Good Enough Governance_Poverty Reduction and Reform in Developing Countries

Good Enough Governance:
Poverty Reduction and Reform in
Developing Countries*
Merilee S. Grindle
Edward S. Mason Professor of International Development
Kennedy School of Government
Harvard University merilee_grindle@harvard.edu November 2002
“People now place their hope in God, since the government is no longer involved in such matters.”1 So lamented a poverty-stricken citizen of Armenia. Indeed, it is all too clear that when governments perform poorly, resources are wasted, services go undelivered, and citizens—especially the poor—are denied social, legal, and economic protection. For many in the development community, good governance has become as imperative to poverty reduction as it has become to development more generally.2
Yet good governance is deeply problematic as a guide to development. Getting good governance calls for improvements that touch virtually all aspects of the public sector—from institutions that set the rules of the game for economic and political interaction, to organizations that manage administrative systems and deliver goods and services to citizens, to human resources that staff government bureaucracies, to the interface of officials and citizens in political and bureaucratic arenas. Getting good governance at times implies changes in political organization, the representation of interests, and processes for public debate and policy decision-making. Not surprisingly, advocating good governance raises a host of questions about what needs to be done, when it needs to be done, and how it needs to be done.
When good governance is advocated as a necessary ingredient for reducing widespread poverty, these questions are compounded. This is particularly so for countries attacking poverty as a condition for debt relief. Among them are the poorest countries in the world. Almost by definition their institutions are weak, vulnerable, and very imperfect; their public organizations are bereft of resources and are usually badly managed; those who work for government are generally poorly trained and motivated.
Frequently, the legitimacy of poor country governments is questionable; their commitments to change are often undermined by political discord; their civil societies may be disenfranchised, deeply divided, and ill equipped to participate effectively in politics.3 Despite these conditions, expectations for what such countries should accomplish are high. While Poverty Reduction Strategy Papers (PRSPs) have

*

Prepared for the Poverty Reduction Group of the World Bank.

encouraged governments to define the tasks they must take on to reduce poverty, ownership does little to make the list of governance reforms more manageable.
What’s a country to do in the face of these problems? And what role should donors take on in efforts to promote the good governance agenda? Among the issues that need to be addressed is the overwhelming nature of the agenda—it is long and expanding.
More important, among the long list of things that “must be done” to encourage development and reduce poverty, there is little guidance about what’s essential and what’s not, what should come first and what should follow, what is feasible and what is not. If more attention is given to sorting out these kinds of issues, the end point of the good governance imperative might be recast as “good enough governance,” that is, a condition of minimally acceptable government performance and civil society engagement that does not significantly hinder economic and political development and that permits poverty reduction initiatives to go forward. After a review of governance concerns as laid out in the PRSPs and the process through which they were developed, the paper outlines some of the ways of building toward a concept of good enough governance.
Even if a more parsimonious agenda for good enough governance can be devised, efforts to date suggest important lessons about how government capacity can be improved and how the role of civil society in building more effective and responsive government can be strengthened. In the third part of the paper, I consider what we have learned in recent years about public sector reform, participation of civil society and the poor in improving government effectiveness and responsiveness, and the politics of institutional and policy change. In the fourth section of the paper, I consider what the foregoing means in terms of the roles that donors can play in advancing the good enough governance agenda. There are no technical or easy fixes to what is inevitably a long, slow, reversible, and frustrating path toward better performing governments, but there may be ways of reducing the burden on those attempting to undertake the journey.

Governance and Process in the PRSPs
Almost all of the Poverty Reduction Strategy Papers reviewed considered governance issues.4 More tellingly, many of the PRSPs indicated that good governance was among the “pillars” of the country strategy to reduce poverty; in most cases, it was one of four to seven such pillars. Almost universally, the country documents noted that good governance was fundamental to the achievement of many of the other goals for poverty reduction set out in the papers. The government of Albania, for example,
“considers that progress in institutional reform is fundamental for the attainment of the objectives” of economic growth and poverty reduction. Guinea’s PRSP argues that
“Good governance is…a critical prerequisite for economic and social development.”
What Was Addressed? Most governments presented good governance as an important objective for reform in their countries; for them, it was set out as one of a limited number of priorities. A large number of PRSPs went a good deal farther and presented a list of governance issues they planned to deal with. Central among these were the reform of judicial systems, public administration reform, anti-corruption,

2

decentralization, and public expenditure management. For most countries, these reform commitments represented important—if very ambitious—steps toward better goverance.
Among the reforms to the judicial system, several countries committed themselves to increase its independence, to strengthen its human resources, to increase its budgetary allocations, and to revise legal codes. While some countries—Honduras.
Guyana, and Mauritania, for example—indicated the importance of human rights and poverty concerns, other countries—Tanzania and Uganda, for example—focused more on the role of the judiciary and the rule of law for goals such as investment and stability.
Many countries also coupled discussions of judicial reform with concerns about public safety and law enforcement. With some exceptions, decisions about which aspects of judicial reform to pursue were not explicitly related to poverty alleviation or to the poor and their access to the legal system. Instead, judicial reform and the rule of law were promoted as general objectives of relevance to all citizens and all sectors of the economy.
Public administration reform was also highlighted as an important aspect of creating good governance in many countries, and proposed reforms were often broad in scope, such as introducing a merit-based civil service system or efforts to decentralize or deconcentrate government. In Tajikistan, improving public administration was the only issue considered under the good governance discussion. Some PRSPs were more specific, indicating the need to improve pay and incentive systems and to give priority to some areas—strengthening the capacity of health and education ministries, as in the case of Uganda, for example. Similarly, Ethiopia indicated the need to strengthen the civil service at the local level, in consequence of its commitment to decentralization. Overall, proposals to strengthen public administration were not clearly related to the goal of poverty alleviation; they were, instead, general commitments to improve government performance. Commitments to fight corruption figured in most of the PRSPs reviewed, often in conjunction with discussions of judicial and public administration reforms. The Malawi report, for example, stressed the importance of improving the prosecution of those accused of corrupt acts and strengthening the government’s ability to detect such activities. Most frequently, the papers indicated the need to improve accountability and transparency of government processes, and some singled out procurement, public information, and parliamentary oversight as critical areas needing attention. Much of the concern about corruption was addressed in discussions about public expenditure management and budgetary processes. Because corruption is a complex issue, and was generally not related to a diagnosis of its causes or cures in the PRSPs, it is not surprising that some governments suggested actions that will probably not bear much fruit.
Nicaragua, for example, proposed a national anti-corruption council and inculcating a culture of integrity in the public administration, Honduras recommended a new culture in the public service, and Burkina Faso indicated plans to enhance probity in the public service. Decentralization also figured in most discussions of good governance in the
PRSPs I reviewed. Indeed, all governments committed themselves to pursuing

3

decentralization. Anumber of countries already had decentralization initiatives in place; the PRSPs indicated plans to continue and strengthen this process. Most frequently, specific actions mentioned involved strengthening the capacity of local governments, particularly in terms of fiscal management and training for a wide variety of new tasks and responsibilities. In several cases, governments committed to legal and regulatory reforms to clarify responsibilities among levels of government. More participatory approaches to local budgetary planning and management are promised in a few of the documents. Overall, decentralization was put forth as a central aspect of good governance and as a process that required further emphasis and support. Like other aspects of good governance, it was promoted as a goal in its own right rather than as a means of promoting poverty reduction.
Public expenditure management was generally dealt with as a separate category in the PRSPs, but related to the discussion of good governance. Where it was discussed in detail, plans included efforts to increase efficiency and particularly to increase the capacity to manage information about public expenditures and to promote greater transparency in the budgetary process. Many countries committed to promoting or continuing a series of auditing, monitoring, and review procedures. In general, improved budgeting and expenditure management were proposed as reforms that must be dealt with if other aspects of the strategies were to be effective. Indeed, among the various governance related reforms—many of which are complicated and require significant periods of time to demonstrate results—budgeting, auditing, expenditure reviews, and other such procedures are probably among the easier to undertake from an administrative perspective and among those whose benefits can be demonstrated most readily.5
This characteristic of budgeting and public expenditure management reforms suggests that they may be a focus for particular World Bank and other donor concern.
Such reforms, if carried out successfully, touch on many other aspects of governance, including anti-corruption, increased management efficiency, transparency, and accountability and could be instrumental in strengthening them. Given that these kinds of reforms may also be more amenable to monitoring and measuring, they also may be of particular interest to donors concerned about benchmarking, achieving measurable goals, and sequencing other reforms around them. Typically, donors also have a comparative advantage in the kinds of expertise that contributes to developing capacity for budgeting and expenditure management. Thus, budgeting and public expenditure management might be a central focus of donor assistance to many of the countries.
Other concerns figured in the discussions of good governance found in the
PRSPs, such as democratization (Ethiopia and Vietnam), political will (Malawi), information on government activities (Rwanda and Zambia), improving social services
(Mauritania), reforming state-owned enterprises (Vietnam), and innovative ideas about citizen monitoring (Bolivia). To their credit, some governments facing significant challenges of political instability recognized the importance of establishing law, order, and stability as critically important governance conditions for poverty reduction.

4

What Was Not Addressed? Some aspects of the governance agenda outlined in the PRSP Sourcebook were largely ignored. For example, the discussion of governance reforms paid very little attention, if any, to the issue of checks and balances among branches of government, beyond concerns about strengthening the judicial process. By and large, checks and balances in the management of public resources were addressed by committing to executive-based action with regard to public expenditure management and information provision. As indicated below, legislatures were not much discussed, and usually only in terms of their participation in the PRSP process rather than through reforms that would strengthen their capacities to hold public officials and public agencies accountable for their actions. Going beyond the three branches of government, some countries announced the creation of public oversight bodies formed of various representatives of civil society. Overall, however, most issues of accountability and transparency—concerns that figured very prominently in the discussions of governance problems—were addressed by putting in place technical rather than political checks.
Checks and balances based in institutional relationships and capacities and in political accountability would no doubt lead to more sustainable reforms.
More generally, while judicial, civil service, anti-corruption, decentralization, and public expenditure reforms received the most consistent and universal treatment in the papers, and most PRSPs presented serious discussions of the good governance imperative—Zambia’s paper had a full chapter discussing the issue—there were consistent weaknesses among them. Most, for example, failed to establish priorities among governance concerns. Similarly, most PRSPs did not acknowledge the time dimension required for improving most aspects of governance. And many failed to link proposed actions to poverty and its reduction. Indeed, the Joint Staff Assessments (JSAs) characteristically “welcomed” the priority given to good governance in the documents and then suggested a variety of ways in which the discussions fell short of expectations.
They indicated that most governments had not provided much detail on proposed reforms and had not indicated priorities, outlined specific steps that need to be taken, or suggested how they would measure, monitor, and evaluate what they committed themselves to do.
In this assessment, the JSAs were largely correct.
These failures suggest that despite considerable discussion of the importance of good governance and indications of a variety of steps that needed to be taken, concern with good governance in the PRSPs often lacked depth in linking a clear diagnosis with clearly relevant remedies. To their credit, several governments indicated specific problems in their countries—patrimonialism in Bolivia, violence and the enduring legacy of genocide in Rwanda, need for democratization in Ethiopia, extensive centralization in
Guyana, corruption in Mozambique, for example—and indicated commitment to remove some of these barriers to development. Yet it cannot be said that any of the PRSPs provided a full diagnosis of the problems of governance prior to indicating what actions needed to be taken. Little data was presented and analysis of the problem was frequently replaced by statements about the importance of good governance to poverty alleviation or of statements about existing weaknesses in the judiciary, the management of public resources, and so forth. Somewhat more analysis was presented in the sector analyses, particularly with regard to the ability of public sector ministries to deliver important

5

social services like education, health, pubic sanitation, and water. Generally, however, such analyses were not used to generate specific plans for governance improvements.
Moreover, as indicated, few of the reform commitments were directly tied to their consequences or impact on poverty reduction. Again, while the poverty analyses indicated a series of ways in which public sector organizations and conditions in the public sector affected poor people, the governance agendas set forth in most of the PRSPs were based on commitments of general import—civil service reform, better public expenditure management, decentralization—that were not linked explicitly to poverty or its reduction. Although this is a weakness in most of the PRSPSs, it should not be surprising to find it. Data on, say, the difficulties that poor people have in gaining access to judicial redress of grievances, or the amount they must pay in bribes and “fees” to see a nurse or send their children to school, or the extent of corruption in public contracting are not likely to be available. It is with such data, however, that better analysis could be made of priorities in attacking governance deficits, considering which kinds of governance failures most affect the poor, and which kinds of actions would be most effective in combating such problems.
Governance Reforms, Capacity, and Indicators. There is also reason to question, as many of the JSAs do, the capacity of governments to make good on the commitments they made in the PRSPs. The reforms they mentioned would be difficult for any government to undertake. Often, governance reforms are interrelated, as are, for example, better public expenditure management and efforts to fight corruption, or more professional public administrations and decentralization. Governance reforms also take time—perhaps decades—to be put in place and to produce results. Often they are expensive. It is worth questioning whether governments that are weak financially, politically, institutionally, and in human resource terms can pursue several such reforms at the same time. Moreover, although the JSAs do not address this issue, there is reason to question the commitments of a number of the governments to carry out any reforms that might undermine their sources of support or their ability to benefit from corrupt activities. Clearly, governments that face civil unrest or open domestic conflict could only proceed with such reforms if they enjoyed widespread legitimacy. Usually, however, governments facing such challenges are accorded very low legitimacy—at least for significant portions of the population—and clearly lack the capacity to make and enforce authoritative decisions. For these reasons, the Joint Staff Assessments are correct in indicating some skepticism about the ability to implement the reforms that governments committed themselves to.
It is understandable that many governments avoided the task of setting targets and performance measures for many governance reforms. To the degree that countries addressed issues of measurements and targets, most were able to devise them for a relatively narrow range of reforms, largely having to do with budgeting and public expenditure management. Other specific goals were those relating to commonly acknowledged measures of human development—infant mortality, school enrolment ratios, incidence of poverty—or those that involved infrastructure—roads, schools, and health clinics built or repaired, for example. Although it may be relatively easy to

6

imagine clear criteria for judging the accomplishment of better management of public expenditures, and to use well-recognized measures related to human development and infrastructure achievements, it is more difficult to assess how progress in accomplishing other governance reforms might be assessed, such controlling corruption or improving local government.
The problems of setting indicators for governance reforms is evident in Zambia’s
PRSP, which was more adventurous than others in proposing such measures. It proposed governance indicators in three areas—better decision making through the consultative process, transparency, equity, and public resource management, and judicial system improvement. In the first category, the government proposed measures such as the advance of decentralization, the proportion of women among members of parliaments, a universal denunciation of corruption among political party leaders, popular perceptions about the level of corruption, and review of executive and legislative branch legislation.
In the second category, the PRSP indicated that the government would monitor public expenditures, the extent to which public servants were selected by the civil service commission, the monetization of civil service benefits, and the reduction of corruption
(but without indicating how this would be measured). In terms of judicial reforms, the government would monitor reduction in the backlog of cases in the system, how many new courtrooms were built, and citizen perception of the level of crime.
Many of these are interesting and innovative measures. Yet whether they are appropriate measures for knowing if in fact government is performing better is a matter for considerable discussion. At the same time, it is not clear that there are better measures available to countries that are usually bereft of information and that lack human and financial resources for acquiring such information. Governments of poor countries can perhaps be excused from avoiding such a complex and often controversial issue as how to measure improvements in governance. As indicated above, in sorting out priorities for action, selecting areas that are easier to monitor and measure—in the area of budgeting and public expenditure management, for example—may be one way of moving the PRSP process forward.
In fact, as I suggest in the second section of the paper, it is not surprising that the
PRSPs were often disappointing in terms of how they addressed the good governance agenda—there is precious little research or discussion to guide countries about which among governance reforms should have priority, the amount of time required to improve how government works, or which among a variety of actions have the greatest payoff in terms of alleviating poverty. More generally, the JSAs are correct in pointing out serious concerns about the ability of most governments to implement and sustain the reforms they promise. This suggests that the PRSP process presents a dilemma for governments.
On the one hand it is important and useful that governments of poor countries explicitly address important limitations in governance, while on the other there must be some skepticism about the capacity of even well meaning governments (and it should not be assumed that all are well-meaning) to achieve the objectives they set for themselves. For donors, the dilemma is more complex, as they must decide whether acknowledging problems of governance, among many other issues of importance, is sufficient for

7

responding with assistance, or if some evidence of commitment to pursue the reforms is needed. This dilemma is compounded by recognizing that it is often the countries most in need of governance reforms that are in the weakest position to undertake them.
Benefits in Process. This is not to argue that the PRSPs are not valuable documents. While most can be faulted for their failure to present realistic plans for improving governance, many of them introduced new dynamics into the process of national planning and priority setting. Indeed, the most valuable aspect of the PRSPs may well be the processes involved in their development. These dynamics included two forms of consultation, one taking place within government and one taking place with civil society.6 Most governments that produced PRSPs set up mechanisms for consultation and interministerial coordination. Often this took the form of a national steering committee
(Albania and Ethiopia, for example), a national policy council (Nicaragua) or commission (Mauritania) or coordinating office (The Gambia). Frequently, these new organizations established a technical agency or committee reporting to them as well as a secretariat to manage the day-to-day development of the document. In other cases, the
PRSP was coordinated through ministries of finance and planning, but with the expectation that they would not prepare these documents alone but rather consult broadly within government about their contents. For the planning process to work, representatives from many different ministries and agencies across government had to meet and communicate regularly. In terms of the goal of poverty reduction, this kind of intergovernmental discussion is a novelty in many countries. Even while many countries have long had mechanisms to coordinate economic policy planning, most have not devised means for discussing and coordinating policies related to poverty.
In particular, drawing social service ministries into cross-government discussions had many potential benefits. In most countries, these ministries tend to be politically and administratively weak. Although they receive considerable budgetary support, most of their expenditures are committed to pay the salaries of large cadres of public service providers—teachers, administrators, nurses, doctors, public health officials—and little is available for promoting better performance or new initiatives. These ministries also tend to suffer from severe capacity limitations—they are often inefficient, they may be colonized by unions of service providers, they are frequently bereft of information or the means to acquire information, their procedures are often arcane, complicated, and centralized. Traditionally, they have usually not been at the table when important policy decisions have been made. With the introduction of the PRSP process, however, these ministries were drawn into serious discussions with powerful economic ministries. Their presence helped bring to light the very real problems that line agencies confront in delivering services to poor and often remote locations, the fragility of their administrative and monitoring systems, the lack of staff capacity, the lack of resources, and other weaknesses commonly suffered by service providing ministries.
Similarly, ministries of agriculture and public works were often involved in interministerial discussions leading to the PRSPs. Given the large tasks and shortage of

8

resources often characterizing such ministries, discussions raised issues of capacity and priorities for them also. Again, although the JSAs frequently questioned the implementation capacity of governments, the process of bringing these often forgotten ministries together with national decision makers and with the leaders of much more powerful economic ministries undoubtedly led to more realistic PRSPs than would have been the case if the document had been left to technocrats in ministries of finance. Thus, one benefit of the process of producing PRSPs is the extent to which governments found themselves talking across ministries about similar kinds of concerns and around the central theme of poverty reduction.
This process, although not always reflected in the PRSP document, undoubtedly increased awareness among national policy makers of the difficulties faced by these organizations and the need to address them if steps were to be made in alleviating poverty. Moreover, discussions of indicators and information undoubtedly raised consciousness about the paucity of information about basic conditions in social, agricultural, and infrastructure sectors and the inability to target service delivery and other activities without such data. Indeed, many PRSPs contained commitments to produce better information so that existing conditions could be better known and monitoring of progress could occur.
Of course, most of the intra-governmental discussions of the PRSPs occurred within the executive branch. With some exceptions (Niger, for example), parliaments were generally not part of PRSP planning, although many of them discussed the documents and were asked to approve them. Of course, in most countries, policy planning is recognized as an executive prerogative and the role of parliaments is that of discussing, approving, or rejecting the proposals that come from the executive branch.
Nevertheless, although parliaments were often provided with information and opportunities to discuss the PRSPs, and although they approved the documents in many countries, it cannot be said that intra-governmental consultation about them was strongly geared toward the legislatures.
Consultations with civil society organizations, the second of the two processes undertaken in the PRSPs, were often very inclusive. While this process was everywhere initiated and managed by government—thus potentially limiting the range of voices to be heard in the discussions—most of them sought to engender fairly broad discussions of the problem of poverty. In some cases, national dialogues were undertaken through a phased process of local, regional, and national consultations (Bolivia). In some cases, thematic working groups composed of representatives of different civil society organizations engaged in ongoing discussions about problems of poverty and the PRSPs and consulted with local governments, traditional authorities, and a range of other groups (Malawi,
Zambia). In other cases, governments disseminated information and solicited feedback through a series of meetings and workshops (Mozambique, Niger). National workshops were scheduled in a number of countries (Guinea, The Gambia). Some countries, such as
Rwanda and The Gambia, introduced innovative techniques for understanding poverty based on the methodology of participatory rural appraisal. In Nicaragua, the organizations responsible for the PRSP process established a web site to disseminate

9

information about the PRSP and its development. In most cases, national dialogues and consultations engaged a large number of organizations and provided an opportunity for a variety of voices to be heard. In some cases, national councils of NGOs and other organizations, such as those representing the private sector, established committees or working groups to contribute to or monitor the PRSPs. Some PRSPs even complained of consultation fatigue (Mozambique).
Those most frequently engaged in the consultations were NGOs, local governments, and organizations representing businesses, and religious establishments. At times, organizations representing workers were involved and only rarely were political parties engaged in the discussions. The PRSPs make very frequent reference to the poor, but in most cases it was surrogates for the poor (such as NGOs) who were actual participants in the process. There was little effort to disaggregate “the poor” in the consultations, although some countries made a point of meeting with representatives of ethnic groups often excluded from the benefits of public policy.
Despite their tendency to be top-down exercises, much of value was harvested from these consultations, at least according to their descriptions in the PRSPs. Among the important information that came from these consultations was the extent to which those advocating for the poor indicated very specific problems that governments should work to correct. In particular, participatory processes regularly pointed to major weaknesses in the delivery of education and health care services and indicated that implementation of national policies was more often the problem than the contents of the policies. Indeed, one message indicated in many reviews of the consultataive process was the extent to which very little in terms of government services or activities was reaching the poor, particularly the rural poor. Similarly, governments were pressed to pay more attention to basic activities of poor people, such as agriculture, and orient their policies more to this sector, and to the needs for basic infrastructure like roads and water.
Repeatedly, PRSP planners were told that more attention needed to be paid to creating jobs and sources of income for poor people. In the reports provided in the PRSPs, the focus on poverty meant that government officials heard much about the vulnerability of the poor and to the failure of government services to reach them. From this perspective, the consultative process created an important vehicle for registering concerns with government and providing a reality check on plans made in capital cities.
Just as important, however, voices and concerns that emerged from the consultations were recorded in many of the PRSPs in the discussion of the participatory process, yet they were not necessarily picked up on in the strategies outlined in the documents, particularly those related to governance. Given that so many of the consultations focused attention on basic parameters of service delivery in health and education, for example, the PRSP documents might have done much to link discussions of governance to problems of service delivery and implementation. In fact, however,
Mauritania was one of the few countries to make a clear connection between participatory feedback and national strategies for poverty alleviation. Its PRSP singled out strengthening of health and education ministries as a key to poverty alleviation. Even here, though, the message did not extend to the governance agenda.

10

Had governments been more assiduous in relating what they heard to concerns about governance, the often-mentioned failure to establish priorities among governance concerns might have been partially resolved. In this sense, for example, taking a cue from the consultations might have encouraged government to focus attention on a few ministries rather than committing to overall civil service reform. This might also have led governments to focus greatest attention in judicial reforms on those that could address sources of vulnerability of poor people, and to take other actions that clearly related governance concerns to poverty reduction. Indubitably, the national consultations contributed much to greater awareness of the concerns of citizens in poor countries, but they might have had greater impact, and have helped deal with the difficulties of the good governance agenda, had they been pursued as part of a national agenda setting process.
Nevertheless, the PRSP process resulted in some positive steps toward addressing the governance agenda. Countries dealt with the issue of poverty reduction in terms of government planning, they acknowledged the importance of governance to poverty reduction, they engaged in important intra-governmental discussions about reform, they engaged various constituencies in the process and, on occasion, reflected feedback from civil society groups in their plans.7 These are important achievements and increase the likelihood that the governance agenda is taken seriously as an aspect of an overall country strategy to reduce poverty.
Of course, the contents of the PRSPs and the process through which they were developed provide many examples of problems not addressed and things that could be done better. Indeed, reviews of early PRSPs found that the discussion of governance issues in many country documents was insufficient.8 Yet, however much criticism is focused on the PRSPs, in many ways, the problems of insufficient attention to priorities, of inattention to data and measurement, and of failures to take implementation capacity into account cannot be laid at the door of the governments that prepared the documents.
The problem is that many governments—particularly those with limited technical, administrative, and political capacities—can easily promise much more than they can deliver in terms of governance reforms. System overload is a real risk for many of them.
This problem is the result of the overdeveloped state of the governance agenda. Above all, this agenda is far too long for governments of poor countries to approach with clarity, commitment, or reasonable expectations. Indeed, the good governance imperative might be more reasonable if it were recast as good enough governance.

Toward a Good Enough Governance Agenda
The good governance agenda is overwhelming.9 If the World Development
Report can be taken as a measure of mainstream development thinking, and if the 1997 report can be taken as a statement of revised perspectives about the role of government in the development process, the list of what is needed for good governance has grown annually. Table 1 provides a list of recommendations taken from World Development
Reports from 1997 to 2002/2003, a list that has grown to include a very wide range of characteristics and actions that development-oriented governments are advised to adopt

11

and practice. Clearly, countries in need of good governance must undertake a great deal to get it—and the longer they wait, it seems, the more things they will need to do to get it.
Indeed, it may be difficult to identify a desirable condition or action that is not conducive to good governance, suggesting that the underlying agenda is actually a search for a cure to underdevelopment.
Table 1
The governance agenda has been compiled in part through research, in part in response to advocacy, in part out of frustration with development experience.
Unfortunately, the confluence of these sources often means that for any given country, a multitude of governance reforms are being undertaken at the same time, differentially supported by a plethora of donors, often with little thought to their sequencing, their interdependence, or their relative contributions to the overall goal of creating governments that are more efficient, effective, and responsive.10 While it is recognized that the ability to move toward good governance is dependent on country-specific characteristics, the agenda does not set priorities or define sequences of actions.11 It does not separate activities that are easier to undertake from those that are more difficult, those that can be achieved in the short term and those that will take years if not decades to accomplish. It does not provide insight into the dynamics that surround efforts to change current conditions. It does not take seriously the contentious nature of the changes it recommends. And it does not separate an ideal state of good governance from one that is
“good enough.”
Highly indebted poor countries preparing PRSPs were encouraged to consider a more parsimonious list of governance related issues, but the number of issues requiring attention was still impressive (see Table 2). These countries were asked to empower the poor by ensuring that there are fair elections and checks and balances in the
“architecture” of the state, that budgets were effective mechanisms for managing government affairs, and that there were pro-poor policies and good representative institutions in place. They were encouraged to adopt mechanisms for the effective delivery of basic services, an objective that is to be reached through budgetary measures, clearly defined responsibilities, committed civil servants, diverse methods for ensuring accountability, diverse modes of delivery, and local capacity. In addition, it is important that there be a variety of economic protections for the poor such as laws and regulations, efforts to reduce exclusion, and investment in physical capital. Further, governments have responsibility to protect the poor from adverse economic shocks by putting in place a series of institutional, organizational, and policy guarantees. And governments must have effective court and police systems to protect the poor from corruption and violence.12 Table 2
As indicated in the first part of the paper, many PRSPs reflect commitments to making a number of governance changes, but even stable and committed governments would have difficulty undertaking and sustaining them.13 Governments that are poor,

12

disorganized, vulnerable to political disruption, and lacking in legitimacy are particularly hard pressed to carry out such extensive commitments. As we saw, countries that prepared PRSPs varied in the analysis of governance issues and were often weak in the ability to link them to poverty reduction. Indeed, many PRSPs promised wholesale reform of government activities while others treated the topic much less seriously.14
Even where governance issues were a significant part of PRSP analyses, stated intentions were often overarching. Rarely were priorities among reforms addressed, their varying degree of difficulty acknowledged, or their differing time frames noted. According to one review, “most [PRSPs] lacked a systematic diagnosis of what the key governance challenges are and the implications of specific governance reforms for poverty reduction.”15 Given these problems with the governance agenda, is there anything that can be done to make it less overwhelming, less additive, more strategic, and more feasible for countries that may lack even basic capacities required to put authoritative changes in place? Possibly. Several actions might lead to a more realistic agenda for good enough governance—addressing the link between governance and poverty reduction more carefully, situating good governance historically and developmentally, asking different questions about change, assessing more carefully who needs to do what, and applying priorities on a country-by-country basis. While the governance agenda is likely to remain a challenging one, there are ways to provide better guidance about what needs to be done, how it needs to be done, and when it needs to be done.
What’s the Payoff for Poverty Reduction? It is hard to find fault with most of the items on the good governance agenda. Usually, a clear and compelling argument can be made about why each action or condition is critical to poverty reduction—reducing corruption, improving accountability, decentralizing government, managing public resources better, restructuring the civil service, and so on. Moreover, many of the conditions of good governance are laudable goals in and of themselves—the efficient use of resources, the effective delivery of services, responsiveness to the poor majority. Who can argue with such goals? And who doubts that conditions of poverty could be alleviated if governments performed better?
Yet logical arguments need to be scrutinized carefully to assess their purported payoff for poverty reduction. In particular, it is important to assess which reforms are encouraged and pursued because they are good for governance, and which are particularly relevant to poverty reduction. Civil service reform, for example, may improve pay and conditions of work for government officials, and it may even reduce corruption and patronage, but may mean little to the poor unless other conditions are in place, such as effective political organization among the poor to ensure that public officials treat them fairly or organizational cultures that encourage a service orientation among public officials. Decentralization may remove bottlenecks in decision-making and might even make regional and local officials more accountable to local citizens, but it does not necessarily do so; and decentralization can easily lead to increased inequality among regions and constituencies (see Box 1).

13

Box 1
The PRSP Sourcebook provides other examples of governance reforms that may be important as objectives for improving the performance of governments but that may be only conditionally connected to poverty reduction. For instance, those who prepare the PRSPs are encouraged to consider the role that legislatures play in oversight, transparency, and monitoring of strategies and policies for poverty reduction. Certainly such activities can increase the checks and balances within the “architecture” of government, but are most likely to result in greater commitment to poverty reduction and more effective strategies for it only when the poor are effectively organized and represented by political parties that gain seats in those legislatures. Similarly, the better off, who generally have access to public services, may be more affected by corruption in their provision than the very poor in rural areas who are rarely reached by such services.
The danger, of course, is that governments will expend precious capacity, resources, and political capital making changes that may not have much impact on poverty, however laudable the changes are overall. One way to begin to reduce the good governance agenda, then, is to assess more carefully and empirically the payoff of particular kinds of reforms for poverty reduction. Do the poor, for example, regularly gain from efforts to reduce corruption? Do some kinds of corruption have greater impact on the poor than others? Through such analysis, it may be possible to generate tradeoffs that help reduce the agenda, as I have suggested elsewhere.
In terms of steps toward poverty reduction, for example, is it more important to have an autonomous central bank or a judicial system that metes out more equitable justice? a judicial system that ensures property rights or one that protects poor people from police harassment? a reformed civil service or increased ability to provide maternal-child health care in remote rural areas? an education system that provides basic literacy to all children or one that focuses on improving linkages between primary and secondary education? a political system in which people vote regularly or one in which the poor are integrated into political parties, interest groups, or labor unions? Questions such as these are difficult to answer, the options are not mutually exclusive, and responses to them may vary by country. Asking them, however, may focus attention on what is usually very limited capacity to achieve all good things at once.16
Thinking about the payoff of various kinds of reforms should also illuminate conditions under which particular reforms are more or less effective in reducing poverty.
For example, the mobilization of the poor into political parties, interest groups, unions, and NGOs may be a condition under which judicial reform, civil service reform, decentralization, and other kinds of changes are most likely to have a significant impact on poverty or on the poor. If this is the case, poverty reduction might be advanced more effectively in some countries by focusing less on government and more on strengthening the political clout of the poor in civil society.

14

Thinking in Time.17 Much of the research and advocacy that has contributed to the good governance agenda is ahistorical. Yet the practice of good governance emerged slowly and haltingly in today’s developed countries and was often the work of generations. Undertaking more analysis of the emergence and consolidation of better performance by government, and the sequences in which different ingredients of good governance were undertaken, can provide clues for whittling down the agenda to more manageable proportions. Studies of this nature would provide insight into changes that are essential and those that are less so, sequences of changes, and alternative paths to better performance by government. Moreover, they would indicate that the goal of good enough is often what occurred historically in developed countries, and often what continues to characterize some of their institutions and policies.
Recent research hints at the advantages of thinking in time. In an effort to assess the institutional reform agenda of international financial institutions, for example, Chang explored the development of different institutions of good governance in the history of now developed countries.18 He found that many factors currently considered preconditions for development were actually consequences of it (see Table 3). Indeed, he demonstrates that considerable economic development occurred long before countries had fully institutionalized democracies, professional bureaucracies, rules for corporate governance, modern financial institutions, and extensive social welfare services.
Similarly researching in time, Moore argues that as governments have become more proficient at tax collection, their overall organizational capacity improves.19
Table 3
While such findings do not mean that today’s agenda should be shaped by the historical emergence of institutions in now developed countries, they provide a platform for questioning the “essentialism” of current good governance reforms. As Chang concludes, “Given that institutions are costly to establish and run, demanding [that developing countries] adopt institutions that are not strictly necessary can have serious opportunity cost implications…Even when we agree that certain institutions are
‘necessary,’ we have to be careful in specifying their exact shapes.”20 Research on such issues can provide insight into the impact of particular governance reforms on conditions of poverty and their connection to the development of pressures for additional changes.
Consulting history can also provide insight into the time dimension of change and promote greater tolerance for less than ideal characteristics even in the midst of improvements over time. Most developing countries are young, and are under pressure to create conditions that took developed countries decades and even centuries to achieve. It is worth remembering that fistfights, duels, and use of firearms were a regular characteristic of the institutional development of the U.S. congress well into the 19th century, that spoils were a central fact of life in U.S. politics well into the 20th, that a devastating civil war occurred long after the country was “stable,” and that its early experience was hedged about by compromises to hold the fledgling republic together, including tolerance for the morally obnoxious condition of slavery. This is certainly not meant to justify violence, spoils, civil war, or slavery, but simply to be a reminder that the

15

consolidation of good governance can take a great deal of time, even while economic growth occurs and poverty is reduced.
Assessing the experience of developing countries that have achieved good enough governance can also help sort out changes that are essential from those that are less so— what factors, for example, contributed to decent conditions of governance in Costa Rica,
Botswana, Kerala, Poland, and Chile or countries such as Sri Lanka during particular periods of their history? From another perspective, what kinds of factors undermined governance accomplishments in Kenya, Argentina, and Indonesia? Knowing more about the emergence of good governance over time can provide additional insight into cause and effect relationships, historical sequences that suggest ways to discriminate between the essential and the merely desirable, and changes that can be instrumented in the short term and those that take longer to emerge and produce benefits. More generally, thinking in time can be important in distinguishing between being developed and getting developed. Asking Other Questions. Typically, those concerned about widespread poverty begin to think about solutions by asking “Why is there so much poverty?” In a similar vein, much of the analysis of poverty and government performance begins with questions about what’s wrong: Why are the poor excluded from public services or given substandard or inequitable treatment? Why are the voices of the poor not heard? Why is there so much corruption and such lack of accountability? These are important questions that shed light on causal relationships, lead to important ideas about what needs to be done, and determine why development tasks are being shunted aside or ignored.
Yet looking for what is not in place is almost certain to add to the governance agenda, given all too evident conditions of low growth, distorted markets, corruption, inequity, poorly provided health and education services, instability, violence, and a host of other characteristics of poor countries. Whatever the country, a great deal is likely to need remediation. Thus, asking about what’s not working is likely to lead to very long lists of things that need to be fixed. Such an approach also adds to the difficulty of assessing priorities, sequences, and feasibility.
While it is important to ask what’s missing in a country’s governance profile, questions about improvements that are occurring and the conditions under which they are doing so can provide important insights into how change occurs, the dynamics of reform, and the kinds of interventions that can produce changes that are good enough for improved performance. In recent years, asking such questions of high performing East and Southeast Asian countries added to understanding of policies that had spurred their development and the role of government in this process.21 Similarly, Judith Tendler’s work on the role of intangible incentives and communication in public health initiatives in Northeast Brazil may have added more to understanding factors that contribute to successful interventions than numerous analyses of what’s wrong with public health delivery programs in numerous countries.22 Similarly, studies of how change occurs— rather than much more common accounts of the impediments to change—provide insight into the conditions and dynamics of policy and institutional reform.23

16

Asking about what’s changing can also bring to light aspects of governance that may be improving without specific programmatic interventions by governments and donors. For example, the mobilization of groups in civil society around issues of public corruption—often as a result of scandal or crisis—has been a factor promoting performance improvements in a number of countries recently. Improvements in the human rights performance of governments and the responsiveness of some service delivery programs have also occurred even when they were not intended by governments nor part of deliberate reform programs. Not all change has to be orchestrated, and careful observation of what is occurring in particular countries might indicate some governance issues that are being addressed unexpectedly. Understanding the roots of this dynamic can add to understanding of how change happens and what needs to be done to promote it. Similarly, asking more critical questions about the sources of particular problems can provide important insights into how they can most effectively be addressed and can narrow down the conditions for good enough governance. Clientelism, for example, is widely decried as an impediment to poverty reduction and a constraint on the ability of poor people to be heard, treated equitably in their encounters with government, or incorporated effectively into political parties, unions, and other political organizations. It is criticized as characteristic of a culture of personalism that needs to be changed in order to control corruption. Frequently, efforts to eradicate it focus on civil service reform, public expenditure management, and project monitoring systems. Yet clientelism generally occurs in political systems of incomplete legitimacy and economic systems in which jobs are scarce and where political elites are able to trade access to jobs or other resources for political support. As such, efforts to increase political legitimacy or to make jobs more widely available might do more to eradicate widespread clientelism than civil service reform, public expenditure management, or a variety of monitoring systems.
This is suggestive of how asking different questions provides new perspectives on old issues and might indicate different ways to remedy governance deficits.
Thinking Strategically about Priorities. Clearly, much of the above has to do with setting priorities among governance reforms. Priorities can be better set if there is more understanding about which actions produce more results in terms of efficiency, effectiveness, and responsiveness, which produce the most benefit for the poor, which logically precede others, which are easier to undertake or produce results in the short term, and under what conditions particular reforms are likely to have the most impact.
While assessing the administrative and fiscal capacity of countries to carry through on various commitments is important, setting priorities for actions requires a broader frame of reference.24
Of course, priorities necessarily differ among countries. It is clear that establishing basic political legitimacy and order is an essential first step for countries with collapsed states. This task inevitably must take precedence over other important, but less urgent, reforms such as public expenditure management, civil service reform, and decentralization of public services. Thus, some countries—Afghanistan and Sierra Leone

17

come to mind—are in need of basic institutions to ensure a modicum of political stability, basic physical protection of citizens, and initiatives that increase the legitimacy and authoritativeness of government, laws, and public policies. Other countries—Nicaragua,
Burkina Faso, and Honduras come to mind—can be assumed to have enough institutional coherence that they can begin to think more about expanding public services to their poor majorities, diminishing the most development-averse forms of corruption, and setting up systems for better management of public resources. Yet other countries—India,
Thailand, Brazil, South Africa, and Mexico come to mind—are in position to undertake more difficult governance reforms such as putting in place transparent budgeting and accounting processes, regulatory frameworks, and risk mitigation systems for the poor.
Countries will also differ in terms of the severity of particular problems. In some situations, for example, the misuse of government resources may be so grave that other reforms cannot be achieved until it is reigned in. An important adage for considering what kinds of changes are needed and what priority they should have is to “begin where the country is” in terms of its governance capacities and feasible actions to improve them.25 The task of setting priorities is inherently political. Priorities for attacking governance deficits differ among distinct reform constituencies. Donors, for example, often emphasize the efficient management of public resources and clarity about the uses of those resources. Politicians are likely to be more concerned about the political implications of different kinds of reforms and might give priority to those that increase satisfaction among their supporters. Economic elites might place priorities on changes that improve their capacity to survive, to generate profits, to eliminate what they see as troublesome government bureaucracies, or to improve the security with which they do business. Poor people, in contrast, might give priority to the availability of services and basic conditions of personal security. None of these groups is homogeneous, adding to the difficulty of specifying what should be undertaken and what not. Poor rural people without access to land may give priority to some changes that conflict with important concerns of those who have land or of poor people in the urban informal sector.
Thus, conflicts about priorities are inevitable, and the more open a political system is, the more likely it is that such conflicts will be publicly aired and debated.
While there may be some overlap among interests—economic elites and poor people may both be concerned about public security, for example—there is likely to be much contention about which are most important and which should be given more emphasis.
Sorting out priorities from the perspective of different interests is a political process, and one that cannot be short-circuited by technical analysis or donor fiat.
There are also differential political payoffs to undertaking governance reforms.
Producing tangible benefits in the short term, for example, might be a wise first step in many countries, particularly those suffering from reform fatigue or weak legitimacy. For example, improved garbage collection in poor neighborhoods and providing more security in public markets are high visibility changes that can build citizen trust that government services are getting better. Similarly, more ambitious and politically difficult

18

reforms need to be assessed in terms of their political feasibility, given the context into which they will be introduced. The overall goal of better government performance is not advanced when governments promise actions that are not politically or bureaucratically feasible or when they cannot deliver on their promises.
The politics of reform can also introduce difficult tradeoffs among priority areas.
Policy reform in Bolivia is a good example. Beginning in 1985, successive policy reforms were made possible by governing pacts established in the aftermath of inconclusive elections, when choices about control over the presidency were thrown into the congress and alliances were cemented among parties about supporting particular candidates.26 With varying degrees of success, these pacts usually lasted through a presidential term and provided the basis for legislative approval of important and highly contentious reform policies. At the same time, however, a number of efforts to reform
Bolivia’s public sector made only sporadic progress. This was because the pacts were based on the exchange of government patronage for political support in advancing the reform agenda. Thus, the cost of widespread policy reform was to allow the old politics of patronage to continue in the public sector. This experience suggests that public sector reformers may thus be faced with difficult tradeoffs and dilemma—might energetic pursuit of specific reforms limit the government’s capacity to promote other important reform goals?
Setting priorities, therefore, is a multidimensional exercise, involving discussions of more technical issues such as sequences and degrees of administrative difficulty as well as tradeoffs among country-specific needs and capacities, interests, and political benefits. Discussions about priorities also highlight again the need to consider tradeoffs in the commitment of resources, energy, and political capital—governance reforms are important, but are they as important to reducing poverty as investing in rural development, infrastructure, better quality education, expanded health care for particularly vulnerable groups, more employment opportunities, and so on? Good answers to these kinds of question do not yet exist.
Who Needs to Do It? Most of the good governance agenda is about what governments need to do to perform better. It is also governments that are asked to prepare the PRSPs and, understandably, they respond with commitments to put their political, administrative, and financial houses in better order. At the same time, many of the poorest governments not only have low capacity to carry out these commitments, they may be captured by corrupt elites, have a history of not complying with their commitments, lack even basic legitimacy in the eyes of many of their citizens, or be locked in conflicts that consume their energies and resources. Such countries raise many questions about the viability of the commitments made in the PRSPs. Are there ways of improving governance when governments are unable or unwilling to make necessary changes in how they carry out their activities? Even when governments have more capacity, legitimacy, and commitment to poverty reduction, are there ways to reduce the burden on their managerial and administrative systems?

19

Development practice in recent years has produced a wide range of innovative ideas about how to improve service delivery through community engagement, contracting out, various forms of delegation, and privatization; how to engage communities and municipalities in development planning and budget management; how to help NGOs take on activities that governments are unable or unwilling to provide; and how to find other surrogates for government in such situations. Fox, for example, has suggested ways in which citizens can monitor and evaluate services, reducing the burden on governments of developing full-blown monitoring and evaluation systems and increasing the extent to which citizens are engaged in governance reforms.27 Similarly, Tendler provides examples of how the availability of better information about programs and client rights helped discipline community level health workers with limited investment in administrative measures to achieve the same result.28 Certainly, many governments have continued to commit themselves to guaranteeing certain kinds of basic services, but have often determined that they no longer have to be exclusive providers of them.
Many such innovations are well-known and have been effectively adapted in a variety of contexts. It is interesting, however, how rarely PRSPs draw on this experience or consider adopting such innovations. Some of the alternatives to government, of course, are politically threatening—empowering citizens to monitor government, for example, or allowing NGOs to deliver sensitive public services. Nevertheless, in the search for ways to make the governance agenda more manageable, and facing the reality of governments of very limited capacity, alternatives to government action and provisioning can be useful. There is room for opening up the question of who needs to do it and revisiting the commitments made in the PRSPs. At the same time, however, it is important to question the extent to which these kinds of innovations provide short-term responses to serious governance deficits, but may not provide long-term solutions to them
(see Box 2).29
Box 2
Conclusion: An Agenda for Good Enough Governance. It is understandable that many PRSPs were found wanting in the discussion of governance reforms—the state of knowledge about them remains limited. Much can be done through research and strategic analysis to make the good governance agenda less overwhelming for poor countries, as the summary in Table 4 suggests. This kind of work can help sort out the essential from the less so, the primary from the secondary, the short term from the long term, and the high priority from the less pressing. Considerable analysis can be done at a general level, but ultimately, setting priorities and developing strategies for improving governance in pursuit of poverty reduction must be determined on a country-by-country basis. Table 4
Inevitably, this kind of analysis will reveal the extent to which governance reforms take place in the midst of conflict, confusion, cross-purposes, inefficiencies, and learning by doing. Currently, many governments that have prepared PRSPs have

20

committed themselves to a range of changes that can easily outpace their capacity to manage reform and the conflicts it produces. Certainly there is much that needs to be done in most countries and the poorer they are, the more likely they are to require extensive change before their governments work well. But it is unlikely that much can be accomplished when such countries are overloaded with commitments to change large numbers of conditions at the same time. From this perspective, it is better to assess capacities and feasibility more carefully, target fewer changes, and work toward good enough rather than ideal conditions of governance.

Getting on with Good Enough
Setting out to get better governance is not a new undertaking for many countries.
A significant number can point to long-standing efforts to improve government capacity to perform with greater efficiency, effectiveness, and responsiveness. In addition, the past two decades have been a watershed of policy and institutional reform in many developing countries. Such changes, at least in some countries, contributed to progress toward good enough governance. Indeed, a survey of expert opinion in 20 countries in
2000/2001 revealed that even while perspectives remained very critical of government performance, many acknowledged that improvements had been made in the preceding five years (see Table 5).
Table 5
The experience of countries that have introduced and sustained policy and institutional reforms has much to offer in terms of lessons about how change happens
(see Appendix). In moving beyond concern about the length of the agenda to thinking about how reforms are introduced and sustained in the real world, this experience provides some guidance about activities that promote good enough governance and the link between more capable governments and more capable civil societies.
More Capable Governments. In the past two decades, reformers in government, development professionals, academics, and donor organizations have focused great attention and resources on improving government performance.30 Among 99 low- and middle-income countries included in a survey carried out by Elaine Kamarck, for example, 24 had announced civil service reforms by late 1999 (see Table 6).31 Sixtythree countries had carried out privatizations to deal with large public sector deficits, reduce the public sector wage bill, and improve the performance of state enterprises and service delivery agencies. Just as assiduously, a large number of countries restructured the relationship between central, provincial, and local governments.32 Technical cooperation for capacity development directed to poor countries from international development agencies was significant throughout the 1990s, reaching $20 billion in 1995, and totaling over $170 billion by the end of the decade.33
Table 6

21

Most of these public sector reforms can be described as first generation initiatives to address performance issues. As with governance issues more generally, these reforms were characteristically guided by a set of important questions: Why do public servants behave badly? Why do public organizations carry out their functions poorly? Why is there corruption and lack of efficiency? In asking such questions, reformers arrived at reasonable answers: Because public sector personnel are badly paid; because incentives are perverse; because clientelism rather than merit prevails; because structures, rules, and accountability are inadequate.
Given such responses, it is not surprising that the resulting reforms focused on bringing order to disordered structures, controlling the activities of public sector workers, keeping them from doing harm, and ensuring that organizations do what is expected of them. These are important considerations in efforts to improve the public sector, particularly when most citizens are poor and vulnerable to the misbehavior of the officials and organizations that represent the state. Nevertheless, while these first generation reforms may be important foundations for more efficiency, effectiveness, or responsiveness in government, they do not necessarily produce these conditions. These kinds of reforms, essential as they are, can even lead to rigidity and aloofness unless public sectors move on to second generation reforms.
Considerable experience in the management of private sector and nongovernmental organizations, and the evidence of “islands of excellence” in the public sector are suggestive of second generation issues that need to be addressed once basic structures and systems are in place. Second generation public sector reforms respond to a different set of questions than those that characterized the first generation of changes:
Why do public servants perform well? Why are they committed and energetic in pursuing the public interest? Why are public organizations efficient, effective, and responsive? Why do public servants resist opportunities for corruption?34 By asking such questions, public sector reformers are directed to a different set of factors than emerge when asking why public officials act badly: Because they are committed to the missions and norms of their organizations; because organizations have mystiques that motivate their workers; because public servants have meaningful jobs and believe they are involved in finding solutions to important problems; because they gain approval for what they do well; because public employees have opportunities to work in teams with others who share their commitments.35 Second generation public sector reforms address how government performance can be improved through such changes as building managerial capacities and developing positive organizational cultures.
Structures, systems, and rules are important, of course, and little progress in improving performance can be sustained in the absence of such changes. Beyond certain basic conditions, however, many organizational specialists recommend “loosening up” rather than “tightening up” on organizational rules and procedures. Current approaches to management in both the public and private sector in developed countries recommend more horizontal and less vertical organizations, greater autonomy for organizational units to make decisions and carry out activities, and increased flexibility in how routine operations are carried out.36 Organizations need to have clear performance standards, but

22

with “room to maneuver” in problem-solving and sufficient control over the decisions that allow them to achieve the results for which they are held accountable. Equally important are meaningful jobs, commitment to missions, positive responses to jobs well done, teamwork, participation in finding solutions to important problems, loyalty, respect, and managers who recognize and reward excellence. Second generation reforms have produced results even when first generation changes are lacking. They are characteristics of organizations regarded as “islands of excellence” in the midst of what are often broken and incapable governments.37
Of course, second generation public sector reforms are difficult, require considerable investment in developing management and leadership skills, and are usually most effectively addressed unit-by-unit and organization-by-organization. Nevertheless, in many countries, they may be the kinds of changes required to move additional steps toward more efficient, effective, and responsive governments. In the spirit of managing the agenda for good enough governance, the creation of more capable governments can be approached with distinct strategies. Where appropriate, first generation reforms might be undertaken with the expectation that they need to be followed, in time, by the kinds of changes suggested by second-generation reforms. Governments that have already moved toward the adoption of first generation reforms might be encouraged to take on the managerial challenges of the next generation. Or, where governments prove to be averse to significant public sector reform, for whatever reasons, changes might go ahead on a target of opportunity basis in the expectation that some islands of excellence can be created. More Capable Civil Societies. Even when it is possible to move on to second generation public sector reforms, organizations that demonstrate the capacity to improve are likely to remain islands of excellence unless civil societies are also strengthened.38
An important incentive for organizations and officials alike is the capacity of citizens and groups to demand fair treatment, to have information about their rights vis-à-vis government, and to be able to hold officials and governments accountable for their actions. Thus, while the supply of good governance is important, so too is demand for it.
As indicated in the first part of this paper, the PRSP process recognized the importance of including discussion with civil society groups in the formulation of country plans. Some countries took this seriously and set up innovative forms of consultation.39
In some cases, governments listened to feedback from various constituencies and amended their plans according to the information they received.40 These are positive achievements. There is reason to be concerned, however, that the PRSP process may have by-passed politics in some countries and established forms of representation that do not necessarily advance the interests of the poor over the longer term.
The last twenty years have seen extraordinary growth of civil society organizations and often clear signs that they are increasing their capacity to interact effectively with government, to organize political pressure to gain attention for their demands, to abandon clientelistic relations in favor of negotiation with government, and to bring to light instances of public malfeasance, ineffective services, and lack of

23

responsiveness. Frequently aided by organizations representing international civil society as well as the media, they have at times been able to join in debates about policy and to raise important criticisms of widely adopted market reforms. Increasingly, civil society groups have organized to denounce corruption and demand basic honesty from government. Ultimately, this kind of counterpoint to government is the essential incentive that governments need to make them more accountable for their actions. Faced with increased pressure from opposition political parties, unionized workers, NGOs, ethnic and regional associations, interest groups, and other organizations, some governments have improved some aspects of governance.
Countries vary in the strength and vitality of their civil societies, and governments differ in the extent to which they encourage, control, or repress citizens and the groups that represent them. Citizens have varied opportunities for participating in political debates and decisions. In most countries, however—even those with wellinstitutionalized democratic institutions—the poor are usually the least organized and have the most difficulty representing their concerns to government and pressing for response to those concerns. In most countries, meaningful voice for the poor in political decision-making is particularly challenging. An important task, then, is to facilitate the political mobilization of poor constituencies, whether these are defined by income category, gender, ethnicity, region, occupation, or other organizing principle.
The participation of the poor in civil society means the ability to mobilize political power and use it in expressing needs, demanding response, and holding officials and individuals accountable for how well they respond. Yet, interests among the poor are likely to vary—rural and urban divides, contention about restructuring gender roles, regional and religious conflicts, competing ethnic identities, and occupational distinctions are among the politically-relevant divisions often found among the poor. In most situations, then, the poor seek political power around concerns that are more specific than that of the general condition of poverty. Further, conditions of illiteracy, intimidation by powerful interests, clientelism, repression, and other factors often mean that the interests of the poor are represented by proxy, frequently by NGOs that define their missions as poverty reduction, empowerment, or the particular needs of categories of poor people.
Although a useful short-term way to increase awareness of the needs of the poor in political decision-making, representation by proxy has limitations. NGOs usually define their own missions and how to achieve them; the poor are often marginal to this process.41 NGOs tend to be wary of partisan politics and thus often shun interactions with political parties and interest groups that are more connected to national political processes and debates. NGOs tend to rely on outside sources of support, not on the poor, and their objectives and actions can be affected by such dependency relations. Clearly,
NGOs play an important role in policy debates, and can be particularly valuable partners for the poor in repressive regimes, but building strong civil societies also means developing the capacities of political parties, labor organizations, autonomous think tanks, independent media, and a variety of interest groups.

24

Strengthening the demand side of good enough governance, therefore, requires considerable tolerance for politics, including public contestation of different interests and the creation of winners and losers in those conflicts. Table 7 sets out a hierarchy of ways in which civil societies interact with governments around public policy issues and national development strategies, such as those outlined in PRSPs. At the least political end of a participation scale is the provision of information, often a one-sided initiative taken on by government ministries and agencies and political leaders. These initiatives do provide some benefits. With greater access to information, citizens become better informed about government programs and policies and their rights to particular kinds of services or particular kinds of behavior in their encounters with public officials and organizations. In most cases, however, those who benefit most from efforts to provide information are likely to be those who live in urban areas—where communications are easiest—and those who have more education and more access to government services— the better off, in fact.
Table 7
Many governments undertaking the participatory process recommended for the
PRSPs sought more than the diffusion of information. In most cases, they set up processes for consultation and consensus building around the documents, as indicated earlier.42 In these activities, government ministries and public agencies generally took the lead; they invited particular groups such as NGOs, interest groups, and religious institutions to be present in fora where the reports were discussed. Although governments remained in control of who was involved in these discussions, these consultations indicated concern that those representing the poor take a more active role in the process than was the case with more information-based approaches. The benefits to these kinds of consensus building exercises can be broad agreement on goals for national development and getting particular kinds of reforms on the political agenda, as well as better-informed publics. They do not usually lead to sustained participation by the poor in policy decision making, however, nor do they easily translate into greater capacity of the poor to make demands on government.
These two modes of interaction are relatively simple and easily controlled from the top. More difficult, but of potentially greater impact in terms of how governments and civil societies interact, is public debate. Public debates about the national goals and the contents of reform initiatives involve a diversity of actors. Entry into the discussion is not easily controlled by governments. Political debates take place in a variety of arenas—the media, legislatures, universities, even the streets. Such debates can move the policy reform agenda toward deeper political commitments to pursue particular actions and create mandates for reform. It can also put a reform agenda at risk when those opposed to change are well organized and vocal and those who might benefit from it— such as particular groups among the poor—are not effectively represented or organized.
Beyond consultation, consensus building, and public discussions are activities geared toward building coalitions in support of important reforms.43 Political leaders, parties, particular interest groups, movements, unions, and others are critical actors in

25

putting together such coalitions. These actors engage in discussions over tradeoffs, negotiate over divergent interests, and debate about the inclusion and exclusion of particular interests. While building coalitions is more difficult than consensus building, consultation, and political debate, the payoffs can also be greater. It is possible, for example, to define the actions to be undertaken with greater specificity, increase the political weight of those in favor of change and thus enhance the chances that reforms will be put in practice, and hammer out compromises about what should be undertaken.
As indicated, efforts to build coalitions for reform are also more likely to produce winners and losers—the more specific the actions agreed upon, the more apparent their costs and benefits for particular groups.
In the PRSP process, governments generally provided information and sought consensus rather than encourage debate and contention over the goals of national development and debt relief. In some countries, governments were aided by the absence of strongly organized and programmatic parties, the lack of groups to represent the interests of poor constituencies, or well-developed institutions such as the media and legislatures where such debates can be carried out. Indeed, efforts to build stronger civil societies through the PRSP process are likely to have limited impact. Because governments are those responsible for preparing the documents, they tend to be in the driver’s seat in determining a process for discussion about them.44 In general, those who participate tend to be proxies for the poor rather than the poor themselves. While this limited form of engagement can provide some short-term support for poverty alleviation strategies, ultimately, effective participation in improving governance must engage political interests and must involve debates about tradeoffs and contention between winners and losers. However much governments and donors might want to shy away from it, policy reform is inherently political and contentious. The role of civil society in governance reforms is often to create more contention around decisions and actions taken by government, but also to provide more incentives for governments to improve their ability to produce results.

Strategic Goals for Promoting Good Enough Governance.
Getting good governance that contributes to poverty reduction is extraordinarily difficult. Even getting good enough governance is fraught with ambiguities, challenges, and the potential for failure and less-than-anticipated results. Good—or good enough— governance is a long-term objective, and efforts to achieve it will often be halting and reversible. In this paper, I have reviewed the strengths and weaknesses of the PRSPs and the process through which they were developed and I have raised a series of issues aimed at making the good governance agenda more manageable. In addition, I have indicated cautionary notes about remedies for bad governance, suggested activities that might advance the cause of better performance, and emphasized the political nature of policy and institutional reform. The perspectives of the paper can be summarized in a series of recommendations for promoting good enough government.
Reformulate the Objective. Good governance is a laudable goal. Strong and compelling reasons can be advanced to demonstrate how important it is to the
26

development process in general and the reduction of poverty in particular. Yet, for many countries, and particularly for the poorest, good governance is a distant possibility. More to the point, in being encouraged to seek this laudable goal, countries that suffer from weak or nonexistent institutions, inefficient organizations, and poorly developed human resources are likely to be overwhelmed by all the things that “must be done” to achieve it.
It makes sense to find ways to reduce this extensive agenda by reformulating the objective of reform activities to be that of good enough governance. Currently, of course, it is not clear what good enough means. Research, analysis, and country-specific assessments can help develop the concept of good enough and can be useful in programming activities and assistance packages that are more feasible and geared to country-specific conditions.
Make the Connection to Poverty Reduction. If the principle goal of the PRSP process and the reforms that come in its wake is direct impact on the extent and depth of poverty in a country, then governance reforms need to be assessed in terms of their contributions to this goal. As indicated, good governance is important for all countries and certainly is a condition that ought to improve as countries become more developed.
In the short and medium term, however, some governance reforms may be less important than others in affecting poverty. Sorting out what’s critical and what’s less so in affecting the life chances of the poor majority can save scarce energy, resources, and political capital as initiatives to reduce poverty are undertaken. It is an activity that requires research and critical analysis.
Consult Longitude as well as Latitude. Cross-national analyses have helped generate important insights about governance; advocacy about particular governance deficits has been important in spurring concern about the link between government performance and poverty reduction. Yet much analysis and advocacy has contributed to the additive nature of the good governance agenda. Historical analysis and country case studies can provide insights about reducing the list of things that “must be done” as well as suggest sequences for putting governance reforms in place. Highly indebted poor countries are not the first to face problems of governance and poverty. More attention to the historical experience and to the lessons that can be drawn from specific countries or groups of countries as they faced up to governance deficits would help clarify good enough governance.
Learn from What’s Working (Well Enough). The good governance agenda has been largely developed by assessing what’s not working or what’s working imperfectly.
Given the very large number of things that don’t work particularly well—or don’t work at all—in very poor countries, it is not surprising that the to-do list is long and growing longer. Yet in almost all countries, even the most destitute, some activities of government work better than most.45 Consulting this experience can provide valuable lessons about why this is the case, what factors make for better (even if not good) performance, and what needs to be changed for progress to occur. Such lessons can suggest the types of reforms that have better than average chances of making a difference, the kinds of conditions that surround more successful activities, and the specific ingredients important to efforts to improve performance.

27

Assess Priorities Strategically. Setting priorities for good enough governance is extremely important, yet is extremely difficult because it means sorting out activities across a series of criteria. It involves discriminating the short term from the longer term, sequences and hierarchies of reform activities, feasibility and capacity, and political as well as efficiency impacts. Priorities will certainly differ by country, and even by political administrations within countries. Efforts to define priorities will also undoubtedly generate conflicts. Despite these difficulties, determining a hierarchy of activities is essential if any progress is to be made toward good enough governance.
Think of Alternatives, but Remember the Public Sector. The prospect for governance reforms aimed at poverty reduction promoted by governments in some countries may be particularly dim. Where countries are controlled by groups uninterested in either good government or poverty reduction, where the obstacles to change are overwhelming, where capacity is unusually low, or where conflict, corruption, or other factors rob governments of the ability to reform, there may be alternative ways to ensure basic services and the capacity of communities to protect themselves and survive. Even where governments are more committed to change, there may be alternatives to traditional methods of improving governance. The past two decades of innovative experiments with new ways of getting the business of government done can provide a range of ideas to facilitate some change, even in very hostile environments. At the same time, many such alternatives provide only short-term solutions. Ultimately, good enough governance has to involve governments and ensure the regular provision of basic public goods, such as order, security, and legitimate authority. That some governments are currently unable to provide these conditions is a good part of the challenge of encouraging the development of ones that can in the future.
Consider Structures, Systems, and Rules, but Move Beyond Them.
Governments are currently under great pressure to increase their capacity to take on difficult development challenges such as the reduction of poverty. Efforts to do this have generally focused on first generation reforms that seek to create or strengthen the structures, systems, and rules that serve as a backbone for government activities. Yet such efforts are often not enough to ensure better performance. A second generation of reforms that focuses on management, leadership, and organizational behavior, culture, and mission is critically important. In best-case scenarios, managerial and organizational reforms complement first generation reforms; in other cases, they can be pursued as part of target of opportunity strategies.
Consider Civil Society. Governments have been the primary focus of governance reforms. They are also important actors in determining what needs to be done to improve their performance. Yet governance cannot be left solely to governments. As civil societies develop and increase their potential to hold politicians, agencies, and governments more generally accountable, performance is likely to improve. Thus, enhancing the capacity of groups in civil society to represent their interests, make demands on government, and hold its agents accountable is an important way to achieve the longer term goal of better public sector performance. Too often, however, efforts to

28

strengthen civil society attempt to avoid politics. The activities that effective civil society organizations engage in are about power and efforts to increase their power vis-à-vis government. While much of the strengthening of such organizations must emerge from their own leadership cadres and strategies, they can be assisted in developing managerial and policy analytic skills, acquiring expertise in communications and lobbying, and gaining protection against the repressive acts of hostile governments.
Roles for donors. Many of the activities suggested above are ones that will engage donor agencies. There are also other activities for which donors are particularly well suited because of their expertise and the resources they command.
•

Central among activities that are particularly appropriate for donors, of course, is investing in more research about governance. This paper has suggested a variety of questions that are not well understood about governance, including those related its emergence over time, sequences of governance capacities, the roots of problems, and so on. Donor organizations commit major resources to research activities and are particularly well paced to contribute to building expertise in this area. In particular, they should resist the temptation simply to add to the list of reforms that must be undertaken and concentrate instead on discriminating among them. •

Donor agencies, given their concern about ownership of the development process, and given their expertise in research and analysis, are also particularly well situated to contribute to building analytic capacity within developing country governments and within civil society organizations about governance issues. In particular, contributing to the capacity of legislatures, the media, NGOs, interest groups, unions, and political organizations to analyze policies and governance issues contributes not only to the fund of knowledge about the importance of government performance, and to the generation of alternatives, but also to the capacity to engage in discussions and debates about national policies and institutional innovations.

•

Donor agencies have played an important role in setting the good governance agenda and in providing advice and support to governments that undertake reforms to promote better performance. Frequently, however, these same agencies have added to the length of the agenda and increased the number of projects and programs that governments pursue, without at the same time providing much input about capacity requirements, feasibility, or the potential for system overload. Donor agencies can do much to focus government attention on critically important yet feasible reforms, at the same time they reduce the overall number of initiatives undertaken if these are making it difficult for governments to manage them.

•

Donor agencies can do much to tailor governance initiatives to conditions and capacities in individual countries. All too often, donors have been accused of promoting “one size fits all” reforms. Considerable experience suggests that

29

reforms need to take account of the specific mix of capacities and resources of each country and that there may be a variety of ways to approach a given problem such as poor performance by government. Donors, with an eye to the comparative experience of many countries, are in a good position to understand the importance of tailoring reforms to that experience.
None of these recommendations will resolve the governance problem. However, governance reformers and donor agencies can do much to set reasonable standards for achievements toward governance reforms by ensuring that standards are in line with capacities and time frames. Donors cannot guarantee that better governance will emerge from the PRSP process or from any other mechanism—including conditionality—that can be developed to influence the practices of governments in poor countries. They and their reformist allies, however, can help move the process along by contributing to a reasonable agenda, setting reasonable standards for accomplishments, and rewarding movement in the right direction.

30

Table 1
The Good Governance Agenda
1997

1998 1999/ 2000/ 2001/ 2002/
2000 2001 2002 2003

Characteristics of good governance:
Administrative capacity

PP

Checks and balances
Coordinated policy, policy making system
Credibility
Decentralization
Effective/responsive institutions
Effective rules/constraints/monitoring
Efficient/equitable/independent judicial system Environmental protection
Free press
Foundation of law
Gender/racial/class/intergenerational
equity
Good investment climate
Incentives for public officials
Institutions for coordination/accountability
Investment in basic social services and infrastructure Learning/innovation institutions
Local capacity
Macroeconomic stability
Non-distortionary policy environment
Participatory political processes/political pluralism/democracy Political stability/conflict management
Poverty reduction
Property rights
Sound regulatory system
Strong and capable state
Transparency/Information

31

To get good governance, countries need:
Institutions for:
Appropriate technology
Agricultural research/extension
Bank and finance regulation
Central banking
Civil service
Conflict resolution
Debt management
Democracy
Effective markets
Effective service delivery
Environmental/natural resource management Inclusive markets
Local development
Local governance
Local problem-solving
Lowering transaction costs
Managing decentralization
Participation
Information and technology for the poor
Local demand
Local decision making
Private sector governance
Representation
Regulation
Risk assessment/mitigation
Socially responsible behavior
Sustainable development
Telecommunications regulation
Training
Transparent budgeting/accounting/disclosure 32

Violence prevention
Laws about:
Accounting/auditing/disclosure standards
Anti-corruption
Trademark protection
Enforcement of contracts
Penalties for dishonesty and fraud
Anti-discrimination
Bankruptcy
Biodiversity
Commercial law
Competition
Contracts
Environmental degradation
Foreign investment
Gender/group equity
Human rights
Information
Labor standards
Monopolies
Pensions/social assistance
Property rights
Intellectual property rights
Quality of goods and services
Policies for:
Agriculture
Land reform/land policy
Rural credit/infrastructure
Agricultural technology
Banking
Biodiversity
Capital markets
Community development

33

Community planning/management of forests Competition
Corporate governance
Disability
Downsizing bureaucracy
Energy
Empowerment of women
Fiscal restraint
Family planning
Fisheries
Foreign investment
Forests
Industrial development
Information and knowledge
Domestic R&D
Information on rights/laws to poor
Insurance
Labor markets
Liberalized markets
Open trade regime
Privatization
Pro-poor fiscal adjustment/market failure for poor
Reducing vulnerability of poor
Regulation
Selective affirmative action
Social safety nets
Taxation
Urban development
Water basin management

Services for:
Education
Certification of producer/consumer skills

34

Girls’ education
Lifelong learning
Local accountability
Tertiary Education
Environmental protection
Health
Cost recovery
HIV/AIDS
Information
Safe drugs
Universal access to basic services
Vaccines/communicable diseases
Physical infrastructure
Communications
Cost recovery
Electricity
Finance capital
Housing
Land use planning
Public transportation
Roads
Safe water
Sanitation
Telecommunications
Poverty reduction
Credit
Job opportunities
Legal aid/legal information for poor
Microcredit/savings
Social funds
Social safety nets
Targeted transfers/subsidies
Strategies for:
Access to services for the poor

35

Alternative provision of social services/ infrastructure Asset creation for the poor
Building social capital among the poor
Capacity building in public sector
Capital markets
Community monitoring of services
Compensating losers in transitions
Comprehensive approach to attacking poverty Comprehensive development framework
Economic growth
Empowering the poor
Engaging NGO sector
Environmental protection
Focusing public action on social priorities
Foreign investment
Innovation
Interdependence of development strategies
International partnerships
Knowledge development
Prevention, preparation for, response to macroshocks Private sector development
Promoting equity
Reducing vulnerability of the poor
Redistributing investment toward poor
Technology
Voice for the poor
Source: World Development Reports, 1997-2002/2003

36

Table 2
Dimensions of Poverty and Governance
Poverty

Governance Issues

Empowering the poor Rules for seeking and holding public office
•Fair, transparent national electoral processes
•Power-sharing arrangements to ensure stability in heterogeneous societies
Oversight by political principals
•Parliamentary oversight with independent audit institutions
•Budget that is credible signal of government policy intentions
•Pro-poor policies
•Sound institutions for local and national representation
Improving
Adequate, predictable resources fro sectors, local authorities coverage, •Pro-poor budget priorities for service provision efficiency, and
•Stable intergovernmental transfers with hard budget constraints sustainability of
•Hierarchical and transparent budgeting processes basic services
Demarcation of responsibilities for delivery
•Assignment of responsibilities according to subsidiarity principle
Capable and motivated civil servants
•Merit-based recruitment and competitive pay
•Hiring to fill real needs, within a hard budget constraint
•Public service that earns respect
Accountability downwards
•Publication of accounts for local-level activities
•Dissemination of basic data on performance
•Mechanisms for client feedback, including report cards and client surveys
Flexible delivery
•Involvement of civic and private (for profit) partners
Development of local capacity
•Incentives to deploy staff to poor and remote areas
•Appropriate autonomy in deploying staff
Increasing access
Legal and regulatory framework to markets
•Enforcement of antidiscrimination legislation
•Incentives for deepening of credit and land markets
Methods for reducing exclusion
•Enforcement of legislation against barriers to entry
•Provision of information on labor and credit markets
Demarcation of responsibilities and budgeting procedures to support development and maintenance of infrastructure (e.g. rural roads) to enable physical access to markets
Providing security
Rules for sound economic management
•from economic
•Hard budget constraint for subnational and aggregate fiscal discipline shocks •Efficient administration of tax and customs
•Independent central bank to carry out monetary policy
Safeguards against economic vulnerability
•Recognition of property rights over physical assets
•Access to speedy social insurance and other services through hub-and-spoke arrangements
•From corruption,
Enforcement mechanisms crime, and violence •Independent and adequately funded court system
•Access to speedy recourse and redress
•Reliable and competent police
Efficient courts with competent judiciary and legal personnel
•Alternative mechanisms for dispute resolution
Source: World Bank, 2001: Ch. 7:9.

37

Table 3
Summary of Institutional Evolution in the Now Developed Countries
First
Adoption
Democracy
Male Suffrage
Universal Suffrage
Modern Bureaucracy

Majority
Adoption

Last
Adoption

United
Kingdom

United
States

1848 (France)
1907 (New Zealand)

1907
1946

1925 (Japan)
1971 (Switzerland)

1918
1928

1870
1965

Mid-1800s

Early 1900s

Early 19th Century

1930s?

Modern Judiciary
Intellectual Property Rights
Patent Law
Modern Patent L.
Mod. Copyright L.
Trademark Law
Corporate Governance
Institutions
Gen. Limited Liability
Bankruptcy Law
Modern Bankruptcy Law
Mod.Auditing/Disclosure
Competition Law
Effective Competition Law
Financial Institutions
Modern Banking
Central Banking
Modern Central Banking
Securities Regulation
Modern Sec. Regulation
Income Tax

1474 (Venice)
1836 (U.S.)

1840s
1960s

1912 (Netherlands)
1990s(Spain, Canada)

1623
1852

1793
1836
1891 (1988)

1862 (U.K.)

1862

1844 (Sweden)

1856 (1862)
1542
1849
1848
1919
1956

1800
1898
1933
1890
1914

Mid-1920s
1694
1844
1679
1939
1842

1913
1929
Mid-19th Century
1933
1913

1890 (U.S.)
1914 (U.S.)
Mid-1920s (U.K.)
1688 (Sweden)
1844 (U.K.)
1679 (U.K.)
1842 (U.K.)

1830s
1900s

1913 (U.S.)
1929 (U.S.)

Social Welfare and Labor
Institutions
Industrial Accident Ins.
Health Insurance
State Pension
Unemployment Insurance
Child Labor Regulation
Mod. Child Labor Reg.

1871 (Germany)
1883 (Germany)
1889 (Germany)
1905 (France)
1802 (U.K.)
1878 (U.K./Prussia)

1898
1911
1909
1920

1930 (U.S.,Canada)
Still absent in U.S.
1946 (Switzerland)
1945 (Australia)

1897
1911
1908
1911
1802
1978

1930
(Still absent)
1946
1935
1904
1935

Note: Institutions entered in italics denote “pre-modern” varieties, which fell so short of modern standards in terms of coverage and enforcement that they are usually better regarded as different categories from their “modern” descendants.
Source: Chang, 2000:Appendix. The original contains many notes that are not reproduced here.

39

Table 4
Reducing the Agenda
Strategy

Action

Actors

Assess payoffs to poverty alleviation Research to explore the impact of specific reforms on poverty alleviation and to compare the payoffs to alternative reforms

Donor agencies, scholars, think tanks, university research centers

Assess historical record of good enough governance in now developed countries and developing countries that have achieved good enough governance Research to explore sequencing and time dimension in reforms, crossnationally over time and country case studies

Donor agencies, scholars, think tanks, university research centers

Ask questions about what’s working, the roots of problems, the dynamics of change Research to explore country specific evidence of governance improvements over time and the characteristics that facilitated change Donor agencies, scholars, think tanks, university research centers, development practitioners, governance NGOs

Set priorities strategically

Assess capacities, interests, political benefits, tradeoffs.
Develop norms for distinct phases of governance development Donor agencies, development practitioners, governance NGOs, political organizations, politicians, civil society groups, surrogates for the poor

Assess responsibility for action Consult innovations for alternative mechanisms for providing good enough governance, assess political tradeoffs among reforms

Donor agencies, development practitioners, NGOs, scholars, politicians, civil society groups

Table 5
Perceptions of Government Performance
2000/2001
Governance
Principles
Participation

Rating of performance “five years ago”
2.82

Rating of current performance Percentage Change

3.12

10.06

Fairness

2.71

2.86

5.5

Decency

2.82

3.10

9.9

Accountability

2.52

2.70

7.1

Transparency

2.66

2.87

7.8

Efficiency

2.77

2.93

5.7

Respondents in 20 countries were asked to rate government performance along several dimensions of governance at the current time (2000/2001) and “five years ago.” Scores range from 1 (very low) to 5 (very high).
Source: Court and Hyden, 2001:22.

42

Table 6
Public Sector Reforms in Poor Countries
1980-1999
Type of Reform
Civil Service Reform
Downsizing
Financial Management/Budgetary
Reform
Privatization
Regulatory Reform
Decentralization
Countries undertaking 2 or more reforms * n = 99
Source: Kamark, 2000:database.

Number of Countries
24
31
22
63
20
39
18

43

Table 7
Participation and Politics
Form of
Participation

Principal
Actors

Political
Dynamic

Potential
Benefits/Costs

Information

Government ministries, public agencies, political leaders Initiated and managed from top down

Better informed publics; citizens with greater awareness of rights, responsibilities. Those most likely to benefit are urban and better off citizens Consultation and consensus building

Government ministries and agencies, political leaders,
NGOs and other civil society groups

Initiated and managed from top down.
Some input from leadership of NGO and civil society groups Better informed publics; citizens with greater awareness of rights, responsibilities; broad agreement on principles for action; agenda setting.
Those most likely to be engaged are those
“approved” by government

Political debate

Government ministries and agencies, political leaders and other politicians, political parties, interest groups, NGOs, and other civil society groups, media

Diverse forms of initiation, often through government announcement of policy direction, leadership of organizations, media

Better informed publics; citizens with greater engagement in setting priorities for public action; awareness of winners and losers, conflict over goals, content of decisions

Coalition building

Government ministries and agencies, political leaders and other politicians, political parties, interest groups, NGOs and other civil society groups, media

Usually initiated by those advocating reform, responded to by those opposing reform, leadership of organizations, media

Better informed publics; citizens with greater engagement in determining costs and benefits of change; definition of specific goals; creation of political weight to advance/hinder change; negotiations and explicit tradeoffs; definition of winners and losers, heightened conflict

44

Box 1
Are Expectations Too High for Decentralization?
Decentralization has been popular reform throughout the world in recent years.
There are, of course, numerous ways to decentralize and numerous sequences through which it can be achieved. Recent experience suggests that the payoffs to decentralization reforms are highly contingent.† For example, the claim that decentralization can bring more efficient, responsive, and accountable government is based on the assumption that local officials will use resources wisely and that local citizens can monitor their behavior effectively. But resources can be captured by local elites, local officials can have bad judgment or little understanding about how to allocate resources effectively, and local bureaucrats can have so little administrative capacity that resources are wasted. Powerful local interests can be as corrupt as national interests. Expected benefits such as accountability and transparency depend very much on the quality of local government and the quality of local civil society. In cases where that pre-existing quality is low, public services can even decline with decentralization.
Moreover, decentralization can increase disparities among regions. In poorer regions, the capacity of local government to use resources wisely may be weaker than in richer regions. Similarly, citizens in poorer areas may be less mobilized and less able to make effective demands for efficiency and responsiveness. There is a special danger where financing for public services is decentralized; poorer regions have a lower capacity to mobilize local resources than do richer regions. Moreover, transfers to local and regional governments may add to fiscal deficits for the central government while diminishing its capacity to control expenditures. Newly empowered local and regional governments can also contribute to the debt burdens of the central state.
Decentralization can have many positive benefits, but by itself, it does not necessarily result in better governance or poverty reduction. Decentralization needs to be approached as a long-term process that requires strengthening of local government and local civil society. It needs to include equity-enhancing mechanisms to protect poorer regions and poorer people. Adequate resources, well-trained local officials, and informed and well-organized civil societies may be critical to whether decentralization provides benefits or simply results in continued inefficiency, ineffectiveness, lack of accountability, and unresponsiveness.

†

Burki, Perry, and Dillinger, 1999; World Bank, 2001.

45

Box 2
Are There Alternatives to Public Sector Service Delivery?
Much has been written about the problems of public sector organizations whose task is to deliver basic services. Often, these organizations are held in very low esteem. They are routinely criticized for being highly centralized and overly bureaucratized, and anecdotes are rife about simple problems that cannot be solved because they have to be referred to central headquarters, assessed through the lenses of hundreds of irrelevant rules and regulations, and signed by dozens of officials and departments. It is frequently charged that ministries have been captured by public sector unions that control personnel and resource allocation decisions. In such cases, the tenure of ministers and vice-ministers is often brief and their power is routinely contested by union officials and those who control internal bureaucratic empires. Along with strong criticism of their efficiency and commitment to the public good come equally strong assertions that public sector organizations are corrupt, that they pay salaries to phantom employees or to those who “parachute in” on payday, that they provide contracts to favored vendors, and that their budgets are used to line the pockets of influential politicians.
As a consequence, efforts to improve service delivery often seek ways to bypass traditional bureaucracies. The creation of special agencies or implementation units, for example, allows for the recruitment of professionals at higher salaries and the provision of inputs such as computers, vehicles, and training that are not available to sector ministries or their staffs. Social development funds established outside line ministries can provide quick, flexible, and demanddriven services. Decentralization, privatization, and contracting out of service provision have also been widely adopted, many NGOs have been enlisted to deliver essential services, and the private sector has been brought into the delivery of some services.
Defenders of these approaches argue that needs are so pressing and line ministries so inefficient that it is impossible for them to respond adequately; it is better to circumvent the bureaucracy and get directly to the actions that respond to the dire problems that beset so many citizens. Understandable as such sentiments are, it is important to assess the long-term consequences of many alternatives to traditional forms of service delivery. Experience to date suggests that when administrations change, special agencies and units are shut down or left to wither for lack of new resources. NGO management capacity can be limited when services need to expand, they may have favored clienteles, and they may resist downward accountability.
Decentralization does not necessarily mean better performance. Privatization can mean extensive political backlash.
Frustrated reformers might well ask what alternative there is to working with line ministries that are inefficient, ineffective, unaccountable, and unresponsive. While some innovations may provide long-term solutions to these problems, any alternatives simply avoid confronting the real issue—governments and citizens need to be able to count on public organizations that perform well. These organizations control budgets, standards and regulations, critically important personnel such as teachers, nurses, doctors, and civil engineers, extensive infrastructure, assignments to fill service delivery positions, and at times the capacity to block initiatives they do not agree with. In reality, there are limited long-term alternatives to the long, hard job of organizational reform in government.

46

Notes
1

Narayan (2000:100) quoting a citizen of Armenia, 1995

2

I understand governance to consist of the distribution of power among institutions of government; the legitimacy and authority of state institutions; the rules and norms that determine who holds power and how decisions are made about the exercise of authority; relationships of accountability among state officials/agencies and between these officials/agencies and citizens; the ability of government to make policy, manage the administrative and fiscal affairs of the state, and deliver goods and services; and the impact of institutions and policies on public welfare.
3

See Bräutigan (2000); Kaufmann, Kraay, and Zoido-Lobatón (1999).

4

PRSPs and Joint Staff Assessments (JSAs) of the PRSPs were reviewed for Albania,
Burkina Faso, Bolivia, Ethiopia, Guinea, Guyana, Honduras, Malawi, Mauritania,
Mozambique, Nicaragua, Niger, Rwanda, Tajikistan, Tanzania, The Gambia, Uganda,
Vietnam, Yemen, and Zambia.
5

Politically, they may be quite difficult, as they threaten important political and bureaucratic interests.
6

Donors were also involved in these consultations, but their importance is probably less than intra-governmental discussions and those with civil society.
7
International Development Association and International Monetary Fund (2002) and
World Bank (2002:55-62).
8

See, for example, World Bank (2002:56).

9

Where did the goverance agenda come from? At times, it has developed from research, when scholars have found an association between particular kinds of policies and institutional arrangements associated with growth or when analysis indicates that corruption and instability constrain it. At times, the good governance agenda has expanded as a result of advocacy by committed partisans of democratic government, sustainable development, empowerment of the poor, free trade, participatory development, or other desirable condition. The expanding agenda may also serve particular programmatic, sectoral, or policy interests of donors, ministries, politicians, and researchers.
10

Brautigan (2000).

11

Recent reviews of the PRSP process (International Development Association and
International Monetary Fund 2002 and World Bank 2002) recognize the importance of setting priorities and sequences, but indicate that these are issues that countries need to

47

address in their PRSPs and the actions that follow from them. They recommend basing priorities on detailed costing of alternatives. The PRSP Sourcebook provides insight into the sequences of actions that need to be taken to introduce specific governance reforms.
The point to be made here is that there is little knowledge currently available to guide strategic thinking about setting priorities and sequencing activities.
12

World Bank (n.d.) presents a somewhat different list and then assesses the extent to which governments address these issues in the PRSPs. In this document, governments are urged to increase the accountability of government and the transparency of its decisions; they are urged to consider the “architecture” of the state and its link to corruption; they are expected to provide evidence of “political will” to undertake difficult changes; they are instructed to consider how decentralization can improve conditions for the poor; they are encouraged to undertake civil service reform and the reform of the legal system. In addition, they should include parliaments and parliamentarians in the discussion of poverty; they should consider how to sequence and measure government performance; they should increase participation, particularly by the poor; and should undertake public expenditure management, and they should increase government revenues 13

For example, Bolivia’s PRSP commitments include the “implementation of a merit based administrative and career development system; an integrated management and administrative modernization system; and a performance based monitoring and evaluation system” as part of its public administration reform, at the same time it
“proposes to accelerate judicial reform, to consolidate independence of judicial branch,”
“undertake to deepen the decentralization process,” “measures to deepen popular participation and social control over the EBRP,” among other commitments. These are all important goals, but their achievement can only be halting and difficult, given limited capacity and the ever-present specter of political instability, even in a country that has undertaken a very significant amount of reform in the past two decades. World Bank
(n.d.: Annex 2).
14

Casson (2001).

15

World Bank (2002:56).

16

Grindle (2001:36-37).

17

The term is from Neustadt and May (1986).

18

Chang (2000).

19

Moore (1998).

20

Chang (2000:no page number).

48

21

World Bank (1993)

22

Tendler (1997).

23

Grindle (2000); Nelson (1990); Grindle (forthcoming).

24

International development Association and International Monetary Fund (2002).

25

This is an adage adopted by the Department for International Development in the U.K. as part of its governance work.
26

Gamarra (1994).

27

Fox (2001).

28

Tendler (1997).

29

In commenting on the role of NGOs in food security in developing countries, for example, Paarlberg (2002:49) observes, “In sum, asking NGOs to provide essential public goods where national governments have failed to do so is usually asking too much.
NGOs are good at many things, but they have not yet demonstrated an ability to keep or restore the peace divided societies, and they are unable to push governments to embrace democracy or to make the research and infrastructure needed to supply the rural poor with better transport, power, water, or technology options. NGOs can help with all of thse tasks if governments are doing their job. But when national governments fail or abdicate, NGOs can compensate only to a limited degree.”
30

See Grindle (1997), Lindauer and Nunberg (1994), Nunberg (2000), World Bank
(1995).
31

Initiatives included establishing clear categories of personnel, linking them to pay scales and career systems, developing job descriptions, and introducing measures to link performance to pay and mobility reward structures. In addition, some countries have sought to devolve personnel decision making—including standards for performance and hiring, promotion, and firing decisions—away from central civil service commissions or human resource units to operational organizations within the public sector. See Kamarck
(2000:244).
32

Of course, countries differed in terms of the alacrity with which they took up such reforms and the extent to which they actually put them in place after committing themselves rhetorically to change. Similarly, countries differed in the extent to which they maintained their commitment to public sector reform over time. Of 31 countries reporting efforts to downsize the public service, for example, 11 actually employed more

49

public workers at the end of the 1990s than they had in the 1980s. Many countries introduced broad reform programs, but implemented them haltingly.
33

World Bank (2001a:87).

34

See DiIulio (1994).

35

DiIulio (1994), Tendler (1997), Grindle (1997a).

36

See, for example, Donahue and Nye (2003).

37

It is notable, for example, how frequently ministries of finance and central banks perform reasonably well, even when most other parts of government do not.
38

Putnam (1993).

39

See, for example, World Bank (2002).

40

International Development Association and International Monetary Fund (2002:9).

41

See, for example, Fowler (1991).

42

For a discussion of participation in the PRSP process, see Whaites (2002), and World
Bank (2002).
43

Kingstone (1999).

44

See, for example, International Development Association and International Monetary
Fund (2002:9)
45

Grindle (1997).

50

References
Bräutigan, Deborah, 2000. Aid Dependence and Governance. Stockholm: Almqvist and
Wiksell International.
Burki, Shahid Javed, Guillermo Perry, and William Dillinger, 1999. Beyond the Center:
Decentralizing the State. Washington, D.C.: The World Bank, 1999
Casson, Kathryn, 2001. “Governance and the Poverty Reduction Strategy Process: A
Review of 23 I-PRSPs and PRSPs.” London: Department for International Development.
Chang, Ha-Joon, 2001. “Institutional Development in Developing Countries in a
Historical Perspective: Lessons from Developed Countries in Earlier Times.”
Unpublished paper, University of Cambridge.
Court, Julius, and Goran Hyden, 2001. Towards a World Governance Assessment:
Preliminary Findings from the Pilot Phase. United Nations University, Working Paper.
Donahue, John, and Joseph Nye, 2003. Public Service in the Twenty-first Century.
Washington, D.C.: The Brookings Institution.
DiIulio Jr., John, 1994. “Principled Agents: The Cultural Bases of Behavior in a Federal
Government Bureaucracy.” Journal of Public Administration Research and Theory, Vol.
4, no. 3.
Fowler, Alan, 1991. “The Role of NGOs in Changing State-Society Relations,”
Development Policy Review, Vol. 9, no. 1.
Fox, Jonathan, 2001. “Vertically Integrated Policy Monitoring: A Tool for Civil Society
Policy Advocacy. Nonprofit and Voluntary Sector Quarterly, Vol. 30, no. 3 (September).
Gamarra, Eduardo, 1994. “Crafting Political Support for Stabilization: Political Pacts and the New Economic Policy in Bolivia.” In William C. Smith, Carlos H. Acuña, and

51

Eduardo A. Gamarra, ed., Democracy, Markets, and Structural Reform in Latin America.
New Brunswick, N.J.: Transaction Publishers.
Grindle, Merilee, forthcoming. Despite the Odds: The Contentious Politics of Education
Reform.
__________, 2001. “The PRSP Process: What Next?” London: Department for
International Development.
__________, 2000. Audacious Reforms: Institutional Invention and Democracy in Latin
America. Baltimore, MD.: The Johns Hopkins University Press.
__________, 1997. “Divergent Cultures? When Public Organizations Perform Well in
Developing Countries.” World Development. Vol. 25, no. 4.
International Development Association and International Monetary Fund, 2002. “Review of the Poverty Reduction Strategy Paper (PRSP) Approach: Main Findings.” Washington,
D.C.: the World Bank and IMF, March 15.
Kamark, Elaine, 2000. “Globalization and Public Administration Reform.” In Joseph Nye and John Donahue, eds., Governance in a Globalizing World. Washington, D.C.: The
Brookings Institution.
Kaufmann, Daniel, Aart Kraay, and Paublo Zoido-Lobatón, 1999. “Aggregating
Governance Indicators.” World Bank Policy Research Working Paper 2195. Washington,
D.C.:The World Bank.
Kingstone, Peter, 1999. Crafting Coalitions for Reform: Business Preferences, Political
Institutions, and Neoliberal Reform in Brazil. University Park, PA: Penn State University
Press.
Lindauer, David L. and Barbara Nunberg. 1994. Rehabilitating Government: Pay and
Employment Reform in Africa. Washington, D.C.: The World Bank.
Moore, Mick, 1998. “Death Without Taxes: Democracy, State Capacity, and Aid
Dependence in the Forth World.” In Mark Robinson and Gordon White, eds., The
Democratic Developmental State. Oxford: Oxford University Press.
Nelson, Joan M. ed., 1990. Economic Crisis and Policy Choice: The Politics of
Adjustment in the Third World. Princeton, N.J.: Princeton University Press.
Narayan, Deepa. 2000. Voices of the Poor: Can Anyone Hear Us? New York: Oxford
University Press for the World Bank.

52

Neustadt, Richard E., and Ernest R. May, 1986. Thinking in Time: The Uses of History for Decision-Makers. New York: The Free Press.
Nunberg, Barbara, ed., 1999. The State after Communism: Administrative Transitions in
Central and Eastern Europe. Washington, D.C.: The World Bank.
Paarlberg, Robert, 2002. “Governance and Food Security in an Age of Globalization.”
IFPRI Food, Agriclture, and the Environment Discussion Paper 36. Washington, D.C.:
International Food Policy Research Institute.
Putnam, Robert, 1993, Making Democracy Work: Civic Traditions in Modern Italy.
Princeton, N.J.: Princeton University Press.
Tendler, Judith, 1997. Good Government in the Tropics. Baltimore, MD: The Johns
Hopkins University Press.
Whaites, Alan, ed., 2002. Masters of Their Own Development? PRSPs and the Prospects for the Poor. Monrovia, CA: World Vision International.
World Bank, 2002. “Review of the Poverty Reduction Strategy Paper (PRSP) Approach:
Early Experience with Interim PRSPs and Full PRSPs, Final Text.” Washington, D.C.:
The World Bank, April 1.
World Bank, 2001a. Global Development Finance, Analysis and Summary Tables.
Washington, D.C.: 2001:87.
World Bank, 2001b. PRSP Sourcebook. Draft for comments. Washington, D.C.: The
World Bank.
__________, 1995. Bureaucrats in Business: The Economics and Politics of Government
Ownership. New York: Oxford University Press.
__________, 1993. The East Asian Miracle: Economic Growth and Public Policy. New
York: Oxford University Press.
___________, n.d. “Thematic Review: Governance and PRSPs.” Washington, D.C.: The
World Bank.

53

54

References: Bräutigan, Deborah, 2000. Aid Dependence and Governance. Stockholm: Almqvist and Wiksell International. Burki, Shahid Javed, Guillermo Perry, and William Dillinger, 1999. Beyond the Center: Decentralizing the State Casson, Kathryn, 2001. “Governance and the Poverty Reduction Strategy Process: A Review of 23 I-PRSPs and PRSPs.” London: Department for International Development. Chang, Ha-Joon, 2001. “Institutional Development in Developing Countries in a Historical Perspective: Lessons from Developed Countries in Earlier Times.” Court, Julius, and Goran Hyden, 2001. Towards a World Governance Assessment: Preliminary Findings from the Pilot Phase Donahue, John, and Joseph Nye, 2003. Public Service in the Twenty-first Century. DiIulio Jr., John, 1994. “Principled Agents: The Cultural Bases of Behavior in a Federal Government Bureaucracy.” Journal of Public Administration Research and Theory, Vol. Fowler, Alan, 1991. “The Role of NGOs in Changing State-Society Relations,” Development Policy Review, Vol Fox, Jonathan, 2001. “Vertically Integrated Policy Monitoring: A Tool for Civil Society Policy Advocacy Gamarra, Eduardo, 1994. “Crafting Political Support for Stabilization: Political Pacts and the New Economic Policy in Bolivia.” In William C __________, 2001. “The PRSP Process: What Next?” London: Department for International Development. __________, 2000. Audacious Reforms: Institutional Invention and Democracy in Latin America __________, 1997. “Divergent Cultures? When Public Organizations Perform Well in Developing Countries.” World Development International Development Association and International Monetary Fund, 2002. “Review of the Poverty Reduction Strategy Paper (PRSP) Approach: Main Findings.” Washington, Kamark, Elaine, 2000. “Globalization and Public Administration Reform.” In Joseph Nye and John Donahue, eds., Governance in a Globalizing World Kaufmann, Daniel, Aart Kraay, and Paublo Zoido-Lobatón, 1999. “Aggregating Governance Indicators.” World Bank Policy Research Working Paper 2195 Kingstone, Peter, 1999. Crafting Coalitions for Reform: Business Preferences, Political Institutions, and Neoliberal Reform in Brazil Lindauer, David L. and Barbara Nunberg. 1994. Rehabilitating Government: Pay and Employment Reform in Africa Moore, Mick, 1998. “Death Without Taxes: Democracy, State Capacity, and Aid Dependence in the Forth World.” In Mark Robinson and Gordon White, eds., The Nelson, Joan M. ed., 1990. Economic Crisis and Policy Choice: The Politics of Adjustment in the Third World Narayan, Deepa. 2000. Voices of the Poor: Can Anyone Hear Us? New York: Oxford University Press for the World Bank. Nunberg, Barbara, ed., 1999. The State after Communism: Administrative Transitions in Central and Eastern Europe Paarlberg, Robert, 2002. “Governance and Food Security in an Age of Globalization.” IFPRI Food, Agriclture, and the Environment Discussion Paper 36 Putnam, Robert, 1993, Making Democracy Work: Civic Traditions in Modern Italy. Tendler, Judith, 1997. Good Government in the Tropics. Baltimore, MD: The Johns Hopkins University Press. Whaites, Alan, ed., 2002. Masters of Their Own Development? PRSPs and the Prospects for the Poor World Bank, 2002. “Review of the Poverty Reduction Strategy Paper (PRSP) Approach: Early Experience with Interim PRSPs and Full PRSPs, Final Text.” Washington, D.C.: World Bank, 2001a. Global Development Finance, Analysis and Summary Tables. World Bank, 2001b. PRSP Sourcebook. Draft for comments. Washington, D.C.: The World Bank. __________, 1995. Bureaucrats in Business: The Economics and Politics of Government Ownership __________, 1993. The East Asian Miracle: Economic Growth and Public Policy. New York: Oxford University Press.