A New Approach to Developing Institutional Structure ErnestoE Betancourt

This article describes an approach developed to assess the institutional capacity of whole governments. It discusses the evolving context in Latin America and the Caribbean for public sector reforms and the impact of structural adjustment and globalization on international agency policies on the issue. It reviews national and sectoral organizational networks, the profile of a modem public administration, and the definition of public management horizontal and vertical systems; explains how integration of horizontal management system reforms is attained at the ministry level, and addresses the issue of performance management and use of an elite managerial and technical corps. Finally, it describes techniques used to plan changes in strategies.

ver the years, the development community has treated institutional development as a subsidiary factor. Macroinstitutional development in this article relates to the organizational capacity required to ensure adequate management of national development implementation and not to its meaning from the perspective of the new field of "institutional economics." Economics and economists have dominated development theory and praxis due to the emphasis given to financial and monetary aspects of development that, quite correctly, must be given priority. Although this emphasis has been expanded to cover social and environmental aspects of development, it continues to neglect macroinstitutional development issues, which are equally important aspects of national underdevelopment. Lack of theoretical macro approaches to institutional

O

Emesto F. Betancourt is Vice-President of International Development and Finance (IDF), in charge of institutional development and public sector reform. He has worked in most countries in Latin America and the Caribbean as a consultant for more than thirty years, and is the author of many essays, articles and reports on Cuban and Latin American politics and public administration. He is also the author of Revolutionary Strategy: A Handbook for Practitioners, a textbook on the analysis of revolutionary propensity, 9 1991 by Transaction Publishers. Address for correspondence is 1815 H Street, N.W., Suite 1100, Washington, D.C. 20006.

Studies in Comparative International Development, Summer 1997, Vol. 32, no. 2, 3-28

4

Studies in ComparativeInternationalDevdopment / Summer 1997

development is another reason for this situation. As a result, public sector reform has not been given the required priority to ensure program or project implementation. Macroeconomic theory can be reduced to formal logic constructs or mathematical formulations. Its variables can be measured in currencies to represent their values or magnitudes, thus allowing mathematical manipulation, as well as cross-time and cross-country comparisons by using price indexes or exchange rates to adjust them. It also offers elegant algorithms to replicate or simulate economic performance. However, as in any effort to represent reality in abstract, the algorithms include only those factors that can be measured and, thus, distort reality. Organization theory, on the other hand, involves many diverse nonquantitative elements. These various elements cannot be reduced to one single unit of measurement. Its praxis deals with discrete and/or diffuse factors----objectives, strategies, structures, people and their inter-relationships, which are hard to reduce to elegant formulas or theoretical constructs for one institution or sector and much less for macroanalysis. As a practitioner, there is little one can do to overcome the inadequacy of organization theory to provide an instrument comparable to economic theory for macroinstitutional policy formulation, although some efforts on the use of the systems approach have been very useful (Katz 1965). To deal with the reality of development one must go through two stages. The first stage addresses imbalances at the macroeconomic level and corrects sectoral situations, which falls in the policy realm. The second stage, which is more complex and time consuming, involves managing implementation not only of macroeconomic policy but of other development policies at the sectoral level. This requires a long term commitment to expand macroinstitutional capacity. This article will explain an approach to macroinstitutional reform developed for some UNDP Management Development Program (MDP) missions, involving overall diagnosis of national public administrations, and some World Bank public sector reform projects. Both experiences are limited to the Latin American and Caribbean regions. The Context for Public Sector Reform In the early period of post-World War II development efforts in Latin America and later, under the Alliance for Progress, the institutional constraints resulting from the introduction of development planning and the expansion of the role of the state into direct economic management to substitute imports were addressed through the creation of ad-hoc institutions. Development banks and specialized agencies or institutes were created to handle specific investment efforts and development functions. There were also some partial efforts at modernizing budgetary, procurement, and civil service systems (OAS 1967). The use of project units or autonomous agencies outside the traditional government ministries, as well as public enterprises, became standard institutional instruments for state action (Moore 1992). Those policies led to a further weakening of traditional government ministries and

Betancourt

5

the creation of a parallel public administration with privileged working conditions, which, in many cases, was more responsive to the guidance of external donors than to national political authorities. Public enterprises attempted to fill in for the weakness of national private enterprises, but, in most cases, became inefficient drains on national treasuries. This dual institutional system undermined accountability to legitimate authority, weakened governance, and generated jealousy from the traditional bureaucracies. Furthermore, it gained a life of its own within national bureaucracies eager to enjoy the better working conditions and diluted accountability they offered. In a 1992 report on Jamaica's government structure (Government of Jamaica 1992), for example, the Nettleford Committee of Advisors, appointed by the Prime Minister, found there were more than 300 statutory bodies and/or public enterprises, one for every 10,000 Jamaicans. Neither the Ministry of Finance nor sectoral ministries were able to provide the committee with details on the financial and manpower resources available to these statutory bodies. In situations like this--the findings in four UNDP MDPs missions in Latin America are similar--the traditional ministerial hardcore of government is left with an increasingly underpaid civil service, unable to attract and retain competent managers and technicians. In addition, due to high centralization of decision making inherited from authoritarian or colonial administrative traditions, management cadres of the civil service were not delegated authority commensurate with their responsibility to manage the resources required to attain the goals and objectives of their respective entities. Therefore, there is a reluctance on their part to accept notions of accountability, efficient service to the people, transparency, and other desirable aspects of modern governance. In many countries of the region, public service is reduced to a form of unemployment compensation for political clienteles, who, in turn, become highly unionized and activist in protecting and expanding their benefits. With the introduction of structural adjustment lending, the situation started to change. The need to improve macroeconomic policy management led to efforts to improve the agencies involved. This has been limited to improved central bank, debt, budgetary and tax management, with increasing attention to the introduction of modem information systems and accounting practices (Lamb 1987). Creation of project units came to be seen as undesirable as a result of the experience that led to government fragmentation and loss of continuity when the project ended, rather than of deliberate structural adjustment policies. The logical institutional consequence of a reduced state role in terms of the functional retrenchment structural adjustment requires, and its impact on the overall structure of the state, initially received scant and certainly not systematic attention. Contemporary economic theory has addressed the issue of the role of the state, but exclusively from the economic perspective (Stiglitz 1989). Some research on the institutional dimension of structural adjustment efforts has been undertaken under World Bank sponsorship (Samuel 1990, Nunberg 1990). The impact on governance of reducing the dimension and role of the state in line with stabilization and structural adjustments programs has received limited atten-

6

Studies in Cemparative International Development / Summer 1997

tion. The impact on poverty has been recognized (Ribe, Carvalho, et al. 1990), and poverty alleviation loans have been provided alongside structural adjustment loans. But the emergency delivery mechanisms used for poverty alleviation are becoming permanent entities in some countries and compound government fragmentation. Despite donor efforts to ensure coordination with existing ministries, duplication of efforts is frequently present. Not much attention has been given to how to improve public management in general to ensure that the substantial percentage of GDP allotted to public expenditures is used effectively and efficiently to provide timely and courteous service to society. The emphasis has been in reducing public expenditures, not in improved management. This is not only an issue of accountability, but it is increasingly important if governments are to provide the infrastructure and support required by the national private sector to compete successfully in world markets. Globalization is one of the basic tenets of structural adjustment in Latin America and the Caribbean. As the experience in Bolivia, Mrxico, and Ecuador shows, among others, privatization and stabilization policies could lead to substantial political turmoil and an initial loss of legitimacy for the government. When combined with arbitrary actions or lack of responsiveness to public opinion, at times they result in violent popular explosions. Governance can be improved by increasing the ability of governments to provide better services to the population through public sector reforms. The impact of civil service payrolls on national fiscal deficits led, through a backward linkage, to focus attention on issues of redundancy, pay, and employment (Nunberg 1990). But, dealing with the problem of excessive manpower in isolation may have unintended consequences. It can cause the loss of scarce competent officials and the dismantling of essential functions. International financing of buyout policies to offset the social and political cost of reducing the dimension of the state is a financially sound proposition in itself. The stream of savings of an abolished post generates more than enough savings to service a loan, even with retraining and other assistance costs added up front to a generous severance pay arrangement. It should be linked to overall manpower or human resources management improvements, and should ensure that posts are abolished and that those personnel bought out are not recruited again. In some cases, structural adjustment conditionality has included across-the-board pay increases to improve government competitiveness in the labor market. Such efforts, isolated from a coherent human resources policy, may be wasteful and even counterproductive. Paying more for a staff member who was recruited on a nonmerit basis does not ensure improved manpower quality and tends to freeze the less competent in public employment. Training of cadres of managers and technical staff is usually limited to very narrow technical goals within a sectoral or project loan. Attempts to change organizations through changes in their organizational structures and procedures alone is unworkable. Qualitative changes in the human factor, more than in structure, are essential to change the behavior of organizations.

Betaacoart

7

Under structural adjustment lending, the existing structure of public administration needs a massive reorganization to adapt it to the new residual role of the state that results. This new role should be determined not only by economic criteria, but also by historical, social, and political factors. A modern administration consistent with better governance goals requires decentralization of decision-making within the central government and delegation to local instances to make it more responsive to citizens' aspirations and a government structure that provides the framework to encourage sound management, and an organizational culture based on performance and accountability (Price 1991). In addition, the transition to democratic rule in the region has resulted in a substantial shift of political power to the national legislatures. This requires an executive branch sophisticated in its dealing with the legislative branch, as well as in mobilizing public opinion in support of its program and legislation. Few presidential offices seem to have organized themselves effectively to deal with this new persuasive and negotiating role of the executive. The attitude is frequently the same that prevailed under the more authoritarian style of presidential rule. In modern public management, a single agency can seldom manage all aspects of national policy formulation and implementation to address a national goal. Organizational networks are essential means of attaining national goals. Unfortunately, their coordination is increasingly left to the creation of commissions upon commissions. The efforts made to expand governmental capacity to manage the increased coordination burden have been either nonexistent or inadequate. The vertiginous rate of change in information technology is seeping into multiple aspects of public sector management in most countries of the region. While the informatics revolution is a factor for change that offers tremendous possibilities to improve public management, used in isolation from other required changes, it may be dysfunctional and wasteful: first, because it may lead to efforts to do more efficiently what is no longer necessary; second, because it is difficult to implement when civil service rules do not allow the government to attract and retain the highly trained staff it requires; and, third, because it is seldom complemented with proper changes in organizational structure. Finally, the problem of government corruption, which is so central to efficient public management and governance, has become a very powerful political issue throughout the region. The existence of an increasingly free press has led to exposure of scandals resulting in the removal and prosecution of presidents in Brazil, Venezuela, and Peril. In Ecuador, it forced the resignation of a Vice-President, and, in Colombia and Mexico, it has resulted in a severe loss of legitimacy for the incumbent administrations. As a result, the Special Inter-American Conference on corruption, convoked early in 1996 at the request of the President of Venezuela, adopted a draft Convention to make it more difficult for corruption to prevail and, when it happens, more difficult for those involved to escape prosecution and enjoy ill-gained wealth outside their countries. To accomplish this, it will not be enough to resort to national and regional punitive measures, it will also require a massive regional effort to reform public administrations as a preventive measure. This

8

Studiesin ComparativeInternationalDevelopment/ Summer1997

involves changes in government structures, procurement procedures, financial controis, and civil service legislation (OAS 1996). All these trends and experiences exert pressures that reinforce the need for a comprehensive institutional development policy. In fact, they reveal that an efficient and effective institutional infrastructure capable of well-coordinated action is as essential to development as the transportation, power, or communications infrastructure. Furthermore, governments are systems in which changes in one subsystem require corresponding changes in the remaining interacting subsystems. At times, the point is raised that to attempt comprehensive institutional changes or reforms exceeds the boundaries of the feasible and therefore is unworkable (Esman 1991). The alternative offered is to concentrate on isolated elements of government at functional or sectoral levels. As already noted, such an approach, rather than generating a demonstration effect, has weakened accountability to elected authorities and fragmented government efforts. It is an accepted axiom that changes in strategies require changes in the structures serving them. Structural adjustment policies that encompass changes in the totality of the role of the state within a society are impossible to implement efficiently and effectively unless parallel changes are made at the macroinstitutional level of society. Recognizing that the task is awesome should not deter us from attempting to accomplish what is needed.

Approach Developed for Macroinstitutional Reform This approach to macroinstitutional reform was not developed in one single sweep, but is the cumulative result of efforts undertaken under UNDP Management Development Program missions as well as several World Bank public sector reform projects. In this article, a brief description of the nature of individual components of the overall approach is provided, with the caveat that not all components have been used in every single instance and that some have been developed as stand-alone elements of those projects. However, the way reform practice is evolving, more and more of these components are being incorporated into what, in fact, is becoming a fully comprehensive approach to macroinstitutional reform. Before moving to the description of each component, it is worthwhile to point out that experience has demonstrated that successful implementation of macroinstitutional reform efforts requires: 1. Clear national commitment to public sector reform through creation of a policy coordinating body with direct access to the president of the country, and representing a broad body of stakeholders in the functioning of public administration, supported by a central coordinating staff to ensure continuity of effort; 2. Definition of the role of the state for the particular country through a broad consultative process that allows the incorporation of the points of view of all sectors of society, rather than only those of the incumbent administration or international consultants, since following economic criteria exclusively, instead of adapting the required changes taking into account historical, social, and political forces in the society, could lead to social conflict and eventual rejection; 3. Establishment of an administrative and technical service corps based on personal rank with

Betancourt

4. 5. 6.

7.

9

adequate working conditions to compete in the local manpower market, to attract competent personnel from the private sector, the existing civil service, autonomous agencies, and internationally financed project units, along with the introduction of performance-based budget and management; Enactmentof buyout schemes to make attractive to redundantcivil servants their transfer back to private activities and the promotion,if necessary,of external financingfor these schemes; Providing substantial resources for a massive training effort at the national level and overseas, depending on available national facilities, to improve managerial and technical capabilities leading to a new governmentalorganizationalculture; Decentralizedreform efforts involvingparallel reform projects at the horizontal public management systems and the sectoral or ministeriallevels, which could be financedby various donors, subject only to broad coordination by the central staff group, under the policy of centralized norm setting and decentralized execution, to ensure consistency with whatever comprehensive macroinstitutional policy has been approved by the government to attain the desired modern state architecture; The use of modern change techniques to develop strategies for the introductionof institutional innovations both at the macro and the sectoral or horizontal systemlevel.

Needless to say, such a comprehensive effort will require the setting of priorities for implementation taking into account areas of action identified by national authorities as requiring urgent attention, as well as the indispensable precedent relationships between changes in horizontal systems and in the sectoral or ministerial entities. To manage such complex project implementation, it is necessary to resort to modem computerized project management systems based on the PERT technique.

Methodolo~for OrganizationalNetwork Institutional CapacityEvaluation The approach followed to undertake the diagnostic stage has evolved as the result of four MDP missions in Guatemala, Nicaragua, El Salvador, Ecuador, and one NATCAP mission in Jamaica. Figure 1 provides a visual portrayal of the steps followed in the methodology that was developed over the years for these,undertakings. The steps that are made explicit in the graph were implicit in the logic followed in designing those missions, rather than their having been followed in an identical pattern in all cases with an a priori design of the methodology. The approach was initially used for sectoral assignments, as will be shown, for tourism, export development, regional or financial system diagnoses. The basic approach was to assess the organizational networks involved in each sector, and it drew on the work of Evan (1976) and others on organizational sets. Afterwards, it was extended to the assessment of the totality of public administration by considering it as a master organizational set or network. There are three phases in the process described in the graph. During the first one, covering steps 1 through 9, the mission, which is multidisciplinary in its composition, sharpens the scope of the mandate received and defines the functional categories that result from the national legislation on the organization of the state, which, in most cases, states the roles assigned to the various ministries. Then, taking into account limitations imposed by available time and resources, as well as information, the mission determines the depth of analysis that is feasible in relation to functional scope, spatial distribution, outputs or products and services, and structures.

10

Studies in ComparativeInternational Development / Summer 1997

~,.o

,= d'"

,,~

.o

0

9o

Q,,

~__~

I

~

0

.,.,=

~.-o

--

o

~_o

o.~

":-_

.=

e~ m

~

~'o~

~s

0

o ,,,-~ o

.~.~

~

.~.-~ ~,

~

.~ ~

"~

-

~.~.~

~-~-~ -~ ~ i ~~ ~

.=~

~ ~ --

~ ~ ~

=~

=,. ~ . ~

~-~.~ _

o~.~ o o

~.~

~

~.

~

~ ,,, . - . _

.

-.-

~._ ~

~ ~,"~ ~ ,_, ~.~

~

o ~

.,

o'~.~- ~ o o

~>_ ~ ~

= ,,, .,=

Betancourt

Il

Dependence on secondary sources requires validation of the information obtained, particularly when it involves criticisms of individuals or agencies. Failure to do so can make the mission a tool of petty bureaucratic revenges and personal grudges. The mission is kept in a total immersion mode during the field data gathering phase to encourage exchange among its members of overall views and perceptions, as well as cross-over specific data inputs. Although surveys of managerial cultures and client satisfaction would provide additional inputs and allow more data gathering relevant to specific mission mandates, budgetary limitations have not allowed use of such tools in the work undertaken so far. In the case of Nicaragua, the Economic Development Institute incorporated client surveys into the diagnostic phase. At this point, the mission moves to the next stage of information analysis and evaluation. This is done through a process of intensive interaction among mission specialists for their sector and relates that to overall criteria that emerge on what may be considered an implicit ideal model of the overall public administration. A more or less detailed profile of the institutional situation is developed either at the sectoral or national level. At this stage, heavy weight is given to the judgments made by national officials familiar with the situation, since experience demonstrates that, in most cases, diagnoses of these officials are validated by the analysis of the data gathered. Once individual mission members prepare the first draft, and the mission leader integrates it into a coherent draft report, the report is submitted to its national counterpart for review and disposition. Whenever possible, before the data gathering mission departs from the country, an effort is made to test hypotheses and tentative recommendations that may have emerged during the field work. Experience indicates that when this is done, the final report is more consistent with national realities.

Profile of a Modern PublicAdministration One of the key judgments usually made in an overall diagnosis of a national public administration relates to the profile of the desired state against which the present administration is being evaluated. This is not necessarily made in an explicit statement, but it is certainly implicit in the criticisms made of the existing situation. Furthermore, criteria used to make such judgments change as the state of the art of public administration evolves. An attempt will be made to define the profile that influences such value judgments. The profile of the desired final state of the public administration usually includes criteria related to the legal context under which the administration functions, the structural aspects of its organization, the decision-making location that prevails in relation to various aspects of its management, and the style of management that prevails. Legal context. In relation to this criterion, there are two specific points to consider: 9 The degree to which there is a clear division of managerialjurisdiction between the executive and the legislative branches of government. When there is an excessive amount of legislative

12

Studies in ComparativeInternational Devdopment / Summer 1997

involvement at the operating level, it is unlikely that good management could prevail. Good management requires a clear cut delegation of power to the executive. This does not mean necessarily authoritarian rule, because through participative approaches, decentralization, and adequate performance evaluation, the legislative branch can ensure accountability of the executive branch consistent with good governance. The degree to which presidential office organization provides adequate staff support for the chief executive to discharge the managerial functions vested in his office at the macro level, to allow him to exert effective management and coordination of the public administration, while relating effectively with the increasingly assertive legislatures and media in the new environment created by the decline of "presidentialism."

O r g a n i z a t i o n structure aspects. I n r e l a t i o n to t h i s c r i t e r i o n , t h e r e a r e f i v e p o i n t s to consider: 9 Whether overall functional coverage is adequate to the mandates emanating from the new role of the state that has been chosen by the country, in particular, to sustain and reinforce democratic rule and a market driven economy capable of competing in world markets. 9 Whether spatial coverage of state responsibilities is adequate to provide the desired services wherever they are needed in the country to ensure prompt, effective, and efficient services to the citizens. 9 The degree of fragmentation in the distribution of functions throughout the public administration. The judgment is that excessive fragmentation is detrimental to good public sector management, although it recognizes that specialization or the convenience of redundancy to encourage competition may be factors influencing the existence of such fragmentation. 9 The degree of state interventionism in operating the economy. Although prevailing economic thought is that the state should not distort market operations through interventions that affect prices and the allocation of resources, the political, social, and historical reality in the countries, at times, makes this a relative rather than an absolute. Nevertheless, for purposes of rating for the profile, a high degree of state intervention is considered negative. 9 The degree of regulatory obstacles and procedural delays that may be manifested by too many regulations or unacceptable delays in providing the permits or documentation required by the clienteles involved. 9 The degree to which the governmental accounting system allows for integrated financial management ensuring that there is need for only one entry per transaction and that the transaction code allows for generation of the reports necessary not only for macroeconomic management but also for performance management and auditing or financial control.

Decision-making

location. In r e l a t i o n to this g r o u p o f e l e m e n t s o f e v a l u a t i o n

c r i t e r i a , the g e n e r a l p r i n c i p l e f o l l o w e d is that, in m o d e m a d m i n i s t r a t i o n , p o l i c y a n d n o r m a t i v e d e c i s i o n s a r e m a d e at t h e c e n t e r , a n d o p e r a t i n g a u t h o r i t y s h o u l d b e d e l e g a t e d as c l o s e as p o s s i b l e to t h e p o i n t o f s e r v i c e d e l i v e r y to the c l i e n t b e i n g s e r v e d . T h e r e a r e six p o i n t s to c o n s i d e r t h e d e g r e e o f d e l e g a t i o n o f a u t h o r i t y m a d e to l i n e managers: 9 Proposing programming decisions at the time of budget formulation, with final approval to be retained at the level of agency management. 9 Operating decisions in relation to execution of the program, once it is approved by the legislature. 9 Personnel decisions in relation to the selection, evaluation, promotion, and termination of staff under their supervision, to ensure a sense of responsibility for the management of human resources within civil service rules oriented to attract and retain staff of the best quality, who should be selected on the basis of merit, rather than political influence. 9 Disbursement decisions, once the financial programming for the year is made, to ensure the necessary cash flow to permit effective and efficient program implementation. Performance budgeting is moot when treasury cash management does not allow a smooth cash flow consistent with its implementation.

Betancourt

13

9 Procurement decisions to acquire the material and equipment inputs required for prompt program implementation to avoid unnecessary delays. 9 Auditing control decisions, which should not be made a priori by the national controller office but a posteriori, since the old a priori discretionary control practice is recognized as one of the biggest sources of bureaucratic delays and corruption. Style o f management. In relation to this group of elements o f the evaluation criteria, there are four points to consider: 9 The degree of use of informational technology, since a modem management system is not feasible today without taking advantage of the capacity for data management and analysis provided by the computer for substantive as well as administrative purposes. 9 Whether the budgeting system offers adequate means to relate inputs to program outputs, thus allowing for performance management. 9 Whether adequate performance indicators have been established to allow measurement of the progress in accomplishing program goals in the evaluation phases of budget implementation. 9 Whether an adequate performance pay system is in place to reward staffers in accordance to their contribution to attainment of program goals, particularly among the management and technical cadres. The judgments for each criterion involved are kept at a certain level o f generality, because in a macroinstitutional evaluation to get down to excessive detail could result in an overwhelming amount o f data. The main purpose o f the profile is to provide a general map o f the situation and an assessment o f the feasible success o f individual changes. The profile provides the basis for the formulation o f a global strategy and plan o f action to r e c o m m e n d to the national authorities. It is at that point in the process that more detailed specialized diagnoses are undertaken. In table l, the above criteria are applied to the five countries where such judgments were made. It must be pointed out that, at the time o f the assignments involved, no formal methodology, such as the one suggested in this article, was explicitly stated and that the criteria emerged from making explicit what was only implicit. Also, criterion inclusion has expanded over the years and, for that reason, in some cases no judgments can be made retrospectively. It is evident from the resulting profiles that it will take a long time and persistent effort for these countries to o v e r c o m e administrative styles and cultures that have evolved over centuries and to attain the m o d e r n public sector m a n a g e m e n t they need to compete in t o d a y ' s internationally integrated e c o n o m y . Public sector reforms u n d e r w a y at present with W o r l d Bank support in several countries are helping improve their public sector management. But it must be pointed out that it will easily take a decade o f persistent and coherent effort for these countries to attain the desired macroinstitutional infrastructure. This means that they will have to survive at least one and probably two changes in government. That is w h y it is essential that these efforts be perceived as national and not partisan undertakings.

Horizontal and Vertical Systems of Public Administration Public administration has evolved o v e r the centuries as the role o f the state became more c o m p l e x and the evolution o f m a n a g e m e n t k n o w l e d g e required in-

14

Studies in CemparativeInternational Development/ Summer 1997

creased specialization, leading to greater functional differentiation. In the assessment of national public administration, the resulting functional division is addressed in terms of two basic groups of systems that have emerged to attend such differentiated functions: horizontal and vertical systems. Figure 2 portrays the dichotomy that is present in all public administrations between the two sets of functions. In this case, the example is drawn from the Bolivian public administration, but in any country a similar situation is found. Bolivia is a unique case because all the horizontal systems apply to the vertical systems responsible for delivery of services to society both at the central government level and at the provincial or municipal level. In societies with federal governments, each state is subject to its own horizontal systems and the same applies to local governmental jurisdictions that, in some countries, are guaranteed autonomy from the central government by their national constitutions. The horizontal systems cut across the entire public sector and provide the normative elements to ensure a minimum level of consistency throughout the whole administration. An argument can be made that such consistency is not desirable and that diversity is better, since it allows for experimentation and competition. However, the fact is that to provide integration of data for national decisions on budget, finance, accounting, personnel, and other aspects of public management, it is necessary to have internal compatibility of the various national systems. Failure to do so in many cases has created administrative Towers of Babel, unable to communicate with each other. This is particularly so in the case of attempts to introduce integrated computerized systems. There are some characteristics of horizontal management systems that have had a negative impact on the quality of national public management. By their nature, these systems do not produce an output addressed to the citizens. Their services or products are consumed within the public administration itself. Yet, they are essential to the provision of services to the ultimate client of the government, society. Their development responded to increasing demands for control of government expenditures to protect society from waste, malfeasance, and corruption. As a result, they tend to generate an organizational culture more oriented to control than to service. In the United States, the National Academy for Public Administration convened a panel of experts under the leadership of Dwight Ink that concluded that these horizontal systems were sapping federal management (Ink 1983). More recently, the campaign to "reinvent government" has targeted several of these horizontal systems as means to reduce the excessive bureaucracy that affects the effectiveness and efficiency of governmental operations (Gore 1993). Similar efforts have been underway in other areas of the world, with New Zealand being the country where the most radical reform of public administration has taken place with a shift from reliance on control of inputs to holding public managers accountable for the outputs and even the results attained from their agencies' efforts, as if they were private enterprises (OECD 1993). These leading edge reforms are seeping slowly into the reform effort in Latin America and the Caribbean and, with the forces pressing for reform of public

~ ~'~'~o

I

9..--

~ 0~ ~0 ~0 0~ o

..

~

~

0

0

o~ ~ <

,.~

.C

4-

I

o " ~0' =.o-~-

0

0

9

~ o~

0

~ 0 ~0

0

0

i

~

0

~

~ 0

0

0

0

0

0

0

0

4.~-

I

0

0

"~

.--

'--

0

0

0

0

0

9

. ~~:~ . ~ --

0

0

0

0

0

0

r

+

$ ..~ I

0

0

..C

+

I

..C

+

:n

[.~

t.,

>."o

o~ t..,

~ o~~

.~_< "~

C

~

o

~8 ~" "--

0

~

~

~

~=

-~.~

~

0

9

o~:

O~

~ ~

ooo

oo.

9

<~

~

16

Studies in ComparativeInternational Development/ Summer 1997

administration increasing every day, it is to be expected that more and more countries will incorporate these approaches in their public sector reform efforts. This article draws on the specific experiences of Bolivia, Ecuador, and Nicaragua, but there are similar efforts in other countries throughout the region. The key reform with respect to the relationship between horizontal and vertical systems in a modem public administration is related to the principle already mentioned: that central agencies responsible for horizontal systems should have policy and normative decision-making authority, but that the operational authority for decision-making for the horizontal systems should be vested as close to the level of service provision as possible. This is consistent with the worldwide trend in public and private management. OrganizationaI Networks

As mentioned, the studies made under the UNDP Management Development Missions were oriented, in some cases, at certain sectors of activities that require the cooperation of many agencies, yet, they cannot be placed in hierarchical relationships because their principals have equivalent ranks. The traditional hierarchical organization structure was predicated on principles of organization based on a command and control approach that rested on the principles of unity of command, which state that each person should be accountable to only one superior, and span of control, which states that no supervisor should be held accountable for more than six or seven subordinates having interrelated responsibilities requiring coordination. However, the fragmentation of specializations and the expansion of government responsibilities have made such principles unworkable in many instances, because, increasingly, it is necessary to pursue national objectives through what are really organizational networks in which the entities interacting toward a national goal have no hierarchical authority over their peer entities in the network. In assessing these networks or sets of entities, the first phase is to identify the macrofunctions that need to be attended at the national level for the selected national objective. Environmental control, alleviating poverty, tourism, and export development are some of the sectors in which the collaboration of many entities is required for a country to succeed. Yet, in these cases, one frequently finds that there is an autonomous entity responsible for the objective, but that it is not related in any effective way to other national agencies whose operational inputs are essential to attain the goal. Figure 3 portrays the organizational network that may be required in a country to expand its non-traditional exports. This is the case if reliance is to be made on small or medium-sized enterprises to undertake the export effort. Large multinational or national corporations usually have, within their organizational boundaries, the capability to take care of most of the functions necessary to get their products into world markets. Nevertheless, as figure 3 shows, there are many contextual functional requirements that even large enterprises may need to be favorable to their efforts if they are going to succeed in developing markets for the country's export products.

Betancourt

17

m o

. , ~ . ~

ONVNOI.LV'dJ.SINII~I(:]V3"1OHM:IHJ_S;~O~IDV3J-V'83clO

~

~

..~

..~

~W O~

o tD

v-. .r

uJ

7~

7J

w

~"

O~

m

7j

,~,.r

~.o. o~

I

0.)

> gl

o L 0

..~ ~

011

smols,~ S l u o m o ~ t m I A I I~luoz!soH

18

Studies in ComparativeInternational Developmem / Snmmer 1997

c~ c c

a

~ L~_ = O0 mZ

Z

t

r &

0

~,~

Betaacoart

19

The first time that use was made of an organizational network to analyze the set of functions and entities involved in export development was in a study undertaken for the Organization of American States in 1978, as part of the design of its export development program. Afterwards, it was used in formulating World Bank assistance to the Jamaican National Export Corporation (JNEC). It was also used to analyze the tourist industry in Mexico in relation to a World Bank loan and Panama's antipoverty efforts under its social investment fund. The first step in defining the network is selecting the macrofunctions that must provide support for the respective clientele, whether exporters, tourist enterprises, or local communities needing social investments. As can be observed in figure 1, for each of the seven macrofunctions required for export development there is a sub-organizational network that includes the pertinent national agencies as well as the nongovernmental associations and individual enterprises active in relation to the particular macrofunction. Somehow, all these entities have to act with a minimum of coordination to provide the individual enterprises the external economies they may need in order to compete successfully. That coordination is provided by the market in the case of most services required for export, and, of course, foreign investors usually have the required capabilities in marketing, product design, and technology within the boundaries of their worldwide organizations. It is the independent national producer who most needs a supportive national institutional context, such as the one described in figure 3. This does not mean necessarily the provision of free goods or services or interventions that distort the functioning of the market. Whatever the situation of the specific country involved, the organizational network for export development, as is the case of any organizational network, requires a focal organization that, due to the nonhierarchical nature of the network, needs to have only power to convene the others as required and power to follow up on the implementation of what is agreed among the various entities. In undertaking the analysis of such a sectoral organizational network, the same approach described under the profile of a modern public administration can be followed. This could provide the initial basis for more detailed surveys of individual sub-networks by specialists to formulate a coherent program that would improve the institutional capability of the country to attain the goals set. Again, it is important to take into account that isolated microinstitutional efforts to improve the agency related to one function could be thwarted by inadequate institutional capability in relation to other complementary functions in the organizational network. Thus, once the diagnosis of the organizational network is made at an aggregate level, a comprehensive institutional development plan is necessary.

MacroinstitutionalPolicyand IntegratedImplementation To undertake a comprehensive reform of the public sector is a complex enterprise. It is necessary to carry out in parallel reforms in the horizontal systems of management and in the ministries and agencies responsible for the vertical systems, whose

20

Studiesin Comparative International Development / Summer 1997

mission is final delivery of services or products to society. This is not easy to accomplish and much less to coordinate not only because of the magnitude and complexity of the undertaking, but also because of the weakness of the central agencies responsible for the horizontal systems and the negative image they have acquired over the years. In addition, there is the problem of the multitude of autonomous agencies created under the project unit phase of development assistance, which resist efforts at restoring the authority of central government agencies over their activities. Therefore, in the first phase of modernizing the horizontal systems--personnel, accounting, informatics, budget, and others--it is also necessary to concentrate on improving or restructuring the central agencies responsible for those functions. Once enough progress has been attained in the design and installation of the new horizontal systems, it is time to move to a second phase involving the ministries and agencies responsible for sectoral or vertical systems. This is a multiyear exercise at both levels that requires a flexible progressive installation calendar. In terms of the horizontal systems, a multiyear installation requires legislation that avoids sweeping one-shot changes throughout the whole administration. Rather, legislation should allow for progressive installation, moving from the center to the more autonomous agencies and, if necessary due to decentralization, to lower levels of government at the provincial or local level. It also requires that the design of the pertinent new horizontal system be conceived in stages, preferably associated with each annual budget cycle. For each stage, it is also advisable to consider the minimum interdependence among systems required to implement the changes. Otherwise, incompatibilities may result in isolated attempts that either fail or are bogged down for lack of inputs from another horizontal system. In terms of the vertical systems, an instrument has been devised, called the Institutional Restructuring or Strengthening Agreement. Under this kind of agreement each ministry or agency manager agrees to undertake a comprehensive effort at modernizing his entity by signing an agreement with the focal agency for the modernization effort. In Bolivia, the focal reform agency is the Public Service Program (PSC); in Ecuador, it is CONAM, the modernization commission; and, in Nicaragua, it is the CERAP, or executive commission for reform of the public administration. In exchange for signing the agreement committing oneself to modernizing the agency--and, without that commitment, it is foolish to attempt the reform of an agency--the executive is offered a variety of incentives such as: 1. financial and technical support for redundant staff buyouts; 2. access to the cadres of management and technical staff recruited under competitive working conditions who could provide the critical mass for the reform effort in the entity; 3. a greater degree of managerial autonomy in managing the horizontal systems; 4. priority in marginal budget allocations and access to international financing; and 5. guarantees of a smooth cash flow during budget implementation against a financial program.

In the case of Bolivia and Ecuador, the comprehensive nature of the reforms is encompassed under the label of performance-based management, which will be addressed later in the article.

Betancourt

21

To address the resistance to change imposed from the center as well as to ensure compatibility of efforts supported by diverse binational and multinational sources of technical assistance, while also ensuring the consistency required to attain a comprehensive and coordinated modernization effort, one of the basic principles of modem public administration is relied upon: To centralize the normative functions required for the reform and decentralize authority for implementation. In the case of Ecuador, this is being done by CONAM through the issuance of a Macroinstitutional Policy, which provides a summary of the basic norms and specifications for the variety of new systems--program-budget, performance measurement, accounting, informatics, and others--that constitute the foundation of the public sector reform effort. This Macroinstitutional Policy will be given to each agency team working in the restructuring of the agency as part of its terms of reference. The only role of the central monitoring entity for the reform effort is to ensure compliance with the standards set in the Macroinstitutional Policy. Of course, an agency or ministry that requests assistance in formulating and implementing its restructuring will be provided with the necessary technical assistance. This will allow those agencies that prefer to undertake their own reform effort, and have resources to pay for it, to implement it on their own, while compliance with the Macroinstitutional Policy ensures that the final product will have the minimum consistency required for a comprehensive public sector reform effort. In the cases of Bolivia and Nicaragua, the public sector reform efforts supported by the World Bank also make provisions for an integrated approach, but in different ways. In Bolivia, as will be seen, reform starts from the provision of an elite managerial and technical corps and links this boost in managerial cadres to a very comprehensive piece of public sector reform legislation. In Nicaragua, the integration is more operational than systemic. It is provided through the central coordination unit established as Secretariat of the CERAP and the use of several large international consulting firms to ensure consistent delivery.

Pub& SectorPe~rmanceManagement Performance-based management is the label used to describe the most recent trend in public administration in the Organization for European Cooperation and Development (OECD) members as well as in state governments in the United States. At the federal government level, the legislation that guides these efforts is the Government Performance and Results Act (GPRA) of 1993. Among OECD members, performance management and measurement has been spreading since the last decade. Its origins come mostly from the reforms started in England and some of the Commonwealth countries, in particular, Australia and New Zealand. As noted earlier, New Zealand is the country that has undertaken the most radical approach to governmental reform by placing the emphasis on results and outputs rather than on control of inputs. Such change in emphasis requires substantial changes in budgeting, accounting, program definition and measurement, personnel policies, and organization, as well as in reporting and evaluation of per-

22

Studies in ComparativeInternational Development ! Summer 1997

formance. Performance pay is perceived as central to the working of the system (OECD 1994). Experience in the United States is not as advanced as in the OECD countries. The National Academy of Public Administration has made a commitment to assist in the implementation of the GPRA and has already issued a report on the initial pilot performance plans. As the evaluation of these pilot plans reveals, performance management is a complex undertaking, and there are many difficulties in defining agency missions and performance measurement indicators. Results, at times, are influenced by many factors outside the control of agency managers. As is to be expected, agency management does not accept accountability for what it cannot control, while outputs are easier to relate to management actions than results. It is accepted that this is an effort that will take time if it is to be effective (NAPA 1994). In the case of Bolivia, the reform effort began with the approval of an overall reform legislation, Law No. 1178 of July 1990. This was a most ambitious piece of legislation since it mandated the modernization of all horizontal systems in the government of Bolivia at all jurisdictions and levels and made allowance for the interrelationship of the diverse horizontal systems. In fact, it is a very comprehensive and coherent public sector reform legislation. However, its implementation has been slow for a variety of reasons, and the operating standards for the various systems were not completed in most cases until 1995. The integrated financial system, the so-called SAFCO project financed by the World Bank and AID, is years behind schedule. More recently, the government of Bolivia started an effort to address the problem of national staffers in management and technical positions financed by international donors, whether bilateral or multilateral, with the intention of eventually having all these line positions financed by the national treasury rather than through donor projects. This resulted in the creation of the Civil Service Program (PSC in Spanish). Under this program, key technical or managerial staffers are recruited through merit processes and paid competitive salaries. They are deployed at agencies willing to enter into performance agreements with the PSC at the level of both the agency and the individual staffer. The PSC found that a critical mass is needed. That is, a large enough number of PSC staffers must be deployed in an agency to overcome resistance to change and have an impact on improved management. A pilot project was commissioned at a Secretarfa, which was under a management-oriented leadership team, to develop a rudimentary performance-oriented program and measuring system, while raising the number of PSC recruited staffers to a number that would provide the required critical mass. Parallel to this effort, it was requested that a study be made leading to the integration of the horizontal reforms, foreseen in Law No. 1178, into a performance-based management system. Figure 4 portrays the components of the Bolivian performance-based management system. The working of the system requires a harmonious interaction among several subsystems. The standards and rules issued in preparation for all the horizontal systems covered under Law No. 1178 are consistent, with minor adjustments, with the needs of a performance-based management system. However, this does not

Betancourt

2~

Z~

W ~

o~5~ ~->~ O0~Ou Z~-->~

o~e~o

z~

~O~E {a

I

/

~ wZ

F--O~

o-,]o

o ~w

o

>oZmE~ ~OWj

<

o~E

o

II

I,L,I

0 Z

i,i I-

iOu.i

>.

(/) I--

iv, ZUJ 0

LU

Q Z

c~ ~WW~Z~

= ,,~

,1," ~0

w Z O- ,~.

\

o~m
n

24

Studies in ComparativeInternational Devdopment / Snmmeg1997

mean that Bolivia is about to transform instantly from a public administration that, as the profile in table 1 reveals, is bureaucratic and traditional into a modern administration. Many years of persistent and patient work will be required to get there.

Technicaland ManagerialCorps The modernization of the public sector is not only a matter of organization charts, systems, flow charts, and computers. These are the inert elements in the process of change. The human factor is the dynamic component in changing the style and culture of an organization. If, as is the case in most countries in Latin America and the Caribbean, the government payroll has been the employer of last resort to reduce unemployment or the payoff to political clienteles, to increase salaries will not improve performance and will tend to increase the resistance of such employees to be removed, with or without buyouts, in case of downsizing efforts. And, as was noted earlier, to increase pay for existing staff in order to make their remuneration competitive with that of the private sector would not necessarily ensure an increase in their competence. In addition, managerial and technical cadres have to provide the leadership for the reform effort and, as the experience in Bolivia reveals, they have to attain a critical mass in order to be effective. There is also a crucial cost factor. It is not financially feasible to convert the whole civil service into a competitive actor in the labor market overnight. Most governmental budgets cannot afford a massive increase in payrolls. Taking into account these two factors, the financial constraints and the need for managerial cadres to lead the public sector reform, an instrument has been suggested: the technical and managerial corps. The idea is to provide for a service that will not have tenure for life, but for a renewable period of up to five years, selected competitively and rewarded with remunerations and other working conditions competitive to those prevailing in the national private sector. This is the source of some debate. Two issues are the most prominent. One is resistance to the dangers of administrative elites usurping the powers of democratically elected rulers, which has been the result of career officer corps in many countries in Latin America and some in the Caribbean. The other is stated in terms of the divisive consequences, internal conflicts and lack of cooperation, of creating two sets of staffers, one paid at competitive labor market rates and the other lagging behind. Both are legitimate concerns that cannot be totally eliminated but can be ameliorated. The first issue is addressed through term limitation of appointments. In contrast with military officers, who enter a career for life, the managerial and technical corps would receive five year appointments that could be renewed. Care should be exercised that such terms overlap the presidential periods to ensure that each administration will have the opportunity of exerting at least partial influence in the composition of the corps. In addition, provision is made for reserving a number of slots for presidential appointments, whose selection is subject only to verification of compat-

Betancourt

25

ibility between their professional credentials and job requirements and whose terms of appointment would end with that of the executive who appointed them. The second issue is addressed by opening to anybody in the existing civil service the opportunity to compete for positions in the managerial and technical corps. This should not give them monopoly access to such positions in the traditional civil service schemes. The idea is to ensure that the government has access to any qualified citizen for its managerial and technical cadres. In addition, remuneration of the rest of civil servants could be adjusted to address increases in the cost of living, always within budgetary limitations faced by the state. The experience in countries such as Argentina has been that the tension generated between the two groups has been less than anticipated. Finally, the increasing concern with corruption within the region makes it more necessary for the public sector to counter with a management and technical corps. The existence of a professional cadre of well paid civil servants is one of the best ways to prevent corruption. The managerial and technical corps concept is working already in Bolivia, as noted in the discussion of the PSC and its impact on introducing performance management. In the case of Nicaragua, it is one of the conditions for a World Bank adjustment loan, and the design of the system is very advanced. It is also one of the most popular incentives offered to agency executives under the restructuring agreements. In Ecuador, the corps system is developing parallel to reforms in civil service legislation, and the CONAM included it among the incentives to be offered to agency executives who sign strengthening agreements.

Strate~ for Introduction of InstitutionalInnovations The final element of the approach to modernize the institutional structure at the macro level is the strategy designed to determine the feasibility and most advisable sequencing of proposed reforms. This is done at two levels. The first level is a component of the initial diagnosis and design of the proposed reforms, and the other level takes place through a workshop given to participants in the teams involved in reform efforts. In both cases, the approach followed is based on combining Kurt Lewin's force field analysis (1951) with the variables defined under the institution building consortium led by Milton Esman and his associates (Eaton 1972). The first provides a simple and easy to understand explanation of the dynamics of change when one is trying to move a social group from one situation to another. It calls for the identification of the forces of resistance and support in the present situation and determines what can be done to increase the positive forces and decrease the negative forces in order to move to the new desired state. The second disaggregates the variables involved in the introduction of institutional change into two groups: 1. Internal:leadership,doctrine,program, structureand resources,which relate to dements of the agencyresponsiblefor introducingthe innovation;

26

Studiesin ComparativeInternationalDevelopment/Summer1997

FIGURE 5 Dynamics of Institutional Innovation Strategy

NORMATIVE

ENABLING

I +

+1

_

_

I FUNCTIONAL

-'1 J

DIFFUSE

Note: The above figure portraits the dynamics of the strategy for the introduction of an institutional innovation. There is an interaction among the internal elements within the circle to ensure that there is consistency in the reform effort. The internal elements of the proposed reform also generate positive or negative interactions with the entities constituting the institutional linkages that will determine whether or not the reform can be implemented. In addition, the entities influence each other on their positive or negative reaction to the reform.

Betancourt

27

2. External: entities that provide the legal and financial resources to ensure the existence of the institution; the entities with which the institution interacts functionally in the provision of services and the acquisition of inputs; the entities that may have indirect influence in the sphere of the reform being introduced due to their normative role in society and those that are not articulated in formal organizations but, in the end, may be stakeholders in the final output of the institution.

Figure 5 portrays the complex set of interactions between the internal variables of institutional reform and its context or environment. In the end, there has to be a prevalence of positive forces over negative ones for the successful introduction of a reform. During the diagnostic missions at the macrolevel, these factors are considered in determining whether there is a favorable climate for public sector reform in the country, as well as in selecting the areas of reform that are more likely to engender support for the modernization effort. During the implementation phase at sectoral agencies or at those responsible for horizontal systems, workshops enhance awareness among the reform team of how to identify negative and positive forces in their specific institutional context, and how to develop change strategies to cope with them. Participation in the workshops is limited to those involved in a specific reform effort and who also share the same database. In these cases, people focus immediately on the key issues and actors relevant to get the reform accepted.

Note I wish to acknowledge the assistance of Mr. Jesus Renzoli in the formulation of the graphics used in this article.

Re~rences EVAN, WILLIAM 1984 Organizationtheory, structures, systems and environments. New York: John Wiley and Sons. GORE, ALBERT 1993 Creating a government that works better and costs less: The report of the national performance review. New York: Penguin Books. GOVERNMENT OF JAMAICA 1992 Report of committee of advisors on government structure, coordinated by Rex Nettleford. Kingston: Bank of Jamaica. INK, DWIGHT A. 1983 Revitalizing federal management: Managers and their overburdened systems. Washington, D.C.: National Academy of Public Administration. KATZ, SAUL M. 1965 System approach to development administration, Papers in comparative public administration. Special Series No. 6. Washington, D.C.: Comparative Administration Group, ASPA. LAMB, GEOFFREY 1987 Managing economic policy change, Institutional dimensions. World Bank Discussion Papers, no. 14. June. Washington, D.C.: The World Bank. LEWIN, KURT 195I Field theory in social science. New York: Harper and Row.

28

Stadiesin ComparativeInternationalDevelopmentI Summer 1997

MOORE, RICHARD J. 1992 A preliminary analysis of the bank's approach to core ministry restructuring with a focus on LAC activities. LATPS Occasional Paper Series no. 4. February 25. Public Sector Management Division, Technical Department. Washington D.C.: The World Bank. NAPA ADVISORY PANEL ON IMPROVING GOVERNMENT PERFORMANCE 1994 Toward useful performance measurement: Lessons learned from initial performance plans prepared under the Government Performance and Results Act. Washington, D.C.: NAPA. NUNBERG, BARBARA 1990 Public sector pay and employment reform: A review o f World Bank's experience. World Bank Discussion Papers no. 98. November. Washington, D.C.: The World Bank. Public sector management issues in structural adjustment lending. World Bank Discus1990 sion Papers, no. 99. August. Washington, D.C.: The World Bank. OAS Progreso de la administraci6n para el desaroUo en america latina. Report of Delibera1967 tions. Inter-American Seminar, Buenos Aires, December 6-15. Washington, D.C.: General Secretariat of OAS. OAS specialized conference on the draft inter-American Convention against corruption. 1996 Caracas. March 29. (OEA/SER. KXXXIV. 1, 14/96 rev. 2). OECD Public management developments: Survey 1993. Paris: OECD. 1993 Performance management in government: Performance measurement and results-ori1994 ented management. Public Management Occasional Papers no. 3. Paris: OECD. PAUL, SAMUEL 1990 Institutional reforms in sector adjustment operations: The World Bank's experience. World Bank Discussion Papers, no. 92. June. Washington, D.C.: The World Bank. PRICE, LISETI'E 1991 Governance: Experience in Latin America and the Caribbean. LATPS Occasional Papers Series, no. 1. March 29. Public Sector Management Division. Technical Department. Washington D.C.: The World Bank. RIBE, HELENA, CARVALHO, SONIA, ET AL. 1990 How adjustment programs can help the poor: The World Bank's experience. World Bank Discussion Papers, no. 71. August. Washington D.C.: The W o r d Bank. STIGLITZ, JOSEPH E., ET AL. 1989 The economic role of the state, edited by Arnold Heertje. Cambridge MA: Basil Blackwell.