,”

New

Zea

land

Jour

nal o

f Pub

lic A

dmin

istra

tion,

30

(196

7), p

. 27

eM

anue

l Gom

ez-R

eino

and

Fra

ncis

co A

ndre

s O

rizo,

“Bu

rocr

acia

Pub

lica

y Pr

ivad

a,”

in A

nale

s de

Mor

al S

ocia

l y E

cono

mic

a, S

ocio

logi

ca d

e la

Adm

inis

traci

on P

ublic

a Es

pano

la(M

adrid

: Ray

car,

1968

), p.

267

fFr

anco

Fer

rare

si, B

uroc

razi

a e

polit

ica

in It

alia

(Mila

n: Il

Mul

ino,

198

0), p

p. 1

20–1

gLe

slie

L. R

oos

and

Nor

alou

P. R

oos,

Man

ager

s of

Mod

erni

zatio

n: O

rgan

izat

ions

and

Elit

es in

Tur

key

(Cam

brid

ge, M

A: H

arva

rd U

nive

rsity

Pre

ss,

1971

), p.

123

hRi

char

d P.

Tau

be, B

urea

ucra

ts U

nder

Stre

ss(B

erke

ley,

CA

: Uni

vers

ity o

f Cal

iforn

ia P

ress

, 196

9)i

Paol

o U

rio,S

ocio

logi

e po

litiq

ue d

e la

hau

te a

dmin

istra

tion

publ

ique

de

la S

uiss

e(P

aris

: Eco

nom

ica,

198

9)j

Roya

l Com

mis

sion

on

Aus

tralia

n G

over

nmen

t Adm

inis

tratio

n, R

epor

t, A

ppen

dix

3 (C

anbe

rra:

Aus

tralia

n G

over

nmen

t Pub

lishi

ng S

ervi

ce, 1

976)

.Pe

rcen

tage

listi

ng e

ach

type

as

“im

porta

nt”

or “

very

impo

rtant

”k

Min

istry

of C

ivil

Serv

ice,

Japa

n, A

nnua

l Rep

ort(

Toky

o: M

inis

try o

f Civ

il Se

rvic

e, 1

990)

, p. 6

administrative hierarchy; they have been in office for some time and may there-fore be expected to have developed greater identification with the organizationrather than with the ostensible purposes of the organization. In Anthony Downs’terminology, they may have become conservers rather than advocates or zealots.73

The limited evidence reported from an Australian survey seems to dispute this,however, and the reasons reported for joining the civil service are less purposivethan those reported for remaining in the service.

The differences between the several sets of administrators for which wehave data are not particularly striking, but do lead to three rather interestingpoints. The first is the extremely high percentage of administrators in the UnitedKingdom who gave answers in terms of solidary incentives when questionedabout their jobs. This result would appear to conform nicely with the stereotype ofBritish administration as a set of ‘‘old boys’’ who conduct administration in a colle-gial, gentlemanly fashion and whose role as talented amateurs may prevent anyeffective policy initiative from arising from within the bureaucracy. Of course,numerous recent studies of the administrative apparatus of the United Kingdomindicate that although they may not be educated as experts in any particular tech-nical specialty, many administrators discharge quite significant roles in the forma-tion of policy – in fact, that has been known by the practitioners themselves forquite a long time.74 Still, it is interesting to note the extent to which the practition-ers give more social reasons for either joining or staying in their positions.

Of the countries on which we have data, the French upper-echelon adminis-trators reported the highest levels of purposive incentives.75 This finding also con-forms to the prevailing conception of the French bureaucracy as the groupedirigeante for the entire society.76 Traditionally, the way of getting things done inFrench government has been through administration, and we may expect thatadministrators would perceive a relatively great ability to accomplish thingsthrough their jobs. Finally, the Italian administrators gave a very high proportionof material answers, indicating the often cited tendency to use the bureaucracy asa means of personal advancement rather than as a force for policy change.77

Another study of motivations and incentives for working in the UnitedKingdom contrasted the motivations of public and private sector employeesdirectly.78 This study also contrasted the incentives deemed most important bymen and women. All groups examined tended to rate the challenge of the workand certain solidary benefits as most important in their choices of jobs (Table 3.3).Public sector employees, however, also tended to rank the security of their posi-tions very highly. In fairness, they also rated their potential contributions tosociety (a purposive incentive) very highly while this value was very low in theratings reported by private sector employees. For public sector employees,material incentives such as high earnings and fringe benefits were among theleast important incentives available. While the differences between the sectorswere not great, they do support most other findings and our general understand-ing of the differences between the public and private sector work forces. Theyalso substantiate the earlier finding of strong solidary incentives in the UnitedKingdom. A similar study in Canada, for example, found less importance attachedto solidary incentives, although employees in the public sector there were alsoless concerned with material rewards than were private sector employees.79


1 0 3

If the evidence on incentive structures for Western administrative systemsis rather spotty, then the information on non-Western administrative systemsappears virtually non-existent. The data for a by now rather dated sample ofTurkish administrators, however, show a close similarity to Western nations.80

Another study of administrative cadres in the People’s Republic of China foundthat material incentives were the most commonly cited, but that purposive incen-tives were a rather close second.81 Further, from a number of more descriptivestudies, we can rather quickly develop the hypothesis that the major incentives forjoining bureaucratic systems in non-Western societies are solidary and materialrather than purposive. In the first place, given the colonial backgrounds of most ofthese societies, the pattern of goal achievement through administration was notwell ingrained into these systems at the time of independence. Moreover, in theLatin American systems, which have been independent longer, the administratorsare not always protected by merit systems and tenure, so any attempt to useadministration to alter the existing social and economic arrangements in otherthan certain ways often meets with a prompt dismissal from office.82 There arealso more positive aspects to the attraction of the bureaucracy for many prospec-tive employees. The public bureaucracy is a stable and relatively remunerativeinstitution of the society, and compared with opportunities that may exist in theprivate economy, the opportunity to work in the public bureaucracy is frequentlyan extremely attractive economic option.83

The operation of the solidary incentives in administrative systems in devel-oping countries is perhaps less obvious. One of the social and cultural bases ofmany underdeveloped countries has been an emphasis on status and rank indefining social behavior. Also, in most of these societies the public bureaucracyhas been able to establish itself as a high-status occupation. This status may be inpart related to the relatively brief separation in time from the period in whichrecruitment to these governmental positions was determined almost entirely byascriptive criteria, and, in fact, the best families frequently chose to send theirsons into the public service. As Kearney and Harris said in speaking of thenCeylon (now Sri Lanka):

The great prestige enjoyed by the public servant has, however, probablycontributed at least as much as material advantage or employment securityto the attractiveness of a bureaucratic career. The social prestige of themodern bureaucrat is in large measure a heritage of Ceylon’s feudal andcolonial past.84

These authors go on to point out that the ‘‘social exclusiveness and supreme confi-dence’’ of colonial administrators tended to reinforce the impression that adminis-trative positions were to be equated with superior social position.85 Even after thepassage of decades from the time of colonial domination, some of these valuespersist. Further, in societies that value social position above the more achieve-ment-based criteria usually associated with Western societies, one may expect ahigh level of solidary incentives among those joining the bureaucracy.

The incentives of administrators joining the bureaucracy in the underdevel-oped world are obviously different from the types of incentives that we would


1 0 4

expect to characterize bureaucrats charged with bringing about important socialand economic changes. We have already noted the load being placed upon admin-istration in these transformations, and we find here a great disparity between therequirements of social change and the motivations of the people being recruited.86

This situation cannot, of course, provide an optimistic outlook for the future ofadministered change.

The study from the United Kingdom shown in Table 3.3 is particularly inter-esting, since it represented the reaction of a group of senior civil servants thenunder fire by an administration that sought to limit bureaucratic control overpolicy and to limit the pay and prestige that the service had enjoyed. Here, too, thereasons given for leaving senior government positions were a balance of incen-tives (or disincentives). It is especially interesting that solidary concerns, such asthe denigration of the civil service and political interference in personnel manage-ment, were more important than some of the more tangible attacks (e.g. on pay)then being made.87 Other subsequent surveys of senior government positions, andsome similar albeit less quantitative work on the British civil service weatheringsimilar attacks, confirmed the impression that pride in service is an importantfactor in recruitment and retention of top civil servants. The same pride in servicehas been found even more strongly in studies of the Canadian public service.

To conclude this discussion of incentives, it is possible to make some highlyconjectural statements about the nature of the bureaucracy in China, the formerSoviet Union and other communist countries. On the basis of descriptive accountsand descriptions of prior administrative systems, one can hypothesize that theincentive structures of these bureaucrats will be rather similar to those found foradministrators in Western societies. That is, there is a balance of material, purpo-sive and solidary incentives. The purposive incentives are rather obvious, giventhat even more than most Western societies, communist regimes are heavily ori-ented toward administration as the means of achieving goals. It might be expectedthat individuals would perceive some real ability to accomplish certain goalsthrough working in the administrative structures. The material incentives mayappear rather odd in supposedly classless societies, but we know well that thereare, if not classes, at least groups for which there are differential economic


1 0 5

Table 3.3 Most important characteristics of jobs for employees in the United Kingdom

Public sector Private sector

Females Males Females Males

Challenge Challenge Challenge ChallengeFriendly people Security Location AutonomySecurity Societal contribution Friendly people Friendly peopleSocietal contribution Friendly people Well-regarded Well-regarded

organization organizationLocation Autonomy Being appreciated Influence

Source: Beverly Alban Metcalfe, “What Motivates Managers: An Investigation of Gender andSectors of Employment,” Public Administration, 67 (1989) pp. 95–108

rewards.88 The public bureaucracy is one such group; being a member of the‘‘apparatus’’ of the state will generally pay off not only directly, but also indirectlythrough access to scarce consumer goods. Finally, one traditional description ofRussian administration was as a set of small and closely knit primary groups oper-ating within the context of a larger governmental structure.89 We may hypothesizethat this same sort of small group were still operating within the Soviet bureau-cracy, so there was a high level of solidary motivation for the workers within suchgroups. Also we might expect that there would be differences by age amongpublic servants in these regimes. Those with their roots in the early days of therevolutions and World War II might think of their work in more patriotic, solidaryterms, while younger workers might see the civil service as simply a well-paidoccupation in a more stable society. Of course, these are only conjectures aboutthe motivations of these administrators, but there is evidence that each of theincentives is likely to be effective. What is not known is the relative strength ofthese motivations and incentives.

It has been established that money may not be the only, or even the best, meansof motivating potential employees to accept governmental careers, or to continuein those careers once they have been employed. However, pay is an importantissue in the public sector, not least because many citizens believe that govern-ment employees receive large salaries for little work.90 In addition, it is not onlythe pay received during the working lifetime of the employees that producespopular resentment, but also the variety of benefits received by civil servants, notleast of which is an inflation-proofed pension. The comparability of compensationin the public sector with that of the private sector is an important consideration indetermining the satisfaction of government workers with their jobs, and for deter-mining the satisfaction of citizens with their public servants.

Determining appropriate levels of compensation for government employeesis not as simple a task as it may appear. It is true that many public sector jobs aredirectly comparable to jobs in the private sector; the tasks of a secretary ingovernment are almost identical to those of a secretary in the private sector.91

However, some jobs in government have no private sector counterparts, and manyjobs that appear comparable may not be. Being a police officer is different fromany private sector job, even that of a private security guard. And although the jobof a government executive may resemble that of a private sector manager of asimilar-size firm, the private sector manager is spared the political responsibility,media exposure, the managerial difficulties (for example, the convoluted person-nel management practices of a civil service system) of a manager in government.In addition, determining the value of an index-linked pension, or of the relativesecurity of public employment, involves a number of assumptions about the futurerates of inflation, future rates of unemployment and the preferences of workers forfuture versus current income. President Ronald Reagan once suggested that civilservants be paid 94 percent of what would be earned for comparable jobs in theprivate sector; the missing six percent was for the pension and for job security.92


1 0 6

Pay in the public sector

This figure was, however, only a guess at best of what the true value is of thosecomponents of the civil service compensation package.

Most studies of pay comparability between the public and private sectorsfind that government employees at the lower echelons are better paid than theirprivate sector counterparts.93 This is especially true of workers in unskilled orsemi-skilled positions, such as sanitation workers or bus drivers. This relativeadvantage of public sector workers derives, at least in part, from the ability ofthese workers to exert pressure on political leaders by real or threatened strikes,as well as the absence of the ‘‘bottom line’’ that would constrain a privateemployer. However, as responsibilities increase, government employees are paidless well than workers in the private sector. Those working at the very top ofpublic organizations frequently earn only a fraction of what they would be earningfor jobs with similar responsibilities in the private sector.94

There are some significant exceptions to the above generalization about therewards of public office. In societies with traditions of strong and prestigiousgovernment, senior civil servants are often well paid. For example, in the UnitedKingdom the very top officials of the civil service (the ‘‘open structure’’) are wellpaid in comparison to the majority of similar executives in the private economy,although their pay has been gradually falling behind since the late 1970s.95 Eventhough the pay of the top British civil servants has been falling (relative to theprivate sector) they have done well relative to the rest of the civil service, whosepay has slipped even more.96 These civil servants lack some of the perquisites,such as an automobile, of highly-placed private sector employees, but they dohave an index-linked pension to look forward to after retirement. Also, in societiesdominated by government, such as the former Soviet Union or many contempor-ary Third World countries, government is by far the most rewarding place to beemployed. Indeed one of the best paid civil service cadres is in the ‘‘Little AsianTiger,’’ Singapore, whose President reputedly has said that ‘‘. . . if you pay peoplepeanuts you get monkeys.’’

The relative position of public and private sector pay is demonstrated inTables 3.4 and 3.5 which relate average pay in the public sector to that in theprivate sector in a number of countries, and more detailed information aboutrewards of high office in a more limited number of countries. While there areinherent difficulties in making these comparisons – for example, differences in theoccupational structure within the civil service and the private sector – these dataare, to some degree, indicative of the relative position of the two sets of workers.In most industrialized democracies, civil servants earn on average about what isearned in the private sector, while in most underdeveloped countries they earnsubstantially more.97 This is indicative of the important position of government,and its employees, in the management and development of these societies. Theposition of top managers in industrialized democracies is not, however, so goodand they tend to be substantially less well paid.

Pay determination in the public sector is more than a question of personnelmanagement. It is also a crucial element in economic management. Since govern-ment now employs a very large proportion of the total labor force (18 percent inthe United States, 31 percent in the United Kingdom and 42 percent in Sweden),pay determination in the public sector influences economic conditions for the


1 0 7

economy as a whole. This is one aspect of the economy that government can influ-ence most directly, and when there is an attempt to implement an incomes policy,public sector wages serve as guidelines for the remainder of the economy.98

Further, when governments come into conflict with labor unions over wages, thesettlement reached can be used as an indicator of the power and resolve of thegovernment; for example, Conservative governments in the United Kingdom in1974, 1982 and in 1985. In contrast, the French government caved in to uniondemands several times during the 1990s with the consequence that the Frenchpublic sector continues to absorb more resources than those in most other Euro-pean countries.

Pay for public sector employees is determined in a number of ways. One isto link pay directly to changes in the private sector, or to consumer prices. Forexample, in the Netherlands, civil service pay has been adjusted biennially to takeinto account changes in private sector wages, while in Australia pay is adjustedannually on the basis of changes in prices.99 Pay for civil servants in the UnitedStates is nominally based upon comparability with the private sector, although the


1 0 8

Table 3.4 Pay of top civil servants as a percentage of average pay in the economy

Country Civil servants’ pay

United Kingdom 541Germany 487France 461Belgium 444Switzerland 385Denmark 302Sweden 223Norway 189

Source: Unpublished data from Christopher Hood and B. Guy Peters, Rewards of High PublicOffice (London: Sage, 1994)

Table 3.5 Ratio of central government wages to per capita income

Nation

Burundi 15.11Senegal 9.90Cameroon 7.39Egypt 5.70India 4.80Austria 1.06Australia 1.16Singapore 1.16Norway 1.48Sweden 1.49

Source: Peter Heller and Alan A. Tait, Government Employment and Pay: Some InternationalComparisons (Washington, DC: International Monetary Fund, 1983)

President and Congress make independent judgments about appropriate levels ofcompensation. In the 1980s the President and Congress consistently passed payincreases below the comparability figures recommended by the more independ-ent Advisory Committee on Federal Pay. The Pay Act of 1990 delinked civilservice pay from Congressional pay so that another political barrier to appropriatecompensation for civil servants was eliminated.100 In several countries pay move-ments in the public sector are directly linked to changes in compensation in theeconomy as a whole. This helps remove the incentive for politicians to hold downthe pay of civil servants. As noted above (pp. 65–7), attacking the civil service intheir pocketbooks is good politics in most countries, and politicians can earn sub-stantial credit by being tough on pay awards.

Negotiation with unions is the other major means of setting public sectorpay. In Denmark this bargaining occurs as a part of negotiations for wagesthroughout the economy, whereas in Italy, Canada and a number of other coun-tries, the negotiation is independent of other labor negotiations. In all negotia-tions, however, comparability is at least an implicit part of the bargaining process.Germany has a modified version of unionized pay setting in which the salaries ofordinary civil servants (Angestelle) are set by negotiations with unions; then Par-liament determines pay for top civil servants (Beamte). The Beamte have a legalright to a living suitable for their status as representatives of the state (the ‘‘ali-mentation principle’’).101

The traditional means of rewarding civil servants is to pay all individuals atthe same grade of the service the same amount, adjusted perhaps for seniority, orsocial factors such as the number of children supported by the employee. Thespread of managerialist ideas during the 1980s and 1990s has produced a greateruse of merit pay in the civil service, with differential rewards to employees basedupon their perceived performance.102 In some instances the better performers aregiven bonuses, while in other systems their salaries are adjusted upward on amore permanent basis. Further, in some of the bonus systems there are well-regulated means of determining those rewards, but in others a huge amount ofdiscretion is available to political and civil service leaders to reward their subordi-nates, and the bonuses are virtually secret.103 In some instances merit is deter-mined in advance, with many top government managers hired on performancecontracts that differ markedly across individuals.104 A few merit pay systems alsoprovide for reductions in salaries for poor performers, although it is generally dif-ficult to prove that performance is sufficiently poor to warrant reductions.

All merit pay systems, however, assume that civil servants are motivatedmore by these material rewards than by the solidary incentives characteristic ofstandard pay systems. These plans also assume that it is possible to measuremerit adequately and fairly.105 That may be relatively easy for many of the lowerlevel positions, e.g. typists or technicians, but becomes difficult for top managerialand professional positions. It becomes especially difficult for the civil servantswhose primary responsibility is policy advice for ministers. To the extent that thisis true, merit pay programs tend to emphasize the managerial and technicalnature of civil service jobs and may devalue the policy advice functions of thosepositions.

Despite differences in the manner in which public sector pay is determined,


1 0 9

several generalities can be made. First, a balance must be struck between fiscalconstraints and the need to attract and retain qualified personnel. This is obvi-ously more of a problem at the upper levels of government than at the bottom butmay become a problem in regard to people with specialized skills, such as com-puter operators, as well as to executives. Second, in times of economic constraintor insecurity, public sector pay is a convenient target for those who want tocontrol the costs of government. This is true despite the fact that personnel costsconstitute a relatively small proportion of total costs of government. For example,central government wages and salaries in Sweden are only six percent of totalexpenditure at that level; in the United States they are 11 percent.106 In both coun-tries wages and salaries for all levels of government are higher than for justcentral government, but still are less than one quarter of all expenditures. Publicsector wages and salaries are a larger proportion of total expenditures in less-developed countries; for example, salaries are over half of total expenditures inKenya and many other African countries.107 For much of the developed world,however, holding down public sector pay, despite its symbolic value, may be asmall portion of the total fiscal restraint needed to control public expenditure.

A third important factor about public sector wages is that public sectoremployees, by virtue of their central positions in the economy and society (publictransportation workers, firemen, defense, etc.) and their increasing levels ofunionization, are in powerful positions to influence their own levels of pay. Finally,pay itself is only a part of the total compensation package; to understand fully thebenefits of public employment, one must also take into account factors such asindex-linked pensions.108 Further, as noted above, employment in the public sectormay be a stepping stone to more lucrative private sector occupations, so thatlower wages in the short term may be an acceptable trade-off for very high long-term rewards.

We have been discussing the methods by which administrators are chosen andsome of the issues involved in the choice of methods. This section examines theeffects of these choices by scrutinizing the actual patterns of recruitment ofadministrators. Again, we are somewhat constrained by the lack of availability ofdata for administrative systems, especially those of the less-developed countries.Despite these constraints, it is possible to identify substantial recruitment informa-tion on the administrative systems of over 25 countries, although the timeliness ofthose data is varied. The data are on several dimensions of social background,academic preparation, and representativeness that can give important informationabout how administrators are chosen. However, some important caveats shouldbe noted. This information was gathered by different individuals, at differenttimes, and on somewhat different segments of the bureaucratic population. Themajority of data is concerned with upper-echelon administrators, but in somecases the coverage is broader. Therefore, care must be exercised in the interpre-tation of differences among these countries. Nevertheless, the data serve animportant function of illustrating the general directions of recruitment in each


1 1 0

Methods of recruitment

country. In each case, the source, year and definition of the administrative popu-lation (if different from “top” administrators) are noted.

The first dimension upon which there are data is the socio-economic back-ground of the administrators. Given that public administration is essentially amiddle-class occupation, the major variance here is in the occupation of thefathers – or the class of origin – of these administrators. We see from Tables 3.6and 3.7 that not only is public administration a middle-class occupation, but theorigins of the administrators are also primarily middle class. The definition ofclass of origin is somewhat fuzzy, especially the difference between upper classand middle class, but the largest single class of origin in each case is the middleclass. This is even more striking when the categories are collapsed into workingclass and bourgeois (by adding together middle and upper classes). In each case,few if any children of workers ever make it into the ranks of upper administration,with the most open system apparently being that of the United States, wherealmost one quarter of the federal executives came from working-class back-grounds. Despite the elitist image, the French civil service also had a relativelylarge percentage of inductees from the working class. Despite this apparent open-ness, the Mitterrand government developed a third means of entering ENA,directed at union officials and others of working-class backgrounds.109 Thisprogram had very little real success, was subsequently terminated, but then rein-stituted in the early 1990s. Germany would appear to be the most unrepresenta-tive of the civil service systems, for in at least one sample of upper administrators,none came from working-class backgrounds.110 Other studies show a morerepresentative recruitment pattern in Germany, but not a great deal morerepresentative than other European countries. Somewhat different samples usedby Aberbach, Putnam and Rockman show France to have fewer civil servantsfrom working-class backgrounds than does Germany, and the number of working-class inductees into the civil service in France appears to be decreasing.111 Asnoted, these differences are a function of different samples, different definitions ofthe higher civil service, and different definitions of class origins.

Lest we become too critical of the bureaucratic systems, we should note thatthis pattern of elitist recruitment is far from entirely the fault of elitism withinthese institutions. The bureaucracies are at the mercy of the educational system,and despite efforts by governments in most industrialized countries to make post-secondary education more widely available, it still remains a sanctuary of theupper and middle classes. Given the job requirements for the vast majority ofhigher administrative positions, be they for specialists or generalists, a post-secondary education is a virtual necessity, and in most countries – even ones withsocial democratic histories – relatively few working-class children are providedthat opportunity. This educational nexus is, in fact, the probable reason for therather positive showing of the United States. Post-secondary education is moreavailable in the United States than elsewhere, and consequently the pool of poten-tial applicants for the senior civil service is that much larger.112

We should not, on the other hand, be too quick to absolve the bureaucraciesof all guilt for their rather unrepresentative nature. All organizations tend to repli-cate themselves, and there is a definite tendency to recruit people like thosealready in the positions. This type of organizational bias is especially strong


1 1 1

Tabl

e 3.

6So

cial

cla

ss b

ackg

roun

ds o

f sen

ior c

ivil

serv

ants

(per

cent

age)

Soci

al c

lass

Uni

ted

Uni

ted

Fran

cec

Ger

man

yd

Den

mar

keSw

eden

fSw

itzer

land

gIta

lyh

orig

inKi

ngdo

ma

Stat

esb

(197

1–5)

(198

7)(1

974)

(197

2)(1

980)

(196

5)(1

995)

(195

9)

Upp

er28

1940

147

4261

17M

iddl

e58

4430

6943

4416

65W

orki

ng14

2114

1110

1423

5O

ther

/unk

now

n—

1616

199

——

13To

tal

100

100

100

100

9910

010

010

0

Can

ada

iIn

dia

jTu

rkey

kRe

publ

icSp

ain

mPa

kista

nn

Zam

bia

oN

ethe

rland

sp

(197

7–88

)(1

947–

63)

(196

2)of

Kor

eal

(196

7)(n

.d.)

(196

9)(1

988)

(196

2)

Upp

er44

1729

13—

225

55M

iddl

e19

6662

7096

7643

23W

orki

ng36

—1

74

223

18O

ther

/unk

now

n—

178

10—

—27

4To

tal

9910

010

010

010

010

098

100

Aus

tralia

qBe

lgiu

mr

Nor

way

sBr

azil

tFi

nlan

du

Israe

lvG

reec

ew

(197

0)(1

973)

(197

6)(1

972)

(197

0–80

)(1

986)

(198

8)

Upp

er13

3631

18—

Mid

dle

}78

59}7

153

5959

86W

orki

ng22

1114

114

23—

Oth

er/u

nkno

wn

—17

15—

6—

14To

tal

100

100

100

100

100

100

100

Sour

ces:

aPe

ter B

arbe

ris, T

he E

lite

of th

e El

ite(A

lder

shot

: Dar

tmou

th, 1

996)

; I. M

. Cre

we,

Soc

ial S

urve

y of

the

Civ

il Se

rvic

e, v

ol. 3

, pt.

1 of

The

Civ

il Se

rvic

e(T

heFu

lton

Repo

rt) (L

ondo

n: H

MSO

, 196

8) p

. 19

bW

. L. W

arne

r et a

l., T

he A

mer

ican

Fed

eral

Exe

cutiv

e(N

ew H

aven

: Yal

e U

nive

rsity

Pre

ss, 1

963)

, p. 2

9c

Pier

re R

acin

e, “

L’or

igin

e so

cial

e de

s en

argu

es,”

La

Fonc

tion

Publ

ique

. Cah

iers

Fra

nçai

s(L

a D

ocum

enta

tion

Fran

çais

e, 1

980)

dH

ans-U

lrich

Del

ien

and

Rena

te M

aynt

z, E

inste

llung

en d

er p

oliti

sch-

adm

inis

trativ

en E

lite

des

Bund

es 1

987

(Bam

berg

: Uni

vers

ity o

f Bam

berg

Leh

rstu

hl fü

rVe

rwal

tung

swis

sens

chaf

t, 19

88)

eT.

Rau

, Den

dan

sk c

enra

ladm

inis

tratio

n og

den

s em

bsm

and

(Cop

enha

gen:

Uni

vers

ity o

f Cop

enha

gen,

Insti

tut f

orsa

mfu

ndsf

ag, 1

974)

fU

. Chr

isto

ffers

on, U

. Mol

in, B

. Man

sson

and

L. S

tröm

berg

, Byr

åkra

ti oc

h po

litik

(Sto

ckho

lm: B

onni

ers,

197

2)g

Paol

o U

rio, S

ocio

logi

e po

litiq

ue d

e la

hau

te a

dmin

istra

tion

publ

ique

en

Suis

se(P

aris

: Eco

nom

ica,

198

9), p

. 146

hPa

olo

Am

mas

sari,

“L’

estra

zion

e so

cial

e de

i fun

zion

ari d

ello

Sta

to e

deg

li en

ti lo

cal,”

in A

mm

assa

ri et

al.,

Il B

uroc

rate

di f

ront

e al

le b

uroc

razi

a(M

ilan:

Giu

ffe, 1

969)

, p. 2

1i

Jacq

ues

Bour

gaul

t and

Sté

phan

e D

ion,

The

Cha

ngin

g Pr

ofile

of F

eder

al D

eput

y M

inis

ters

186

7–19

88(O

ttaw

a: C

anad

ian

Cen

ter f

or M

anag

emen

tD

evel

opm

ent,

1991

)j

V. S

ubra

man

ian,

Soc

ial B

ackg

roun

ds o

f Ind

ia’s

Adm

inis

trato

rs(N

ew D

elhi

: Min

istry

of I

nfor

mat

ion,

197

1), p

. 145

kC

. H. D

odd,

“Th

e So

cial

and

Edu

catio

nal B

ackg

roun

ds o

f Tur

kish

Offi

cial

s,”

Mid

dle

Easte

rn S

tudi

es, 1

(196

4), p

. 271

lD

ong

Suh

bark

, ‘‘K

orea

n H

ighe

r Civ

il Se

rvan

ts: T

heir

Soci

al B

ackg

roun

ds a

nd M

oral

e,’’

in B

yung

Chu

i Koh

, Asp

ects

of A

dmin

istra

tive

Dev

elop

men

t in

Sout

h Ko

rea

(Kal

amaz

oo, M

I: Ko

rea

Rese

arch

Pub

licat

ion,

196

7), p

. 27

mJu

an J.

Lin

z an

d A

man

do d

e M

igue

l, “L

a el

ite fu

nctio

naria

l Esp

añol

a an

te la

refo

rma

adm

inis

trativ

a,”

in A

ñale

s de

Mor

al S

ocia

l y E

cono

mic

a,So

ciol

ogic

al d

e la

Adm

inis

traci

on P

ublic

a Es

paño

la(M

adrid

: Ray

car,

1968

), pp

. 208

–9n

Ralp

h Br

aiba

nti,

“The

Hig

her B

urea

ucra

cy o

f Pak

ista

n,”

in R

alph

Bra

iban

ti, e

d., A

sian

Bur

eauc

ratic

Sys

tem

s Em

erge

nt fr

om th

e Br

itish

Impe

rial T

radi

tion

(Dur

ham

, NC

: Duk

e U

nive

rsity

Pre

ss, 1

966)

, p. 2

71o

Den

nis

L. D

resa

ng, “

Ethn

ic P

oliti

cs, R

epre

sent

ativ

e Bu

reau

crac

y an

d D

evel

opm

ent A

dmin

istra

tion;

The

Zam

bian

Cas

e,”

Am

eric

an P

oliti

cal S

cien

ceRe

view

, 68

(197

4), p

. 160

9p

F. M

. van

der

Mee

r and

L. J

. Rob

orgh

(199

3), A

mbe

tena

ren

in N

eder

land

(Alp

hen

aan

den

Rijn

: Will

ink)

, p. 3

34q

Roya

l Com

mis

sion

on

Aus

tralia

n G

over

nmen

t Adm

inis

tratio

n, R

epor

t(C

anbe

rra:

Aus

tralia

n G

over

nmen

t Pub

lishi

ng S

ervi

ce, 1

976)

; App

endi

x 3

rA

ndre

Mol

itor,

L’A

dmin

istra

tion

en B

elgi

que

(Bru

ssel

s: C

entre

de

Rech

erch

e et

d’in

form

atio

n so

cio-

polit

ique

, 197

4)s

Per L

aegr

eid

and

Paul

Ron

ess,

“D

e sta

tliga

ans

tälld

a i N

orge

,” in

Len

nart

Lund

quis

t and

Kris

ter S

tåhl

berg

, Byr

åkra

ter i

Nor

den

(Åbo

: Åbo

Aka

dem

i,19

83)

tPa

olo

Robe

rto M

otta

, “Th

e Br

azili

an B

urea

ucra

tic E

lite,

” un

publ

ishe

d Ph

.D. d

isse

rtatio

n, U

nive

rsity

of N

orth

Car

olin

a, 1

972

uKr

iste

r Stå

hlbe

rg, “

De

statli

ga A

nstä

llda

i Fin

land

,” in

Lun

dqui

st an

d St

åhlb

erg,

op.

cit.

vD

avid

R. N

achm

ias,

“Isr

ael’s

Bur

eauc

ratic

Elit

e: S

ocia

l Stru

ctur

e an

d Pa

trona

ge,”

Pub

lic A

dmin

istra

tion

Revi

ew, 5

1 (1

991)

, p. 4

14w

Geo

rge

Vend

endr

akis

and

Cat

herin

e D

. Pap

asta

thop

oulo

us, “

The

Hig

her C

ivil

Serv

ice

in G

reec

e,”

Inte

rnat

iona

l Rev

iew

of A

dmin

istra

tive

Scie

nce,

56

(199

0), p

. 474

Tabl

e 3.

7Ed

ucat

iona

l lev

el o

f sen

ior c

ivil

serv

ants

(per

cent

age)

Uni

ted

Uni

ted

Fran

cec

Ger

man

yd

Switz

erla

nde

Can

ada

fU

SSR

gJa

pan

hRe

publ

icKi

ngdo

ma

Stat

esb

(198

4)(1

983)

(198

7)(1

977–

88)

(195

0–66

)(1

975–

86)

of K

orea

i

(196

7)(1

984)

(198

8)

Hig

h sc

hool

23

012

101

2So

me

colle

ge27

107

5—

250

—1

Col

lege

gra

duat

e52

5593

2942

3499

Col

lege

+19

32—

6645

6540

—97

Tota

l10

010

010

010

099

101

100

100

100

Turk

eyj

Paki

stan

kBu

rma

lIsr

aelm

Net

herla

ndsn

Finl

and

oC

hina

pBe

lgiu

mq

Italy

r

(196

2)(1

948–

64)

(196

2)(1

969)

(198

8)(1

977)

(198

9)(1

973)

(196

1)

Hig

h sc

hool

20

——

3330

Som

e co

llege

——

2449

728

1918

Col

lege

gra

duat

e98

3576

2867

2147

Col

lege

+—

65—

237

1581

82

Tota

l10

010

010

010

010

010

010

010

010

0

Spai

ns

Swed

ent

Den

mar

kuN

orw

ayv

Braz

ilwG

reec

ex

Nig

eria

y

(199

6)(1

978)

(199

6)(1

970–

74)

(197

2)(1

988)

(197

9–83

)

Hig

h sc

hool

40So

me

colle

ge—

206

1114

102

Col

lege

gra

duat

e50

75C

olle

ge+

6080

9489

8640

19To

tal

100

100

100

100

100

100

96

Sour

ces:

aA

. H. H

alse

y an

d I.

M. C

rew

e, S

ocia

l Sur

vey

of th

e C

ivil

Serv

ice,

Vol

. 3, p

t. 1

of T

he C

ivil

Serv

ice

(The

Ful

ton

Repo

rt) (L

ondo

n: H

MSO

, 196

8), p

. 64

bU

S O

ffice

of P

erso

nnel

Man

agem

ent,

Stat

istic

s of

Fed

eral

Civ

ilian

Wor

kfor

ce, 1

986

(Was

hing

ton,

DC

: USO

PM, 1

986)

cJe

an-Lo

uis

Que

rmon

ne,L

’app

arei

l adm

inis

tratif

de

l’Éta

t(Pa

ris: E

ditio

ns d

e Se

uil,

1991

)

dH

ans-U

lrich

Der

lien

and

Rena

te M

aynt

z, E

inste

llung

en d

er p

oliti

sch-

adm

inis

trativ

e El

ite d

es B

unde

s 19

87(B

ambe

rg: U

nive

rsity

of B

ambe

rg, L

ehrs

tuhl

für

Verw

altu

ngsw

isse

nsch

aft,

1998

)e

Paol

o U

rio, S

ocio

logi

e po

litiq

ue d

e la

hau

te a

dmin

istra

tion

publ

ique

en

Suis

se(P

aris

: Eco

nom

ica,

198

9)f

Jacq

ues

Bour

gaul

t and

Sté

phan

e D

ion,

The

Cha

ngin

g Pr

ofile

of F

eder

al D

eput

y M

inis

ters

186

7–19

88(O

ttaw

a: C

anad

ian

Cen

ter f

or M

anag

emen

tD

evel

opm

ent,

1991

)g

Phili

p D

. Ste

war

t, Po

litic

al P

ower

in th

e So

viet

Uni

on(In

dian

apol

is: B

obbs

-Mer

rill,

1968

), p.

142

hB.

C. K

oh, J

apan

’s A

dmin

istra

tive

Elite

(Ber

kele

y: U

nive

rsity

of C

alifo

rnia

Pre

ss, 1

989)

, p. 2

21i

Pan

S. K

im, “

Who

Ser

ves

the

Stat

e: E

duca

tiona

l Bac

kgro

unds

of S

outh

Kor

ean

and

Japa

nese

Bur

eauc

rats,

” un

publ

ishe

d pa

per,

Dep

artm

ent o

f Urb

anSt

udie

s an

d Pu

blic

Adm

inis

tratio

n, O

ld D

omin

ion

Uni

vers

ity, 1

991

jC

. H. D

odd,

“Th

e So

cial

and

Edu

catio

nal B

ackg

roun

d of

Tur

kish

Offi

cial

s,”

Mid

dle

Easte

rn S

tudi

es, 1

(196

4), p

. 273

kRa

lph

Brai

bant

i, “T

he H

ighe

r Bur

eauc

racy

of P

akis

tan,

” in

Ral

ph B

raib

anti,

ed.

, Asi

an B

urea

ucra

tic S

yste

ms

Emer

gent

from

the

Briti

sh Im

peria

l Tra

ditio

n(D

urha

m, N

C: D

uke

Uni

vers

ity P

ress

, 196

6), p

p. 2

79–8

1l

Jam

es F

. Guy

out,

“Bur

eauc

ratic

Tra

nsfo

rmat

ion

in B

urm

a,”

in ib

id.,

p. 4

24m

Nim

rod

Raph

aeli,

“Th

e Se

nior

Civ

il Se

rvic

e in

Isra

el; N

otes

on

Som

e C

hara

cter

istic

s,”

Publ

ic A

dmin

istra

tion,

48

(197

0), p

. 174

nF.

Van

der

Mee

r and

J. R

aads

chel

ders

, “Se

nior

Civ

il Se

rvic

e in

the

Net

herla

nds,

” in

E. C

. Pag

e an

d V.

Wrig

ht, e

ds, B

urea

ucra

tic E

lites

in W

est

Euro

pean

Stat

es(O

xfor

d: O

xfor

d U

nive

rsity

Pre

ss, 1

999)

oC

harle

s D

ebba

sch,

La

fonc

tion

publ

ique

en

Euro

pe(P

aris

: CN

RS, 1

980)

; tot

al c

ivil

serv

ice

pXi

aow

ei Z

ang,

“El

ite F

orm

atio

n an

d th

e Bu

reau

crat

ic-T

echn

ocra

cy in

Pos

t-Mao

Chi

na,”

Stu

dies

in C

ompa

rativ

e C

omm

unis

m, 2

4 (1

991)

, pp.

114

–23

qA

ndre

Mol

itor,

L’A

dmin

istra

tion

en B

elgi

que

(Bru

ssel

s: C

entre

de

rech

erch

e et

d’in

form

atio

n so

cio-

polit

ique

, 197

4)r

Fran

co F

erra

resi

, Bur

ocra

zia

e po

litic

a in

Ital

ia(M

ilan:

Il M

ulin

o, 1

980)

sI.

Mol

ina,

“Sp

ain:

Stil

l the

Prim

acy

of C

orpo

ratis

m,”

in E

. C. P

age

and

V. W

right

, eds

, op.

cit.

tU

lf C

hris

toffe

rson

, “D

e sta

tligt

ans

tälli

da i

Sver

ige,

” in

Len

nart

Lund

quis

t and

Kris

ter S

tåhl

berg

, eds

., By

råkr

ater

i N

orde

n(Å

bo: Å

bo A

kade

mi,

1983

)u

H. N

. Jen

sen

and

T. K

nuds

on, “

The

Dan

ish

Cen

tral A

dmin

istra

tion,

” in

E. C

. Pag

e an

d V.

Wrig

ht, e

ds, o

p. c

it.v

Per L

aegr

eid

and

Paul

Ron

ess,

“D

e sta

tligt

ans

tålid

a i N

orge

,” in

ibid

.w

Paol

o Ro

berto

Mot

ta, T

he B

razi

lian

Bure

aucr

atic

Elit

e, u

npub

lishe

d Ph

.D. d

isse

rtatio

n, U

nive

rsity

of N

orth

Car

olin

a, 1

972.

xG

eorg

e Ve

nden

drak

is a

nd C

athe

rine

D. P

apas

tath

opou

los,

“Th

e H

ighe

r Civ

il Se

rvic

e in

Gre

ece,

” In

tern

atio

nal R

evie

w o

f Adm

inis

trativ

e Sc

ienc

es, 5

5(1

989)

, pp.

603

–29

yBo

la D

auda

, “Fa

llaci

es a

nd D

ilem

mas

: The

The

ory

of R

epre

sent

ativ

e Bu

reau

crac

y w

ith a

Par

ticul

ar R

efer

ence

to th

e N

iger

ian

Publ

ic S

ervi

ce19

50–1

986,

” In

tern

atio

nal R

evie

w o

f Adm

inis

trativ

e Sc

ienc

es, 5

6 (1

990)

, p. 4

77

during the personal interviews generally required for appointment to upper-echelon positions; the foreign service of most countries is the most obvious locusfor class bias in personnel selection. Bureaucracies also utilize formal rules, suchas the requirement for degrees or the difficulties in moving from one class of civilservant to another (for example, the administrative and executive classes in theBritish civil service prior to the Fulton Report), as a means of maintaining theirrecruitment patterns even in the face of democratization of the society and cul-tures of most Western societies. This tendency to preserve a more elitist recruit-ment pattern may not even be conscious, and those doing the recruiting maysimply be functioning with a mental picture of a good candidate that eliminatespotentially very good working-class talent.

We now have some inkling that education may be an important characteristic indescribing public administrators, and especially those at the upper levels of thehierarchy. Again, there is less than comprehensive data, but we can get someimpression about the educational levels that differ across cultures and educationalsystems. As much as possible, we have attempted to group the data into cate-gories that are comparable across countries and are meaningful to most readers.Doing this may have some costs in lost precision, but that disadvantage should beoffset by increased comparability of the data.

The previous discussion of the relationship of education and class in theselection of administrative personnel should have led us to expect a well-educatedgroup of people serving as upper-echelon administrators. This expectation is welljustified by the data. Almost universally, higher-level administrative personneltend to have some form of post-secondary education, with the majority havingcompleted the equivalent of a bachelor’s degree. In some cases, this educationmay be within the confines of a specialized administrative college, but there isnonetheless a definite post-secondary phase of education for most administrators.This is true even for less-developed countries that have a scarcity of educated per-sonnel. The data taken from Aberbach, Putnam and Rockman indicate that civilservants tend to be more educated than employees in similar types of positions inthe rest of the economy.113 This should be expected in most cases simply becausesuch education is a formal requirement for appointment. Interesting here is thatthe United States and Canada, which are frequently cited as having more ‘‘demo-cratic’’ political cultures, tend to have larger percentages of their upper civil ser-vices lacking any post-secondary education than do most other industrializeddemocracies. Israel, however, has by far the most open administrative structures,in part because of the newness of the country and in part because of the relativelypoor pay levels.

As well as having completed college or its equivalent, upper-level publicadministrators frequently have attended the more prestigious colleges and univer-sities. Studies of the British civil service, for example, have shown over two-thirdsof the senior civil service as having gone to Oxford or Cambridge. The civilservice appears to have declining appeal for ‘‘Oxbridge’’ graduates, however.114


1 1 6

Education

In his study of the backgrounds of Indian administrators, Subramanian reportedthat ‘‘the majority of recruits come from the six older and better known [universi-ties] . . . The significance of education in the right college is unmistakable.’’115

Suleiman also reports that the majority – and, in fact, over three-fourths – of theentrants to the ENA had their university education in Paris. Forty-two percent ofthese ENA entrants had their entire education in Paris. A later study shows overhalf of all ENA students having had all their education there, with another quarterhaving had their higher education there.116 The dominance of top universities isespecially pronounced in Japan, where the University of Tokyo and the Universityof Kyoto are the principal sources of talent for the senior civil service.117 A similarpattern is found for Seoul National University in South Korea and for the Universi-ties of Athens and Salonika in Greece.118 Thus, in these cases, the importance ofnot only attending college but also attending the right college is indeed unmistak-able.

The American pattern of recruitment from universities is somewhat differ-ent from that of other political systems. The (somewhat dated) analysis by Warneret al. of the college attendance of American career executives shows a ratherstrong influence of large state-supported universities in the education of adminis-trators.119 If foreign-service executives are excluded, none of the Ivy Leagueschools is among the top ten in terms of number of degrees held, and only threeare in the top 30. Among foreign-service executives, however, three of the IvyLeague are in the top ten, and all eight schools are in the top 30. Later evidencetends to confirm the position of large state universities, especially those of theMidwest, as the breeding ground of future civil servants. These data would appearto offer some support for the conception of American society and its administra-tive system as being somewhat more open than most. It further supports the con-tention that public service in the United States has been an important means ofsocial mobility, a factor we will elaborate on when discussing the ethnic represen-tativeness of the civil service.

The last question to be asked concerning the educational backgrounds ofthese administrators is the type of degree obtained. Here we are interested in thedegree of technical or functional expertise that the administrators are likely to becarrying into their work as a function of their college education. We have alreadyreported some data of this type for the United Kingdom, and they are reproducedin Tables 3.7 and 3.8 along with data for such other countries as were available.There is considerably more variance in the types of education received than in thelevel of education, with apparently three rather distinct groups of educationaldegree types in this non-random sampling of administrative systems.

The first pattern is represented by the United Kingdom, with a greatemphasis on general education, the arts and humanities, and consequently lessemphasis on technical ability. As was noted, this is indicative of the generalistconception of administration in the United Kingdom. A second type is typified byGermany, Austria and Sweden. These systems place heavy emphasis on legaltraining, and consequently about two-thirds of their administrators have legalbackgrounds. Many of the remainder also possess some form of professionalqualification, such as engineering, medical, or educational degrees. France issomewhat similar but goes a step further by providing most of the future


1 1 7

Tabl

e 3.

8C

olle

ge m

ajor

s of

sen

ior c

ivil

serv

ants

(for t

hose

with

col

lege

bac

kgro

unds

)

Uni

ted

Uni

ted

Ger

man

yc

Japa

nd

Repu

blic

Turk

eyf

Indi

ag

Can

ada

hSw

itzer

land

iIsr

aelj

King

dom

aSt

ates

b(1

987)

(198

6)of

Kor

eae

(196

3)(1

947–

63)

(197

7–88

)(1

987)

(198

6)M

ajor

(197

0–4)

(198

6–8)

(198

2)

Nat

ural

sci

ence

2632

823

534

3328

2017

Soci

al s

cien

ce—

——

166

9—

721

Hum

aniti

es52

23—

1813

1321

4211

27Ec

onom

ics

and

busi

ness

—28

18—

3216

3653

1225

Law

319

6359

3218

—19

4910

Oth

er18

—12

—15

131

—2

—To

tal

9910

210

110

010

010

010

014

210

110

0

Net

herla

ndsk

Belg

ium

lIta

lym

Den

mar

knFi

nlan

do

Nor

way

pBr

azilq

Gre

ece

rSp

ain

sFr

ance

t

Maj

or(1

988)

(197

6–84

)(1

972–

4)(1

993)

(198

0)(1

980)

(197

2)(1

988)

(199

6)(5

th R

epub

lic)

Nat

ural

sci

ence

2020

10—

148

2214

202

Soci

al s

cien

ce39

28—

1119

2622

542

Hum

aniti

es—

1237

——

12—

677

7Ec

onom

ics

and

busi

ness

——

—20

——

—19

206

Law

3235

5410

4338

2245

73O

ther

196

—19

2417

34—

3—

Tota

l10

010

110

110

010

010

110

010

010

014

0

Sour

ces:

aJo

el D

. Abe

rbac

h, R

ober

t D. P

utna

m a

nd B

ert A

. Roc

kman

, Bur

eauc

rats

and

Polit

icia

ns in

Wes

tern

Dem

ocra

cies

(Cam

brid

ge, M

A: H

arva

rd U

nive

rsity

Pres

s, 1

981)

, p. 5

2b

Joel

D. A

berb

ach,

Han

s-Ulri

ch D

erlie

n, R

anat

e M

aynt

z an

d Be

rt A

. Roc

kman

, “H

auts

fonc

tionn

aire

s fé

dera

ux a

mer

icai

ns e

t alle

man

ds: a

ttitu

des

tech

nocr

atiq

ues

et p

oliti

ques

,” R

evue

inte

rnat

iona

le d

es s

cien

ces

soci

ales

, 12

(199

0), p

. 7

cH

ans

Ulri

ch D

erlie

n an

d Re

nate

May

ntz,

Ein

stellu

ngen

der

pol

itisc

h-ad

min

istra

tiven

Elit

e de

s Bu

ndes

198

7(B

ambe

rg: U

nive

rsity

of B

ambe

rg, L

ehrtu

hl fü

rVe

rwal

tung

swis

sens

chaf

t, 19

88)

dB.

C. K

oh, J

apan

’s A

dmin

istra

tive

Elite

(Ber

kele

y: U

nive

rsity

of C

alifo

rnia

Pre

ss, 1

989)

p. 1

37e

Pan

S. K

im, “

Who

Ser

ves

the

Stat

e: E

duca

tiona

l Bac

kgro

unds

of S

outh

Kor

ean

and

Japa

nese

Bur

eauc

rats,

” un

publ

ishe

d pa

per,

Dep

artm

ent o

f Urb

anSt

udie

s an

d Pu

blic

Adm

inis

tratio

n, O

ld D

omin

ion

Uni

vers

ity, 1

991

fRe

publ

ic o

f Tur

key,

Offi

ce o

f the

Prim

e M

inis

ter,

Stat

e In

stitu

te o

f Sta

tistic

s, G

over

nmen

t Per

sonn

el S

ervi

ce S

tatis

tics

I (A

nkar

a: S

tate

Insti

tute

of S

tatis

tics,

1965

), pp

. 32–

9g

V. S

ubre

man

ian,

Soc

ial B

ackg

roun

d of

Indi

a’s

Adm

inis

trato

rs(N

ew D

elhi

, Min

istry

of E

duca

tion,

197

1), p

. 155

hJa

cque

s Bo

urga

ult a

nd S

téph

ane

Dio

n, T

he C

hang

ing

Profi

le o

f Fed

eral

Dep

uty

Min

iste

rs 1

867

to 1

988

(Otta

wa:

Can

adia

n C

ente

r for

Man

agem

ent

Dev

elop

men

t, 19

91)

iPa

olo

Urio

, Soc

iolo

gie

polit

ique

de

la h

aute

adm

inis

tratio

n pu

bliq

ue e

n Su

isse

(Par

is: E

cono

mic

a, 1

989)

, p. 4

0j

Dav

id N

achm

ias,

“Isr

ael’s

Bur

eauc

ratic

Elit

e: S

ocia

l Stru

ctur

e an

d Pa

trona

ge,”

Pub

lic A

dmin

istra

tion

Revi

ew, 5

1 (1

991)

, pp.

418

–19

kF.

van

der

Mee

r and

J. R

aads

chel

ders

, “Se

nior

Civ

il Se

rvic

e in

the

Net

herla

nds,

” in

E. C

. Pag

e an

d V.

Wrig

ht, e

ds, o

p. c

it.l

R. D

epre

and

Ann

ie H

onde

ghem

, “Re

crui

tmen

t, C

arrie

re e

t for

mat

ion

des

fonc

tionn

aire

s su

perie

urs

en B

elgi

que,

” in

Des

ire D

e Sa

edel

eer e

t al.,

La

haut

efo

nctio

n pu

bliq

ue e

n Be

lgiq

ue e

t dan

s le

s pa

ys in

dustr

ialis

és(B

russ

els:

Inte

rnat

iona

l Ins

titut

e of

Adm

inis

trativ

e Sc

ienc

e, 1

988)

mA

bera

ch, P

utna

m a

nd R

ockm

an, B

urea

ucra

ts an

d Po

litic

ians

nH

. N. J

ense

n an

d T.

Knu

dsen

, “D

anis

h C

entra

l Adm

inis

tratio

n,”

in E

. C. P

age

and

V. W

right

, eds

, op.

cit.

oKr

iste

r Sta

hlbe

rg, “

De

statli

ga a

nstå

lida

i Fin

land

,” in

ibid

.p

Per L

aegr

eid

and

Paul

Ron

ess,

“D

e sta

tligt

ans

tålid

a i N

orge

,” in

ibid

.q

Paol

o Ro

berto

Mot

ta, “

The

Braz

ilian

Bur

eauc

ratic

Elit

e,”

unpu

blis

hed

Ph.D

. dis

serta

tion,

Uni

vers

ity o

f Nor

th C

arol

ina,

197

2r

Geo

rge

Vend

endr

akis

and

Cat

herin

e D

. Pap

asta

thop

oulo

s, “

The

Hig

her C

ivil

Serv

ice

in G

reec

e,”

Inte

rnat

iona

l Rev

iew

of A

dmin

istra

tive

Scie

nces

, 55

(198

9), p

p. 6

03–2

9s

I. M

olin

a, “

Spai

n: S

till t

he P

rimac

y of

Cor

pora

tism

,” in

E. C

. Pag

e an

d V.

Wrig

ht, e

ds, B

urea

ucra

tic E

lite

in W

est E

urop

ean

Stat

es(O

xfor

d: O

xfor

dU

nive

rsity

Pre

ss, 1

999)

tL.

Rou

ban,

Les

Pre

fets

de la

Rep

ubliq

ue(P

aris

: Cev

ipof

, 200

0)

upper-echelon of the administration legal, management and economics trainingthrough ENA. All of these systems have a strong emphasis on the legal role ofthe civil service so that this pattern of education is crucial for success within thesystem.

The final pattern of educational backgrounds is typified by the United Statesand several underdeveloped countries. The principal characteristic of these coun-tries is the relatively large percentage of natural science (including engineering)backgrounds in the civil service. As was noted above, the United States tends tohire people with specialized backgrounds to do specialized jobs rather than hiringgenerally qualified personnel. In the Third World countries there is a need to con-centrate the available technical talent in the country and to make the most effi-cient use of this scarce resource. One way of doing this is to hire as much talentas possible in government and then use the government as the means of allocat-ing total societal resources. Moreover, given the relatively underdeveloped stateof the economies of many of these countries, the only real employment options foreducated individuals may be to work for the government.120 Many political consid-erations may prevent the public bureaucracy in underdeveloped countries fromfulfilling their potential for administering programs of social and economicchange. It would appear, however, from these data that many of the countries dohave the raw material, in terms of personnel within their bureaucracies (especiallyrelative to the pool of educated and trained talent available) that might make thosesocio-economic reforms successful.

Another question to be looked at in the presentation of background data on civilservices is the ethnic representativeness of the bureaucracies. Just as there issome cause for concern about the representativeness of public bureaucraciesaccording to social class, so is there concern over their equality in recruitment ofvarious minorities within the society. We may expect the same sort of pattern aswas found with respect to class, with the dominant community having a dispropor-tionate share of the members of the civil service, especially in elite positions. Asthe data in Table 3.9 show, these suspicions are confirmed. In most cases, there isa distinct over-representation of the dominant racial, language, or religious group.As with the findings for social class, this may be the result of applying the usualeducational criteria, not the result of overt discrimination.

Three special points should be made with respect to ethnic representationin the public bureaucracy. The first is that the data that are presented are primar-ily for upper-echelon personnel; as we go farther down the bureaucracy, theimportance of the representativeness of the organization should increase ratherthan decrease. We have noted the importance of the client-contact personnel ofagencies for the success of the agency in serving its clients. Such limited informa-tion as does exist on the lower echelons of public agencies indicates that theyare more representative than are upper, managerial positions.121 These morerepresentative lower echelons may therefore be expected to be more successful indealing with their clientele than would top management. Also, the public sector


1 2 0

Ethnic representativeness

Tabl

e 3.

9Et

hnic

repr

esen

tativ

enes

s of

pub

lic b

urea

ucra

cies

(per

cent

age)

Uni

ted

Stat

esa

Can

ada

bIsr

aelc

Mal

aysi

ad

Indi

ae

Zam

bia

fPa

pua

New

Gui

nea

g

Ethn

icTo

tal

Hig

her

Tota

lH

ighe

r(1

985)

(196

0)(1

974)

Tota

lH

ighe

rgr

oup

(199

0)(1

990)

(197

8)(1

977–

88)

(197

5)(1

975)

Tota

lH

ighe

r

Dom

inan

t63

873

7081

6796

7227

9250

Min

ority

3792

2730

1933

428

738

50To

tal

100

100

100

100

100

100

100

100

100

100

100

Sout

h A

frica

hLe

bano

ni

Switz

erla

ndj

Nig

eria

k

Tota

lEx

clud

ing

(195

5)La

ngua

geRe

ligio

n(1

988)

(198

9)la

bore

rs(1

992)

(198

7)(1

989)

Whi

te41

82M

aron

ite40

Fren

ch24

Cat

holic

38Ig

bo18

Col

ored

173

Sunn

i27

Ger

man

72Pr

otes

tant

58H

ausa

/A

sian

41

Shi’i

te4

Italia

n4

Oth

er/n

one

4Fu

lani

6Bl

ack

3814

Gre

ek O

rthod

ox12

Rom

ansh

—To

tal

100

Yoru

ba54

Tota

l10

010

0G

reek

Cat

holic

9To

tal

100

Oth

er/D

on’t

Dru

ze7

know

22To

tal

99To

tal

100

Sour

ces:

aU

nite

d St

ates

Offi

ce o

f Per

sonn

el M

anag

emen

t, A

ffirm

ativ

e Em

ploy

men

t Sta

tistic

s(W

ashi

ngto

n, D

C: U

SOPM

, bie

nnia

l); m

inor

ity�

non-

whi

teb

P. K

. Kur

uvill

a, “

Publ

ic S

ecto

r Rec

ruitm

ent i

n C

anad

a,”

Indi

an Jo

urna

l of P

ublic

Adm

inis

tratio

n, 2

6 (1

980)

, p. 8

6 Ja

cque

s Bo

urga

ult a

nd S

téph

ane

Dio

n, T

he C

hang

ing

Profi

le o

f Fed

eral

Dep

uty

Min

iste

rs 1

897–

1988

(Otta

wa:

Can

adia

n C

ente

r for

Man

agem

ent D

evel

opm

ent,

1991

);m

inor

ity�

franc

opho

nec

Dav

id N

achm

ias,

“Isr

ael’s

Bur

eauc

ratic

Elit

e: S

ocia

l Stru

ctur

e an

d Pa

trona

ge,”

Pub

lic A

dmin

istra

tion

Revi

ew, 5

1 (1

991)

, p. 4

15; m

inor

ity�

Seph

ardi

md

Robe

rt O

. Tilm

an, “

Publ

ic S

ervi

ce C

omis

sion

s in

the

Fede

ratio

n of

Mal

aya,

” Jo

urna

l of A

sian

Stu

dies

, 10

(196

1), p

. 194

; min

ority

�no

n-M

alay

se

B. A

. V. S

harm

a an

d K.

M. R

eddy

, Res

erva

tion

Polic

y in

Indi

a(N

ew D

elhi

; Lig

ht a

nd L

ife, 1

982)

; min

ority

�sc

hedu

led

caste

s an

d tri

bes

fD

enni

s O

. Dre

sang

, The

Zam

bia

Civ

il Se

rvic

e(N

airo

bi, 1

975)

; min

ority

�no

n-Za

mbi

ang

P. P

itil,

“Pub

lic P

erso

nnel

Adm

inis

tratio

n,”

in O

. P. D

wev

idi a

nd N

elso

n E.

Pau

lia, e

ds, T

he P

ublic

Ser

vice

of P

apua

New

Gui

nea

(Bor

oko;

Adm

inis

trativ

e C

olle

ge o

f Pap

ua N

ew G

uine

a, 1

986)

, p. 1

68; m

inor

ity�

expa

triat

esh

Sune

tte v

an d

er W

alt,

“Em

ploy

men

t in

the

Publ

ic S

ecto

r of S

outh

Afri

ca,”

SA

IPA

, 27

(199

2), p

. 23

iRa

lph

E. C

row

, “C

onfe

ssio

nalis

m, P

ublic

Adm

inis

tratio

n an

d Ef

ficie

ncy

in L

eban

on,”

in L

eona

rd B

inde

r, ed

., Po

litic

s in

Leb

anon

(New

Yor

k: Jo

hn W

iley,

1966

), p.

172

jR.

E. G

erm

ann,

Adm

inis

tratio

n Pu

bliq

ue e

n Su

isse

(Ber

ne: H

aupt

, 199

6), p

. 155

kBo

la D

auda

, “Fa

llaci

es a

nd D

ilem

mas

: The

The

ory

of R

epre

sent

ativ

e Bu

reau

crac

y w

ith a

Par

ticul

ar R

efer

ence

to th

e N

iger

ian

Publ

ic S

ervi

ce19

50–1

986,

Inte

rnat

iona

l Rev

iew

of A

dmin

istra

tive

Scie

nces

, 56

(199

0), 4

67–9

5; a

ppro

xim

ate

figur

es b

ased

on

state

of o

rigin

may not be perceived as being as unrepresentative as it actually is, simply becausethe clients may deal only with the relatively more representative lower echelons.

The second point about ethnicity and representativeness is that this is fre-quently a point of bargaining in societies attempting to manage severe internalethnic divisions. In some societies, most noticeably Belgium, this has gone to theextent of dividing several ministries by ethnicity (in this case, language) and actu-ally providing two ethnically homogeneous units instead of one integrated unitthat might tend to advantage one group or another. Another variant of the samepattern is the Austrian method of carefully dividing the posts in each ministryaccording to ethnicity or, more specifically in this case, religious or non-religiouspreferences.122 Similarly, the division of posts in the Lebanese administrativesystem among the numerous religious groups in that society was an importantpart of the bargain holding that otherwise tenuous union together prior to 1975.123

Thus there is no necessity for having unrepresentative bureaucracies in ethnicallyplural societies, but the equalization of the service often requires explicit bargain-ing and a recognition of the role of the bureaucracy in institutionalizing ethniccleavage.

The third point is that even when there are active programs to recruitmembers of minority populations, they may not alter the representativeness of thecivil service. Affirmative action programs do not appear to make that much dif-ference in the recruitment of non-whites in the American civil service. Similarly,legal provisions to benefit certain ‘‘scheduled’’ castes and tribes in India that havebeen discriminated against historically have produced very few members in thecivil service, and especially few at the upper echelons.124 As with the representa-tiveness of the civil service by class, educational and other social barriers mayhave to be overcome before legal efforts at greater ethnic representativeness havethe intended effects.

A final point to be made about the ethnic representativeness of the civilservice is that the civil service has served as a means of social advancement forminority groups, in part because of its reliance upon relatively objective criteriafor recruitment. For example, while blacks and Hispanics constitute a rather smallpercentage of the civil service in the United States, they actually comprise ahigher percentage there than in total employment in the economy. This would beespecially true of non-white employment in white-collar and managerial positions,although the relative decline of public sector salaries combined with the increas-ing openness of business to minority employees has made the public sector a lessattractive employer in the United States. Similarly, the civil service has served as ameans of social and economic advancement for Italians from the poorer southernregion; in 1978, 56 percent of the Italian civil service came from the southern partof the peninsula and Sicily, although only 33 percent of the population lived inthese regions.125 This pattern of regional recruitment does not appear to be chang-ing rapidly, even with pressures from northern political groups who increasinglyappear to resent being governed by southerners.126

The European Community represents a special case of a political systemattempting to ensure equality of recruitment in the public service. Communitylaws and regulations call for recruitment of civil service personnel in proportion tothe populations of the member countries, but there are wide disparities in the


1 2 2

extent to which member nationals have been offered and accepted positions in theEuropean bureaucracy (Table 3.10). It is not surprising that there are relativelymore Belgian and Luxembourger employees, given that most EC offices arelocated in Brussels and Luxembourg, but there still is a differential attraction ofsenior positions in the EC bureaucracy. Spain is markedly under-represented at alllevels of the EC bureaucracy, as are the United Kingdom and Germany to a lesserextent. This pattern appears related to the differential attractiveness – in careerand financial terms – for senior civil servants in different countries, as well assome attempt to ensure that all countries have some representation in the upperlevels of the Brussels bureaucracy. The differential recruitment may, however,represent a political problem for countries who believe that they do not receiveenough good jobs for their people and that EC decision makers may be stackedagainst them.

A final dimension of representativeness is gender. The issue of equal treatment ofwomen has become increasingly important in all phases of social and political life,and the public bureaucracy is no different. The issue in the civil service is, ingeneral, not about the total number of women employed; most governmentsalready employ large numbers of women, and in many countries more than half oftotal public employment is female.127 Even in countries that historically haveassigned an unequal role to women, there have been some increases inemployment of women.128 The issue of gender equality centers primarily on thetypes of positions in which women are employed. Women often comprise a large


1 2 3

Table 3.10 Employment in the European Community bureaucracy and population bycountry (percentage)

Upper echelon Total Populationpositions positions

Belgium 7.9 12.0 3.1Denmark 3.6 6.5 1.6France 17.8 15.2 17.2Germany 15.5 10.2 18.9Greece 5.6 6.2 3.1Ireland 3.6. 2.0 1.1Italy 12.5 17.3 11.5Luxembourg 2.3 8.7 0.1Netherlands 5.6 6.3 4.5Portugal 4.3 3.7 3.2Spain 6.3 3.9 12.0United Kingdom 14.9 8.1 17.6

Source: House of Lords Select Committee on the European Communities, 11th Report, Staffing ofCommunity Institutions, HL 66 (London: HMSO, 1986)

Sexual equality

proportion of public sector employment, and over half in some, but they are con-centrated in lower level positions. The majority of women in the civil service ofalmost every country are employed in lower-level jobs, such as clerks and typists,rather than in the higher civil service.

Table 3.11 clearly shows the disparity between the total number of womenemployed in the public sector and the number employed in upper-echelon govern-ment positions. Although, in some cases, over half of total civil service employ-ment is comprised of women, in no case does the higher civil service have morethan 15 percent women. Further, that 15 percent figure is reached only in Norway;the average is only 4.6 percent women in the higher civil service.129 Women havefared somewhat better in subnational governments, with over a quarter of topemployees in some German Lander being women.130 There is, however, someevidence that the proportion of women in senior positions has been increasingrapidly.

As with the cases of class and ethnic representativeness, we must considerwhether these observed employment patterns are the result of overt discrimina-tion or reflect other social, economic or historical factors. It would be difficult todispute that there has been overt discrimination against women in recruitmentto senior posts in government, although the civil service has probably beenmore open than many other occupations. In addition, historically there havebeen relatively few women putting themselves forward for the top governmentpositions, or even having the necessary educational qualifications. As the moreovert discrimination lessens, it will still require time for larger numbers ofwomen to be recruited into entry-level positions for the higher civil service andthen to work their way up the career ladder. Countries that have more opencareer structures, such as the United States, will be able to accommodate tothese changes more rapidly than will those with closed structures requiringyears to work up a ladder.

Noted here are several more general points about the composition and the repre-sentativeness of public bureaucracies. The first is that although these may behighly unrepresentative institutions, they are generally less unrepresentative thanother public elites in the same countries. Parris notes, for example, that in Britainthe membership of the House of Commons is at least as unrepresentative, if notmore so, than the administrative class of the civil service:

If there is an excessive proportion of Oxbridge graduates in the Adminis-trative Class, so is there in the House of Commons. The electorate oughtto be blamed for making the wrong choice just as much as the CivilService Commissioners. If too few civil servants have scientific andtechnological backgrounds, the same criticism can be made of industrialmanagers. Thus, management in both the public and private sectorsreflects the values of the society about what is needed to make organ-izations function well.131


1 2 4

Summary


1 2 5

Table 3.11 Employment of women in the civil service (percentage)

Total in In higherCountry civil service civil service

Australia 39 2Belgium 40 7Canada 47 23Czechoslovakia 67 ?Finland 51 4France 48 13Greece 31 7Guadeloupe 51 ?Guyana 39 ?Hong Kong ? 7Israel 52 3Italy 35 4Netherlands 19 2New Zealand ? 2Norway 47 15Spain 14 ?Sweden 42 5Switzerland 26 2United Kingdom 55 7United States 51 16Germany 47 9

Sources: Charles Debbasch, ed., La fonction publique en Europe (Paris: CNRS, 1981); AndreMolitor, L’Administration de la Belgique (Brussels: Centre de recherche et d’information socio-politique, 1974); Per Laegreid and Johan Olsen, Byråkrati og Beslutningar (Bergen:Universitetsforlaget, 1978); John P. Burns, “The Changing Pattern of BureaucraticRepresentation: The Case of the Hong Kong Civil Service,” Indian Journal of Political Science,27 (1981), pp. 398–429; Royal Commission on Australian Government Administration,Report, appendix 3 (Canberra: Australian Government Publishing Service, 1976); P. K.Kuruvilla, “Public Sector Recruitment in Canada: Some Perspectives and Problems,” IndianJournal of Public Administration, 26 (1980), pp. 62–90; Jacques Bourgault and StéphaneDion, The Changing Profile of Federal Deputy Ministers 1867–1988 (Ottawa: CanadianCenter for Management Development, 1991); Phillipe Clerdieu and Christian Theresine, “Lafonction publique en Guadeloupe – Comparisons avec les autres DOM,” Revue Françaised’Administration Publique, 31 (1984), pp. 89–120; La fonction publique de l’État (Paris: LaDocumentation Française, 1998); Richard Rose et al., Public Employment in Western Nations(Cambridge: Cambidge University Press, 1985); David Nachmias, “Israel’s Bureaucratic Elite:Social Structure and Patronage,” Public Administration Review, 51 (1991), p. 415; RobinWilliams, “The Staffing and Structure of the Public Sector,” in R. M. Alley, ed., State Servantsand the Public in the 1980s (Wellington: New Zealand Institute of Public Administration,1986); US Office of Personal Management, Affirmative Employment Statistics (Washington,DC: USOPM, biennial); Data from Comparative Public Service Project, Department of Politics,University of Strathclyde

The simple point is that elites are unrepresentative by the very function of theirbeing elites. Success in society is related to social background, educationalopportunities and interests, and the elite that a society may select to govern it willdiffer only at the margins in most cases from an elite appointed to govern – at leastin terms of their social and educational backgrounds. The dangers of elitism andunrepresentativeness in public life are general, therefore, and not confined simplyto the public bureaucracy. They are only more apparent in the bureaucracy wherethe emphasis on merit criteria and open recruitment makes it a more ostensiblydemocratic institution in its selection. But, as Max Weber pointed out:

Democracy takes an ambivalent attitude toward the system of examinationsfor expertise. On the one hand the system of examination means, or at leastappears to mean, selection of the qualified from all social strata in place ofrule by the notables. But on the other, democracy fears that examinationsand patents of education will create a privileged ‘‘caste’’ and for that reasonopposes such a system.132

These words should not be taken as an exoneration of bureaucracies for theiroften elitist practices, but rather as a means of placing the problem of representat-ive and unrepresentative bureaucracy in clearer perspective. Further, we shouldremember that most of the studies pointing to the unrepresentativeness of thecivil service are studies of the higher civil service, and that the service as a wholedoes tend to be more representative. As the lower echelons of the service interactdirectly with clients, government may in fact appear more representative andtherefore more democratic than it is.

The second point is that all the furor over social class and ethnic back-ground of administrators, especially top administrators, may be a somewhatmisplaced attack on the institutions. Much analysis has shown that social back-ground tends to have a rather slight effect on behavior in public office. This istrue of legislators, judges and administrators. A more important determinant ofbehavior would appear to be the nature of the organization and the goals of theagency. Again, this may be especially true at upper echelons; there may need tobe greater representativeness at lower levels simply to be able to cope adequatelywith the clientele that an agency may serve. This is not to say that this need besimply a cosmetic gesture on the part of the agency; rather, it is a real need to beeffective in interacting with and serving the clientele. But the more general pointremains that in order to change the policy outcomes from the public bureau-cracy, one may have to do more than simply gradually replace administratorsdrawn from one social class with administrators recruited more broadly fromsociety. The operating routines of agencies, the tendency toward conservatism inorganizations in general, and the process of organizational socialization all tendto reduce the variability of individuals in the organization, regardless of theirsocial background. Thus, changing policy may be a considerably more complextopic, and it is one that will be probed extensively during the remainder of thisvolume.

Public policy does involve a human element. This chapter opened with a dis-cussion of the failure of traditional models of bureaucracy to take into account


1 2 6

human differences and variability. The differences, however, may be as much (ormore) in values, motives and goals than in social background. We touched on thisbriefly when discussing the incentive structures of public bureaucracies, and alsowhen discussing the administrative cultures of society. Thus, studies of recruit-ment need to delve somewhat into the nature of the personnel recruited to admin-istrative careers, to determine not only where they came from but, moreimportant, where they think they (and the society) are going. The values that arerelevant are not just those about the specific programs that the individuals will beadministering, but should also extend to questions of accountability and politicalcontrol of bureaucracy. These values are not necessarily related to education orbackground but may say more about the suitability of the recruit for working ingovernment, especially a democratic government.

1 Max Weber, ‘‘Bureaucracy,’’ in H. H. Gerth and C. Wright Mills, From Max Weber: Essaysin Sociology (New York: Oxford University Press, 1964), pp. 196–244; Frederick W. Taylor,Principles and Methods of Scientific Management (New York: Harper, 1911).

2 Robert D. Putnam, ‘‘The Political Attitudes of Senior Civil Servants in Western Europe,’’British Journal of Political Science, 3 (1973), 275–90; Joel D. Aberbach, Robert D. Putnamand Bert A. Rockman, Bureaucrats and Politicians in Western Democracies (Cambridge,MA: Harvard University Press, 1981).

3 Herbert Kaufman, ‘‘Emerging Conflicts in the Doctrine of Public Administration,’’ Ameri-can Political Science Review, 50 (1956), 1059–73.

4 Even in South Africa, issues of equal employment opportunities and affirmative action inthe public sector have been placed on the agenda. See the special issue of SAIPA (SouthAfrican Institute of Public Administration), 27 (1992).

5 Ari Hoogenboom, Outlawing the Spoils (Urbana, IL; University of Illinois Press,1968); JaneCaplan, ‘‘Profession as Vocation: The German Civil Service,’’ in G. Cocks and K. Jarusch,German Professions, 1800–1950 (Oxford: Oxford University Press, 1990).

6 See Paul C. Light, Thickening Government (Washington, DC: The Brookings Institution,1997).

7 Some other smaller systems, e.g. Finland, have even fewer political appointees than theBritish, to the point that ministers find that they do not have adequate support fordecision-making.

8 Chris Skelcher, The Appointed State: Quasi-Governmental Organizations and Democracy(Buckingham: Open University Press, 1997).

9 See Bernard S. Silberman, Cages of Reason: The Rise of the Rational State in France, Japan,the United States and Great Britain (Chicago: University of Chicago Press, 1993).

10 These political appointees are usually in personal offices or ministerial cabinets. ForFrance, see Monique Dagnaud and Dominique Mehl, ‘‘L’Elite de la Cohabitation,’’ Pou-voirs, 42 (1987), 137–53; see also Axel Murswieck, ‘‘Policy Advice and Decision-making inthe German Federal Bureaucracy,’’ in B. Guy Peters and Anthony Barker, eds, AdvisingWest European Governments (Pittsburgh: University of Pittsburgh Press, 1993).

11 See B. Guy Peters and Jon Pierre, The Politicization of the Civil Service? (forthcoming).12 Skelcher, op. cit.13 J. Donald Kingsley, Representative Bureaucracy (Yellow Springs, Ohio: Antioch University

Press, 1944); Kenneth J. Meier, ‘‘Representative Bureaucracy: An Empirical Assessment,’’American Political Science Review, 69 (1975), 526–42.

14 For example see Ellen E. Pinderhughes, ‘‘The Delivery of Child Welfare Services to


1 2 7

Notes

African American Clients,’’ American Journal of Orthopsychiatry, 61 (1991), 599–605; LauraA. Schmidt, ‘‘Problem Drinkers and the Welfare Bureaucracy,’’ Social Service Review, 64(1990), 390–406; David Burn, ‘‘Ethical Implications of Cross-Cultural Counseling andTraining,’’ Journal of Counseling and Development, 70 (1992), 578–83.

15 T. Edward Kellough, ‘‘Affirmative Action in Government Employment,’’ The Annals, 583(September, 1992), 117–30.

16 B. A. V. Sharma and K. M. Reddy, Reservation Policy in India (New Delhi: Light and Life, 1982).17 The intense political activity that often surrounds personnel issues in the European

Community is one example of this point. See Morten Egeberg, ‘‘Organization and Nation-ality in European Commission Services,’’ Paper presented at ECPR Joint Sessions Bour-deaux, France, April, 1995. These appointments are perceived to be crucial to the influenceof each country in the Community.

18 For the argument for government as a ‘‘model employer,’’ see P. B. Beaumont, Govern-ment as Employer – Setting an Example? (London: Royal Institute of Public Administration,1981).

19 John O’Leary, ‘‘Four-year Study Looks for Bias in Entry to Oxford,’’ The Times 14 August,2000.

20 See, for example, Pierre Bourdieu, La Noblesse d’État (Paris: Minuit, 1989), pp. 101–39.21 See Xiaowei Zang, ‘‘Elite Formation and the Bureaucratic–Technocracy in Post-Mao

China,’’ Studies in Comparative Communism, 24 (1991), 114–23.22 Samuel J. Eldersveld, Political Elites in Modern Societies: Empirical Research and Demo-

cratic Theory (Ann Arbor, MI: University of Michigan Press, 1989).23 Robert D. Putnam, The Comparative Study of Political Elites (Englewood Cliffs, NJ: Pren-

tice Hall, 1967), p. 44.24 See Albert Somit and Stephen A. Peterson, ‘‘Political Socialization of US and Japanese

Adults,’’ Comparative Political Studies, 13 (1980), 3–32.25 This may be in part self-serving. The working class or minority members who are success-

ful can say that they ‘‘made it,’’ and therefore others can as well, if they have the talent anddetermination.

26 See Jack Rabin, ed., ‘‘The Future of Affirmative Action and Equal Employment Opportun-ity: A Symposium,’’ Review of Public Personnel Administration, 4 (1984), 1–82.

27 For a classic statement see Anthony Downs, ‘‘Why the Public Budget is Too Small in aDemocracy,’’ World Politics, 12 (1960), 541–63.

28 One official of the Reagan administration in the United States argued that he only wanted acompetent civil service, so that the ‘‘best and brightest’’ would be in the private sector.Terry W. Culler, ‘‘Most Federal Workers Need Only be Competent,’’ Wall Street Journal,21 May 1986.

29 F. F. Ridley, Specialists and Generalists: A Comparative Study of the Civil Service at Homeand Abroad (London: Routledge and Kegan Paul, 1968).

30 Sir Stafford Northcote and Sir Charles Trevelyan, Report on the Organization of thePermanent Civil Service, reprinted in Committee on the Civil Service Report (the FultonCommittee), Vol. I, Appendix B (London: HMSO, 1968).

31 See John Garrett, Managing the Civil Service (London: Heinemann, 1980).32 Hans-Ulrich Derlien, ‘‘Repercussions of Government Change on the Career Civil Service of

West Germany: The Case of 1969 and 1982,’’ Governance, 1 (1988), 50–78.33 Aberbach, Putnam and Rockman, Bureaucrats and Politicians, p. 52; James W. Fesler,

‘‘The Higher Public Service in Western Europe,’’ in Ralph Clark Chandler, ed., A Centen-nial History of the American Administrative State (New York: Free Press, 1987).

34 Heinrich Neisser, ‘‘Die Rolle des Burokratie,’’ in Heinz Fischer, ed., Das Politische SystemOsterreichs (Vienna: Europaverlag, 1978).

35 For the Senior Executive Service almost one-third of the members are scientists or engi-neers, with another quarter having other professional qualifications. US Office of Person-nel Management, The Fact Book: Federal Civilian Workforce (Washington, DC: OPM,June, 1992).


1 2 8

36 Serge Salon, ‘‘Recrutement et formation,’’ in La Fonction Publique, Vol. 2 (Paris: Lescahiers Français, 197, 1980), 2–7

37 At the extreme, the bureaucracies of Latin America are still argued to have been influ-enced by their colonial experiences over a century and a half ago. See M. Hanson, ‘‘Organ-izational Bureaucracy in Latin America and the Legacy of Spanish Colonialism,’’ Journal ofInter-American Studies, 16 (1974), 199–219. At the other extreme, countries of EasternEurope are only recently left with bureaucracies trained according to the style of theformer Soviet Union. See Jaroslaw Piekalkiewicz and Christopher Hamilton, PublicBureaucracies Between Reform and Resistance (Providence, RI: Berg, 1991).

38 Fred A. Clemente, ‘‘Philippine Bureaucratic Behavior,’’ Philippine Journal of Public Admin-istration, 15 (1971), 119–47.

39 Harry Taylor, ‘‘Public Sector Personnel Management in Three African Countries: CurrentProblems and Possibilities,’’ Public Administration and Development, 12 (1992), 193–208.

40 See John A. Armstrong, The European Administrative Elite (Princeton, NJ: Princeton Uni-versity Press, 1974); Rolf Torstendahl, Bureaucratization in Northwestern Europe,1880–1985: Domination and Governance (London: Routledge, 1991).

41 Jerry Hough and Merle Fainsod, How the Soviet Union is Governed (Cambridge, MA:Harvard University Press, 1979); E. Huskey, Executive Power and Soviet Politics: The Riseand Decline of the Soviet State (Armonk, NY: M. E. Sharpe, 1992).

42 Kenneth Liberthal and Michel Oksenberg, Policymaking in China: Leaders, Structures andProcesses (Princeton, NJ: Princeton University Press, 1988).

43 T. H. Rigby, Political Elites in the USSR: Central Leaders and Local Cadres from Lenin toGorbachev (Aldershot: Edward Elgar, 1990).

44 Jerry F. Hough, The Soviet Prefects (Cambridge, MA: Harvard University Press, 1969),pp. 292–305.

45 Rolf H. W. Theen, ‘‘Party and Bureaucracy,’’ in The Soviet Polity in the Modern Era, Erik P.Hoffman and Robbin F. Laird, eds (New York: Aldine, 1984), pp. 254–65.

46 There are strong legal requirements that prevent blatant use of patronage for positions.Also, the ‘‘Swedish’’ model was beginning to change prior to the election of the right-of-center government. T. Petterson and K. Geyer, Varderingsforandringar i Sverige: Densvenska modellen, individualism och rattvisa (Stockholm: Brevskolan, 1992).

47 Klaus von Beyme and Manfred Schmidt, Policy and Politics in the Federal Republic ofGermany (London: Gower, 1985).

48 Paolo Urio et al., Sociologie politique de la haute administration publique de la Suisse (Paris:Economica, 1989).

49 Wolfgang Pippke, Karrierdeterminaten in der offentlichen Verwaltung (Baden-Baden:Nomos, 1975).

50 Jean-Louis Quermonne, L’Appareil Administratif de l’Etat (Paris: Editions de Seuil, 1991),pp. 185–6.

51 This has been referred to as the ‘‘descent from heaven.’’ See Chalmers Johnson, MITI andthe Japanese Miracle (Stanford, CA: Stanford University Press, 1982), pp. 65ff.

52 See G. Calvin Mackenzie, The Politics of Presidential Appointments (New York: The FreePress, 1981); The In and Outers (Baltimore, MD: Johns Hopkins University Press, 1987).

53 The Act, however, lacks the teeth that critics of the military-industrial complex would likefor it to have. Also, other sectors of the government are not even this closely regulated.See US General Accounting Office, DOD Revolving Door: Processes Have Improved butPost-DOD Employment Reporting Still Low (Washington, DC: USGAO, 1989), ReportGAO/NSIAD 89–211.

54 Kenneth Kernaghan, ‘‘Promoting Public Service Ethics: The Codification Option,’’ inRichard A. Chapman, ed., Ethics in Public Service (Edinburgh: University of EdinburghPress, 1993).

55 This is often less than fully successful. See Harvey Feigenbaum, The Politics of French Oil(Princeton: Princeton University Press, 1986).

56 This career distinctiveness of Swedish civil servants is beginning to change. Attempts to


1 2 9

impose a managerialist conception of its role on the civil service have opened the systemto outsiders, although not to the extent hoped by its advocates.

57 Kieran Walsh and John Stewart, ‘‘Change in the Management of Public Services,’’ PublicAdministration, 70 (1992), 499–518.

58 Some of the managerialism associated with administrative reform in the 1980s attemptedto open the public service to outsiders, especially from business. See Andrew Massey,Managing the Public Sector (Aldershot: Edward Elgar, 1993).

59 Peta E. Sherif, ‘‘Outsiders in a Closed Career: The Example of the British Civil Service,’’Public Administration, 50 (1972), 397–418.

60 See Anne Davies and John Willman, What Next?: Agencies, Departments and the CivilService (London: Institute for Public Policy Research, 1992).

61 D. Soderlind and O. Petersson, Svensk forvaltningspolitik (Uppsala: Diskurs, 1988), p. 169.62 Paolo Urio et al., op. cit., p. 46.63 Hugh Heclo, A Government of Strangers (Washington, DC: The Brookings Institution,

1978).64 As the number of appointees has increased and experience in the civil service has

declined, with many people leaving because of low pay and morale, the continuity of thesystem is becoming increasingly threatened.

65 Jean-Luc Bodiguel, ‘‘A French Style Spoils System,’’ Public Administration, 61 (1983),pp. 295–300; Monique Dagnaud and Dominique Mehl, ‘‘L’elite rose confirmee,’’ Pouvoirs,50 (1989), 149ff.

66 For a detailed description, see B. C. Koh, Japan’s Administrative Elite (Berkeley, CA:University of California Press, 1989), Chapter 8.

67 Hans-Ulrich Derlien, ‘‘Wer macht in Bonn Karriere? Spitzenverbande und ihr beruflicherWerdegang,’’ Die offentliche Verwaltung, 43 (1990), 311–19.

68 This figure includes school teachers and lower level public servants as well as Beamte; oneauthority cites just over one third of the Bundestag as being from the Beamte group.Russell J. Dalton, Politics in Germany, 2nd edn (New York: HarperCollins, 1993),pp. 216–20.

69 Peter B. Clark and James Q. Wilson, ‘‘Incentive Systems: A Theory of Organizations,’’Administrative Science Quarterly, 6 (1962), 129–66.

70 Hood and Peters, op. cit.71 Charles H. Levine, ‘‘The Quiet Crisis of the Civil Service,’’ Governance, 1 (1988), 115–43.72 National Personnel Authority, Handbook on Japan’s Civil Service Statistical Overview

(Tokyo: National Personnel Authority, 1990).73 Anthony Downs, Inside Bureaucracy (Boston: Little, Brown, 1967), pp. 96–101.74 Christopher Hood, ‘‘De-Sir Humphreying the Westminster Model of Bureaucracy,’’ Gover-

nance, 3 (1990), 205–14.75 For a less quantitative view of the incentives, see Direction Generale de l’Administration et

de la Fonction Publique, Les Administrateurs civils dans les services centraux de l’Etat face aleur carriere et a leur travail (Paris: La Documentation Francaise, 1991).

76 This characterization of the French civil service may be less true in the 1990s than it hadbeen in the past. See Jean-Luc Bodiguel and Luc Rouban, Le fonctionnaire detrone? (Paris:Presses de la Fondation Nationale des Sciences Politiques, 1991).

77 The factor of self-advancement is especially important for Italy where the civil service isoften utilized as a route of economic advancement for people from the southern part of thecountry. See Sabino Cassesse, ‘‘The Higher Civil Service in Italy,’’ in Ezra N. Suleiman,ed., Bureaucrats and Public Policy (New York: Holmes and Meier, 1984).

78 Beverly Alban Metcalfe, ‘‘What Motivates Managers: An Investigation by Gender andSector of Employment,’’ Public Administration, 67 (1989), 95–108.

79 Jak Jabes and David Zussman, ‘‘Motivation, Rewards and Satisfaction in the CanadianFederal Public Service,’’ Canadian Public Administration, 31 (1988), 204–25.

80 Leslie L. Roos and Noralou P. Roos, Managers of Modernization: Organizations and Elitesin Turkey (Cambridge, MA: Harvard University Press, 1971).


1 3 0

81 Wen-Fang Tang, William Parish and Guansan Yang, ‘‘Bureaucracy in China,’’ unpublishedpaper, Department of Political Science, University of Pittsburgh, 1993.

82 Ben Ross Schneider, Politics Within the State: Elite Bureaucrats and Industrial Policy inAuthoritarian Brazil (Pittsburgh: University of Pittsburgh Press, 1991); Fernando Uri-coechea, Estado y burocracia en Colombia (Bogota: Universidad Nacional de Colombia,1986).

83 M. A. H. Wallis, Bureaucracy: Its Role in Third World Development (London: Macmillan,1989).

84 Robert N. Kearney and Richard L. Harris, ‘‘Bureaucracy and Environment in Ceylon,’’Journal of Commonwealth Political Studies, 2 (1964), 254–5.

85 Ibid., p. 255.86 John W. Thomas and Merilee S. Grindle, ‘‘After the Decision: Implementing Policy

Reforms in Developing Countries,’’ World Development, 18 (1990), 1163–81.87 See Patricia W. Ingraham, ‘‘Building Bridges or Burning Them?: The President, the

Appointees and the Bureaucracy,’’ Public Administration Review, 47 (1987), 425–35.88 M. S. Voslenski, Nomenklatura: The Soviet Ruling Class (Garden City, NY: Doubleday,

1984); Bruno Rizzi, The Bureaucratization of the World (New York: Free Press, 1985).89 Michel Crozier, The Bureaucratic Phenomenon (Chicago: University of Chicago Press,

1964), pp. 228–9.90 Fortunately or unfortunately the recent politicization of issues of public sector pay has

made the rewards of political and administrative officials in government more apparent tothe public. See Christopher Hood and B. Guy Peters, eds, The Rewards of High PublicOffice (London: Sage, 1993).

91 Even here, however, a public sector employee may have special obligations and duties. Asecretary in a government office may type numerous letters of the utmost political sensitiv-ity and is assumed to be able to exercise the appropriate discretion in not recounting toothers what she or he has seen.

92 Peter M. Benda and Charles H. Levine, ‘‘Reagan and the Bureaucracy: The Bequest, thePromise, the Legacy,’’ in Charles O. Jones, ed., The Reagan Legacy (Chatham, NJ:Chatham House, 1988).

93 Siv Gustaffson, Lonebildning och lonestruktyr inom den statliga sektorn (Stockholm:Almqvist and Wicksell, 1972); INSEE, ‘‘Remunerations des Agents de l’Etat en 1987 et1988,’’ INSEE Premiere, 50 (1989), 1–4.

94 One Swedish civil servant is reported as having said in a meeting with private sector coun-terparts that he expected their comments to be twice as good as his, given that they werepaid twice as much. See Barbara Czarniawska, ‘‘The Ugly Sister: On the Relationship ofthe Private and the Public in Sweden,’’ Scandinavian Journal of Management Studies,(1985), p. 93.

95 Review Body on Top Salaries, Report No. 33, Cmnd. 2015 (London: HMSO, 1992) reporteda three percent drop in the real earnings of senior civil servants, in contrast to a 41 percentincrease of income for top managers in the private sector in the period of 1985 to 1992.

96 Christopher Hood, ‘‘Rewards at the Top,’’ in Hood and Peters, Rewards of High Public Office.97 Peter S. Heller and Alan A. Tait, Government Employment and Pay: Some International

Comparisons (Washington, DC: International Monetary Fund, 1983).98 J. L. Fallick and R. F. Elliott, Incomes Policies, Inflation and Relative Pay (London: George

Allen and Unwin, 1981).99 Renk Robough and Wendy Kooistra, ‘‘Rewards in the Public Sector: The Case of the

Netherlands,’’ Paper presented at Conference on the Rewards of Higher Public Office,Erasmus University/Riksuniversiteit Leiden, Februrary, 1993.

100 Desmond S. King and B. Guy Peters, ‘‘Rewards of Higher Public Office: The UnitedStates,’’ in Hood and Peters, The Rewards of Higher Public Office.

101 Hans Ulrich Derlien, ‘‘The Structure and Dynamics of the Reward System for GermanBureaucratic and Political Elites,’’ in Christopher Hood and B. Guy Peters, The Rewards ofHigher Office (London: Sage, 1993).


1 3 1

102 This is true even in countries such as Sweden, Norway and the Netherlands with historiesof low differentiation of wages in the public sector and a ‘‘solidaristic’’ wage policy in theeconomy as a whole. Per Laegreid, ‘‘Lonspolitsike reformforsok i staten,’’ LOS SenteretNotar No. 52 (Bergen: LOS Center, 1989); Lennart Lundquist, Ambetsman eller direktor(Stockholm: Norstedts, 1993); Robough and Kooistra, ‘‘Rewards in the Public Sector: TheCase of the Netherlands,’’ in Christopher Hood and B. Guy Peters, Rewards of High PublicOffice (London: Sage, 1994).

103 Luc Rouban, ‘‘Rewards of High Public Offices in France: Political Arguments and Institu-tional Changes,’’ in Christopher Hood and B. Guy Peters, The Rewards of High PublicOffice (London: Sage, 1994); Marlene Brans, ‘‘Public Office and Private Rewards: Rewardsfor High Public Office in Belgium,’’ in Hood and Peters.

104 On Sweden, see Maivor Sjolund, Statens Lonepolitik 1966–1988 (Stockholm: Publica, 1988).105 See James L. Perry, ‘‘Merit Pay in the Public Sector: The Case for a Failure of Theory,’’

Review of Public Personnel Administration, 7 (1986), 57–69.106 International Monetary Fund, Government Finance Statistics Yearbook, Vol. 14 (Washing-

ton, DC: International Monetary Fund, 1990). Much of the total expenditure of the govern-ment of any developed country is transfer expenditures such as pensions and other socialbenefits. These tend to far exceed civil service costs.

107 Ibid. In part, government may become an employer of last resort in order to prevent highlevels of unemployed people in urban areas. Also, overt corruption and putting friends andsupporters on the public payroll accounts for some of these inflated figures.

108 Christopher Hood and B. Guy Peters, The Rewards of High Public Office (London: Sage,1994).

109 Jean-Luc Bodiguel, ‘‘Nouveaux concours, nouveaux enarques,’’ La Revue Administrative,186 (1978), 610–18. Also, there has been some increase in the number of direct entrants,even into the grands corps such as the Cours de Comptes.

110 Wolfgang Zapf, Wandlungen der Deutschen Elite (Munich: Piper, 1966), pp. 180–2. Thelater study by Derlien and Mayntz found only five percent of senior civil servants with aworking-class background. Hans-Ulrich Derlien and Renate Mayntz, Einstellungen der poli-tisch-administrativen Elite des Bundes 1987 (Bamberg: Universitat Bamberg, Lehrstuhlefür Verwaltungswissenschaft, 1988).

111 Joel D. Aberbach, Robert D. Putnam and Bert A. Rockman, Bureaucrats and Politicians inWestern Democracies (Cambridge, MA: Harvard University Press, 1981). See also Bodigueland Rouban, Le fonctionnaire detrone, pp. 104–10.

112 This appears to be true even though American higher education is fee-paying while it isfree in many European countries.

113 Aberbach, Putnam and Rockman, Bureaucrats and Politicians, pp. 48–9.114 Gavin Drewry and Tony Butcher, The Civil Service Today, 2nd edn (Oxford: Blackwells,

1991); Nicholas Holgate, ‘‘Is Fifty-Nine Percent Enough?,’’ FDA News (June, 1990), 2115 V. Subramanian, The Social Background of India’s Administrators (New Delhi: Ministry of

Information and Broadcasting, 1971), p. 39.116 Pierre Bourdieu, La Noblesse de L’Etat (Paris: de Minuit, 1989).117 Koh, Japan’s Administrative Elite, pp. 86–94.118 Pan S. Kim, ‘‘Who Serves the People: Educational Backgrounds of South Korean and

Japanese Bureaucrats,’’ unpublished paper, Old Dominion University, Department ofPublic Administration, 1990; George Vernardakia and Catherine D. Papastathopoulos,‘‘The Higher Civil Service in Greece,’’ International Review of Administrative Sciences, 55(1989), 603–29.

119 W. Lloyd Warner, The American Federal Executive (New Haven: Yale University Press,1963), p. 372.

120 Critics would argue that this state domination is one of the major barriers to socio-eco-nomic development in these countries.

121 As noted above, representativeness may be particularly important at this level of the publicsector. See M. Adler and S. Asquith, Discretion and Power (London: Heinemann, 1981).


1 3 2

122 This practice is becoming less established. See Barbara Liegl and Wolfgang C. Mueller,‘‘Senior Officials in Austria,’’ in Edward C. Page and Vincent Wright, eds, BureaucraticElites in Western European States (Oxford: Oxford University Press, 1999).

123 Ralph E. Crow, ‘‘Confessionalism, Public Administration and Efficiency in Lebanon,’’ inLeonard Binder, ed., Politics in Lebanon (New York: John Wiley, 1966), pp. 71ff.

124 Sharma and Reddy, Reservation Policy in India.125 Franco Ferraresi, Burocrazia e Politica in Italia (Milan: Il Mulino, 1980), 114–17.126 Sabino Cassese, ‘‘Italy’s Senior Civil Service,’’ in Page and Wright, op. cit.127 See the evidence presented in Richard Rose et al., Public Employment in Western Nations

(Cambridge: Cambridge University Press, 1985).128 Monirah Rawaf, ‘‘The Changing Status of Women in Management in the Public Adminis-

tration of Saudi Arabia,’’ Public Administration and Development, 10 (1990), 209–20.129 Several of the individual Lander in Germany have achieved over 15 percent employment of

women in the senior civil service. See This Week in Germany, September 10 (1993), 7.130 ‘‘Higher-Echelon Positions Remain Largely a Male Domain in the German Bureaucracy,

Though Some Improvement May Be Seen,’’ This Week in Germany, September 10(1993), 7.

131 Henry Parris, Constitutional Bureaucracy (London: Allen and Unwin, 1969), p. 315.132 Gerth and Mills, From Max Weber, p. 240.


1 3 3

Peters (Ch.3) - The Recruitment of Public Administrators

Documents

public servants

public organizations

public service

public program

purposes public

public sector

group of public administrators

public decision makers