Exposing the Unfinished Business of Building Public Administration in Late Democracies: Lessons from the COVID-19 Response in Brazil.

The COVID-19 pandemic has exposed fundamental flaws in the design of public administration in late democracies. While much writing to date focuses on the initial and vital responses to COVID-19, the magnitude of this event also furthers insights into the risks of incomplete institutional designs and practices, such as the case of Brazil, an example of the administrative flaws in late democracies. This article is not a critique of responses to COVID-19 per se, but an examination of these considering democratization processes that include state-building and the need for another push in administrative and political reforms. Shortcomings in state-building, which existed before COVID-19, inflict heavy costs on society and, if left unaddressed, add to the costs of future disasters and unraveling of support for state and democratic institutions.

Evidence for Practice

Crises like COVID‐19 are an opportunity to further improve public administration systems in late democracies.

Poor design prevents effective intergovernmental cooperation as a rule, not exception.

Performance and accountability requirements for political appointees could improve performance.

Managerial, political, and constitutional reforms are needed to improve the public management systems in many late democracies.

This article examines inadequate public administration practices in late democracies and how the COVID‐19 pandemic has highlighted these long‐standing risks. Late democracies are countries that became democratic in the second part of the twentieth century, typically after periods of dictatorships and colonization, some as recent as the late 1980s or 1990s (e.g., Brazil and Indonesia). Many late democracies have structurally or constitutionally fragmented—and sometimes conflicting—administrative governance systems, as well as systematically inadequate performance accountability and poor public leadership (Pertiwi and Ainsworth 2020; Tarverdi, Saha, and Campbell 2019). While several mature democracies have also shown issues with their COVID‐19 responses (Bouckaert et al. 2020; Kettl 2020a), the above factors have inflicted additionally heavy costs, especially notable in low‐ and middle‐income countries such as Brazil, India, Pakistan, and South Africa (World Bank 2021). We argue that another push is needed to address these public administration shortcomings in late democracies. If left unaddressed, these shortcomings increase the costs of future disasters and further undermine public support for state institutions.

As many countries became democracies in the last decades of the twentieth century, initial optimism was great that democracy, along with increased education, would build up an effective public administration. However, many late democracies inherited small or inadequate public administration systems, often from colonial powers in Asia and Africa (e.g., Malaysia and Nigeria) or repressive, military regimes in Latin America (e.g., Brazil and Argentina). Several decades later, only a few of those countries achieved highly functional public administration systems that can deliver good public services under democratization. The few that did often have a strong prior base in authoritarian development regimes (e.g., Singapore and South Korea; see de Avila Gomide 2021; Haque and Puppim de Oliveira 2021a). Indeed, more‐authoritarian regimes can perform better in some governance indicators compared with partially democratized countries (Tarverdi, Saha, and Campbell 2019). Thus, an important question is: What are the main problems in state‐building that remain for many late democracies?

The outbreak of COVID‐19 exposes several of those problems and offers an opportunity to revisit unfinished reforms in several countries. The first part of this article lays out key challenges of unfinished state‐building, discussed for the case of Brazil as exposed by COVID‐19. The second part argues for both incremental and far‐reaching reforms; the article discusses these as offering rich, new challenges to public administration theory and practice.

Questions about the link between democracy and state‐building have a long history in the literature. Scholars and practitioners have extensively debated whether democracy can build an effective public administration (Pertiwi and Ainsworth 2020; Wheare 1951). A key issue is what should come first, an effective state that manages the democratization process or a democratic system that builds a functional public administration that is responsive to the wishes of the people (Carbone and Memoli 2015). In the case of late democracies, the latter strategy has been often followed, and scholars note that subsequent capacity building in low‐ and middle‐income countries has been a tough challenge (Farazmand 2009; Haque et al. 2021b). They often have not built strong mechanisms of good governance in government effectiveness, rule of law, and control of corruption (Tarverdi, Saha, and Campbell 2019), nor overcome inadequacies in technology, trained staff, and resources to invest in health, education, and other public services (El‐Taliawi and van der Wal 2019). For example, in Argentina and Mexico, clientelism practices, such as rewarding public jobs to party members, have remained after democratization and reforms (Donadelli, Cunha, and Dussauge‐Laguna 2020; Grindle 2012; Panizza, Larraburu, and Scherlis 2018). In Indonesia, reform efforts to address corruption problems in provinces have yet to show effectiveness (Lewis and Hendrawan 2020). In the economic development literature, there is some evidence that authoritarian regimes led to rapid economic development at the beginning of the development process, but growth tended to stagnate when countries reach middle‐income status (Ruttan 2003, 132). The key is having a transition to democratic regimes while keeping improvements in economic and social development, which few late democracies were able to do. Evidence links such issues to struggles in state‐building, though countries where authoritarian regimes built up strong state capacity before democratization feared better in democratic transitions (de Avila Gomide 2021). For example, South Korea is an example of the latter which, while appearing to lose state capacity just after democratization in late 1980, regained it in subsequent reforms (Hellmann 2020).

We extend the above by focusing on the role of public administration and remaining reforms that are now needed in many late democracies. This article focuses on two key issues faced by late democracies, namely, having fragmented administrative governance systems, and systematically inadequate performance accountability for public managers and politicians, shown in great detail through its COVID‐19 responses. Brazil, which had a military government between 1964 and 1985, began its democratization thereafter and has the world's sixth largest population (211.8 million, IBGE 2020). Brazil is also relevant to understanding the unfinished business of reforming public administration in late democracies, having attempted many reforms in past decades. In the following, we first discuss how structurally, and constitutionally fragmented, ill‐defined and uncoordinated responsibilities hinder response effectiveness. This is made worse by inadequate performance accountability and job preparedness by appointed public officials, rooted in political cultures of patronage. The second part then discusses reforms that improve these matters.

Flaws in Late Democracies Exposed: The Case of Brazil

Challenges of Effective Federal and Intergovernmental Cooperation Under Poor Design

The pace of state‐building has not always gone hand in hand with the development of capacity and intergovernmental collaboration for dealing with emergencies, particularly at the local level (McGuire and Silvia 2010). In late democracies, decentralization processes often occurred fast, with great fervor, resulting in coordination and decision‐making gaps in intergovernmental relations (Smoke 2015). Arguably, designers were more concerned with democratic rights and arrangements among institutions to protect these compared to matters of performance and issues involving coordination and responsibilities in intergovernmental relations. Yet, public policies and state responses to crises and other issues have become increasingly dependent on how well different levels of government work together (Puppim de Oliveira et al. 2021; Agranoff 2011). Key decision‐making authorities and processes for solving policy problems that require a high level of coordination and expediency are seldom sufficiently established in either law or practice; instead, intergovernmental relations have often been caught up in issues such as fiscal governance and, in recent years, recentralization and political polarization (Conlan 2017; de Mello and Jalles 2020).

The Brazil case exemplifies this issue. The Brazilian constitution of 1988 establishes not only three levels of government (federal government [union], 26 states plus the Federal District, and municipalities [5,570]), but also great autonomy for each level. Following decades of authoritarianism and centralization, the spirit was strong for deep decentralization of power. Today, Brazil is one of a few countries where municipalities have large political autonomy from states defined by the national constitution. This constitution specifies a number of functions for them, ranging from primary education and health to sanitation. In addition, it gives subnational legislative bodies extensive autonomy in policy making on matters that are not preempted by higher levels of governments. This contrasts with the United States, where not all cities have home rule, states may withhold functions, and higher government may impose requirements on cities (Kettl 2020b). Even though subnational law cannot contradict national laws, Brazil's Supreme Court has often upheld subnational autonomy, including during COVID‐19 (Richer 2020).

Specifically, the 1988 Federal Constitution in Brazil (Title 3 about the Organization of the State) prevents the federal government from intervening in states and municipalities, except in specific cases mentioned in Article 18, such as to re‐establish public order or due to foreign invasion (Constituição Federal 1988). In 2020, courts backed both São Paulo and Rio de Janeiro governors’ refusal to follow federal directives for COVID‐19. The disagreements ranged from the protocols for COVID‐19 treatment to the kind of vaccines. At issue are not conflicts that emerge between the federal government and states, as in any multi‐jurisdictional context. Instead, the federal government lacks adequate constitutional powers to force and lead in centralized and coordinated decision‐making and implementation. Succinctly, in the United States, President Trump chose for the federal government not to lead; states drifted in different directions in terms of responses to COVID‐19 (Kettl 2020a). In Brazil, President Bolsonaro, besides having made a similar decision to Trump's, also lacked essential powers to lead, even if he or any other president had wanted. A similar situation occurs with regard to the lack of states' constitutional powers to coordinate decision‐making and implementation with municipalities in Brazil.

Constitutional arrangements also lead to fragmentation with insufficient regard for capacity and performance. The constitution requires municipalities to perform certain functions and to have their own political and administrative systems. As a result, Brazil has 5,570 systems of health and education. On the one hand, this is a wonderful laboratory for experimentation and innovation. On the other hand, several capacity issues and coordination gaps with higher levels of government emerge. Problems in coordination are neither merely ineffective communication nor caused by political rivalries. The coordination gap is structural. Concurrent responsibilities, such as health, are neither clearly defined regarding the responsibilities of different levels nor clear about which level should be blamed in case of failure, including for emergency responses such as COVID‐19. Additionally, as discussed in the next section, it is also unclear what consequences, if any, may occur for public managers who perform poorly. Moreover, as the number of municipalities has grown gradually since the 1960s without any clear criteria for administrative and financial capacity and performance for their creation or extinction, fragmentation, scale, and capacity issues have arisen. Municipalities like São Paulo (over 12 million inhabitants) and Borá (only 837 inhabitants) have the same autonomy and responsibilities for providing health, education, and urban infrastructure although their capacities and resources are different. Moreover, there is no requirement or incentive for collaboration to overcome scale and coordination issues.

The outbreak of COVID‐19 exposed this fragmentation of the Brazilian federation with a chaotic response to the public health emergency. As a result, Brazil's tripartite national health system (SUS), where the union, states, and municipalities share responsibilities and contributions of resources, is barely integrated across the three levels. Each municipality and state have their own health system. Little information is shared among them, let alone between public and private systems, unlike well‐integrated and coordinated healthcare systems like in Taiwan (Huang 2020). Most municipalities lack adequate resources. Supplemental funding from the federal government, which is required to transfer resources to states and municipalities (even if it disagrees with their policies), has been typically inadequate to ensure sufficient and equitable healthcare investment and management capacity. Even with federal aid, many municipalities, both large and small, lacked adequate intensive care facilities, resulting in many COVID‐19 patients who died waiting to receive care. The different levels of government blame each other for the failures. These are surely some reasons why Brazil has among the largest number of COVID‐19 deaths in the world, with over 500,000 as of June 22, 2021 (Our World in Data 2021).

Fragmentation, supported by law, often leads to unexpected consequences that add to adverse outcomes. Municipalities and states manage their own public medical facilities. However, by national law, residents can go to any public medical facility, even if they are not necessarily in their municipalities or states of residence. This can delay and complicate COVID‐19 tracing, treatment coordination, and resource planning. Sometimes, it is easier for a municipality or state to send its citizens to neighbor entities than invest in its own facilities. For example, the Federal District, which is surrounded by the poorer Goiás State, wanted to prevent residents with COVID‐19 from this state from coming to the capital's facilities due to overcrowding its hospitals and stressing its health funds (Valente 2020). There is no interjurisdictional coordination for managing public services and resolving problems affecting more than one entity. Such issues should be settled by court, law, or further reform, but they are not.

The same conditions are found in other late democracies in Latin America (Ramirez de la Cruz et al. 2020). Good performance and intergovernmental coordination are the exceptions rather than the rule. There are instances of effective agency for improving public services, where a group of politicians, bureaucrats, nonprofit organizations, and civil society actors develop a network or policy field to make policy changes and improve implementation (Moulton and Sandfort 2017; Stone and Sandfort 2009). However, the establishment of those networks takes time, which is a crucial element in an emergency response. Structural fragmentation, backed by constitutional power, aggravates poor coordination and leadership in policy making during national emergencies like COVID‐19. While many unanswered questions exist about building intergovernmental relations for effective policy responses (Hegele 2018; Kincaid and Stenberg 2011), it is well established that emergencies require a coordinated response and intense intergovernmental collaboration (McGuire and Silvia 2010), especially against the backdrop of the uneven and often inadequate capacity of state and municipal healthcare systems during peak times. Centralized and coordinated COVID‐19 responses, such as those of China and Australia (Moon 2020; Moonley and Moonley 2020), show that countries need the capacity for intergovernmental coordination, even if it is challenging in many federal systems (Paquet and Schertzer 2020). Finally, COVID‐19 responses by leaders also show an unexpected need by subnational governments for constitutional mechanisms to demand responses from the federal government where the president decided not to act. Given the prior repressive and authoritarian regimes, this is quite the unexpected and sensitive ask, indeed. There is unfinished reform business to ensure the authority, structure, and capacity of public administration and leadership to provide performance.

Weak Performance Accountability Made Worse by Too Many Political Appointees

Many late democracies were founded on the desire for freedom and protection of rights. For example, the Brazilian constitution mentions ‘right(s)’ hundreds of times in over 400 pages (Constituição Federal 1988). However, it is often not clear who should be responsible for guaranteeing the rights and costs for providing those rights. The constitution has been amended more than 100 times during the first 30 years, but not on these issues, rather, with matters like allowing foreign investment in radio and television stations and changing minimum contribution time for social security (Calgaro 2018). These protections of institutional and legal rights include an abundance of rules and processes for accountability in response to prior repression and corruption by cliques, factions, and cronies, which often continue in late democracies, despite the attempts to create anti‐corruption bodies (Doig, Watt, and Williams 2007). Alas, the protections involve heavy rule compliance and oversight, often by many layers of bureaucracy and legal offices; managers have legitimate fears about being accused of rule‐breaking and other wrong‐doing. Not only does such heavy rule compliance thwart promptness and stifle innovation. Lacking in all this is the development of cultures, norms, and processes of performance and performance accountability that ensures decision‐making processes and programs are effective as well. This void is vitally exposed in the COVID‐19 crisis; rule compliance does not guarantee performance.

In Brazil, the legal‐ and rule‐oriented culture exists, rooted in the heritage of Portuguese law, without reward or consequence for performance. Traditional, outdated personnel systems reinforce weak performance accountability (Puppim de Oliveira and Gomes 2018). While administrative reforms have occurred, they remain inadequate. Career civil servants enter the civil service through difficult examinations. After probation, promotion is almost automatic, based on seniority and personal relationships. Promotion incentives are close to politicians to be recommended (indicated) for senior managerial posts, including political appointees (DAS—Direção e Assessoramento Superior). For senior positions, selection seldom involves open or competitive processes or the assessment of CVs, past performance, and/or qualification. There is little performance accountability, making it both difficult and rare to be reprimanded or terminated, even for gross negligence.

Of course, exemplary senior officials and public programs exist. Brazil is recognized for these, but they are not the norm. In fact, they often occur in exceptional circumstances like political priority or senior public managers finding creative workarounds. The situation is made worse by the large number of political appointees (cargos de confiança), for which it is even difficult to gauge accurate numbers. There are estimates between 23,000 political appointees in federal ministries to 600,000 among all levels and state companies (Costas 2016; Pires 2011). This far exceeds numbers reported for larger countries, including the United States, which has 4,000–7,000 federal political appointees (US 2020). Structurally, these positions are used to reach political agreements between the executive and legislative branches in negotiation processes by exchanging favors. As in other countries, political appointees often have little expertise in the areas they are supposed to manage, even at senior levels, and face little performance accountability. Political appointees increase risks of incompetence, corruption, and poor public administration leadership. Anecdotally, many stay only a year or two in their role (Costas 2016; Gallo and Lewis 2012).

The rapid spread of COVID‐19, as well as the need for leadership responses, exposed these weaknesses. During the aggravation of COVID‐19 in the end of the first quarter of 2020, Brazil had three health ministers in less than one month. Many technical decisions were left in the hands of political appointees who lacked skills and public health expertise. The (now former) minister of health during the peak of the first COVID‐19 wave in 2020 was an army general who appointed several other active and reserve military personnel in key posts in the ministry, many without any public health experience. In some municipalities, the health official, and even the mayor, is barely literate. The health secretary of a municipality could be the ambulance driver. There are no job requirements for these posts with responsibilities that could determine life or death of citizens during a crisis. This lack of accountability and control affected COVID‐19 responses. In Rio de Janeiro State, expedited contracting was used to hire an organization to build and manage seven emergency field hospitals; however, only one was completed within an extended deadline, causing a loss of R$700 million (USD 130 million) (Exame 2020). The governor was impeached after months of investigation and officials were prosecuted and imprisoned for receiving kickbacks. In another episode, despite the lack of COVID‐19 tests in many hospitals countrywide, seven million tests were found stored that risked expiration during the first pandemic peak (Barbieri and Rodrigues 2020).

Additionally, decisions by political leaders in many late democracies are often strongly shaped by political cultures of patronage and short‐term electoral interests (Grindle 2012; Panizza, Larraburu, and Scherlis 2018). While they ensured programs of emergency support for families, protective equipment, and planned nationwide distribution of coronavirus vaccines, patronage politics are linked to economic favors, corruption, and the enrichment of family members. In the above case of Rio de Janeiro, the governor's wife was implicated through business dealing with firms behind the scheme (Exame 2020). Politicians in other states and cities are involved in similar cases of COVID‐19‐related corruption, sometimes with senior civil servants (Borges 2020). Alas, patronage extends to judicial appointments; political corruption cases and convictions are routinely dismissed, overturned, or reduced to time served.

In the design of democracy, voting is assumed to hold elected officials and their parties accountable for performance but, through multiple decades, it has been ineffective in addressing the above problems. National emergencies like COVID‐19 heighten and expose the need for senior public managers and appointees in positions of program leadership to have the right knowledge, experience, accountabilities, and incentives for effective responses (e.g., Christensen and Lægreid 2020). Note also that the above occurs in addition to other issues of democratic leadership, such as elected officials pandering to voters or ineffectively enforcing mask regulations. Finally, a hidden cost is that entrusting governance to elected and political officials has also led to serious depletion of leadership capacity and experience by senior public managers, including the underdevelopment of performance culture among public managers. Political appointees do little to foster the above, and, with so many political officials, there is little leadership that is asked, demanded, or even possible for senior public managers. The frequent changes in key posts also undermine policy continuity and experience accumulation. The above issues are structural, raising the cost of addressing key issues such as COVID‐19. In many late democracies, an urgent need exists for new practices of performance‐based leadership that are both consistent with democracy and reduce patronage‐related corruption.

The Road Ahead for Late Democracies

The response to COVID‐19 in late democracies like Brazil openly exposed a series of public management and policy problems that have been recurrent in public administration systems in those countries for decades. Key issues include structurally and constitutionally fragmented, uncoordinated, and sometimes conflicting administrative governance systems. Administrative systems protect and safeguard democratically won freedoms and rights, but lack a performance culture that furthers performance accountability, innovation, and integrity among politicians and senior civil servants. In addition, it leaves too many political officials who are ill‐prepared for the job of public leadership.

Public administration knowledge offers practical directions for addressing these issues in a democratic context. For example, the United States is constitutionally fragmented, devolving considerable power to the states. The arrangement is contestable; history shows many efforts to redress ineffective or unjust outcomes, leading to new arrangements (Kettl 2020b). As a political accommodation, such efforts continue today, contested in legislatures and courts. In other instances, management arrangements that do not produce effective outcomes are overhauled and replaced. Countries like New Zealand have often been at the forefront of managerial reforms, showing the importance of establishing performance culture (Berman and Prasojo 2018). Not surprisingly, New Zealand has been effective in watering down the COVID‐19 crisis (Husted and Sinkovics 2021). In short, democracies must continuously work to improve administrative systems and perform better lest they lose legitimacy and support.

The following draws attention to improvements and processes of making change. Some improvements increment and find support in practice; others are far‐reaching and a bit new. The following is offered with both the certainty of known cures and avenues for learning. While many late democracies have also demonstrated capacity for some major and innovative reforms (e.g., Donadelli, Cunha, and Dussauge‐Laguna 2020), clearly more is needed. A key issue is creating pathways for change and reforms. We offer the following suggestions of reforms for improving administrative systems in Brazil. They could also serve other late democracies and beyond, adapting to their context:

Ensuring power for emergency policy making and coordination, as well as being able to compel federal policy making and coordination

Increasing authority and accountability of the federal government to provide or ensure planning for major functions at all levels of government

Increasing the capacity of subnational governments to deliver performance, including collaboration as needed

Increasing performance accountability and integrity of all senior officials, including political appointees, through a nonpolitical civil service commission

Reducing the number of political appointees and ensuring merit‐based appointment of all senior officials

Strengthening administrative and legal redress mechanisms for jurisdictions to rapidly contest policy outcomes created by other political bodies

These six suggestions address the two main concerns. Coordination is strengthened by 1, 2, and 3; improving accountability is addressed in 4, 5, and 6. While all seek to increase performance, 3 and 6 target performance in a broad way for administration and policy, respectively. In the following, we first elaborate on how these reforms address the problems, paying special attention to what is novel and what can be learned from other countries. Then, we turn to processes and concluding thoughts for realizing reform change.

The first suggestion addresses the need for transferring or assuming leadership by higher levels of government in moments of crisis. Many countries, including Brazil, allow chief executives to trigger emergency executive powers when they assess that the protection and safety of citizens is endangered. Regarding many late democracies, COVID‐19 events show that such emergency executive power systems were inadequate in its functions, as systems had uncertain available assets and lacked crisis leadership. However, few late democracies have mechanisms in place for subnational governments to demand assets and leadership from the national government when they are no longer able to protect safety of its citizens (e.g., state public health systems dealing with COVID‐19). By contrast, in the United States, federal resources are available to subnational governments when governors or mayors declare emergencies, such as for flooding. Developing such practices will improve the performance of subnational governments during crises, perhaps in novel ways. In doing so, late democracies will also need to develop rules and establish practices that allay deep‐seated suspicions about the possible recentralization of power. Rules can be developed to avoid arbitrarily, broad, or open‐ended emergency declarations, ensuring the proper transfer of power back to subnational governments.

The second and third suggestions further planning and delivery capabilities by subnational governments. Higher governments must ensure that lower governments do their planning in a myriad of areas and that they have the capacity and funding to deliver. As the case of Brazil shows, much is made of the constitutional autonomy of subnational governments. Still, many other countries show that higher governments in federal systems can use the power of the purse to ensure that lower governments meet standards (e.g., U.S., Germany). Beyond this, the Indonesian experiences show the national government increasing subnational planning and public management capabilities through incentives, technical assistance, and audits of local management and delivery processes (Ateh, Berman, and Prasojo 2020). Legal authority is needed to ensure that higher governments can hold lower governments accountable for meeting minimal performance standards. Among efforts to increase planning and delivery may be efforts that also encourage smaller jurisdictions to collaborate and even jointly provide services, though the extensive research on polycentricity of Elinor Ostrom and others has shown that having larger jurisdictions per se is not always more efficient to produce a public service (Ostrom 2010; Ostrom, Tiebout, and Warren 1961). By setting standards and providing funding, they can also encourage subnational jurisdictions with inadequate resourcing or capabilities to band together in the delivery of services. Performance requires standard setting and inspection by higher governments.

The fourth, fifth, and sixth suggestions deal with the role of political officials and the need to ensure and be accountable for performance. A fundamental lesson of many late democracies is that political officials have often been insufficiently concerned with performance of the public good. In addition, they are inadequately held to account for their performance and ethical conduct through electoral processes. Perhaps it has been a naïve mistake to assume that these matters could or should be addressed. To start, regarding political officials in leadership positions, there is no reason why they should not be held to the same performance qualifications and evaluations as those in civil service when performing the same roles, duties, and tasks (Kim 2009). Civil service commissions have long fulfilled roles in executives’ promotion and evaluation, including merit‐based assurances. Asking them to fulfill similar screening roles for all positions, regardless of applicants' status, does not overstretch the imagination, even if somewhat novel. In some countries, such as Brazil, governments might consider reducing the number of political appointees as well.

More is needed to improve the quality of political appointees. Cultures and roles of performance need to be strengthened for civil servants (Behn 1998; Meier, O'Toole, and Jr. 2006). Novel rules are needed for the allocation and supervision of responsibilities by senior managers and political appointees (in the case of Brazil, ensuring adequate roles for senior public managers). To create a culture of performance, review is needed of compliance rules and regulations that stifle performance and efficiency. Prior efforts such as Reinventing Government and New Public Management show that they can often be replaced with more efficient and equally effective ones. Performance‐based management tools should be used, including strategic management, performance management, performance contracts, performance‐based awards, and whole‐of‐government efforts (Allen et al. 2017; George 2020). For example, in the United States, 60% of senior manager performance is appraised based on their contributions to organizational objectives (e.g., GSA 2017). In New Zealand, senior public managers are on contracts, subject to performance‐based evaluation that affects compensation and contract renewal. There is much to be done to build performance cultures.

Late democracies often have much to improve and another round of reforms is needed. However, identifying practical reforms is but the first step. Showing how the above reforms can come about is needed, too. Table 1 shows three reform processes: (i) managerial, (ii) political, and (iii) constitutional and the list of reforms that could have short‐ and long‐term effects on the improvement of public administration and management. This coverage is consistent with Raadschelders and Bemelmans‐Videc (2015) who conceptualize constitutional‐level regime change, ‘collective’ (political and administrative) system change, and operational‐level reforms. Table 1 also shows how these linked to the six suggestions mentioned earlier in this work.

Table 1

Categories of Reform Processes and Their Effects

Type of Process	Effects of Changes
	Short‐Term Effects	Long‐Term Effects
Managerial Processes	Formal policy coordination and planning mechanisms in and across sectors (2) Use performance‐oriented management tools (strategic management, performance evaluation) (3) Merit‐based appointment and evaluation of all senior appointments (incl. political appointees) (5) Increase open recruitment (5)	Establish a civil service commission for hiring, career homogenization, and promotion assessment (4, 5) Programs to assure capacity and performance of municipalities (3) Strengthen performance cultures (4)
Political Processes	Procedures to require/revoke centralized policy making and coordination during emergencies (1) Political party reforms (6) Require laws and practices that assess and strengthen performance, integrity, and redress at all levels (6)	Create more incentives for collaboration and agglomeration among jurisdictions as needed to ensure capacity (3) Reduce the role of the supreme court from cases involving political corruption and political defendants (4)
Constitutional Processes	Ensuring political powers of centralized policy making and coordination during emergencies (1) Establish electoral districts (6)	Clear constitutional roles and division of responsibilities among the union, states, and municipalities (2, 3) Increase capacity for administrative reforms (3) Limit the supreme court to cases involving constitutional law (6)

Note: Numbers in brackets refer to the six reforms mentioned in the article.

Source: Made by the authors and based on the Brazilian situation.

First, some reforms can be accomplished with gradual managerial changes and without major political implications. For example, expanding the roles of a civil service commission to manage the qualifications and appraisal of political appointees requires change in personnel and administrative law. Guidelines, job descriptions, and selection processes can be done by internal rule or regulation by ministries. Other reforms that strengthen performance accountability or strategic management can readily be implemented based on existing law, requiring oversight and training to further use (Berman and Prasojo 2018).

The second kind of reform in public administration and management requires deeper political consensus and momentum through democratic process. For example, the establishment of state metropolitan agencies for coordinating health or sanitation policies among the municipalities is likely to involve considerable public debate (Puppim de Oliveira et al. 2021). Incentives or requirements for municipalities to work together increase the coordination and capacity of small municipalities, particularly for dealing with complex situations like a virus pandemic. Enacting these in the case of Brazil will require new decrees and laws, involving significant political negotiation and leadership.

Finally, the third category of changes would only come about through more radical processes and constitutional changes, with a greater emphasis on performance. In Brazil, decisions over public investment are a struggle between career bureaucrats and politicians, with the latter generally having more influence (Burrier 2019). Discourse is now occurring to introduce district elections to ensure that representatives have increased knowledge of, and accountability for, specific districts; at present, all city councilors and state representatives in Brazil are elected city‐ or state‐wide. Yet, at issue is not that only each election type, as each has its own benefits and risks. District politics need to be designed to avoid pork barrel politics and control by local elites, for example. From the perspective of public administration, the bigger point is that, regardless of voting system, late democracies have been unable to ensure political performance and integrity, in all matters and at all times. Constitutional reform must open pathways for strengthening systems and cultures of accountability for politicians, such as legal and administrative mechanisms for rapid redress when performance by political bodies and officials falls short. Reforms should include courts to test outcomes against public interest, such as in the matters of small cities that lack capabilities for effective healthcare. At issue is not policy development through courts, but of courts testing for policy effectiveness. Constitutional reform must focus on the importance of performance at all levels of government and by all actors.

Some late democracies adopted constitutions that are lengthy and difficult to change (e.g., with unchangeable clauses) to prevent a return to their pre‐democratic pasts. While the above changes could be done piecemeal or as a larger reform package, paradoxically or ironically, many Latin American and African late democracies have adopted difficult‐to‐change constitutions and are built on underdeveloped administrative systems that need change the most.

As Brazil prepares a new administrative reform, it could be an opportunity to advance some of the unfinished business of previous attempts (Governo do Brasil 2020). A re‐examination of current public administration structures and processes should occur 30–40 years after initial democratization, particularly in the face of increasingly known successes and progress of other countries. However, it is important that leaders of bureaucratic reform are assisted with credible ideas and examples of reforms that could work in their countries. Senior political leaders have traditionally understood the need for bureaucratic reform. It takes only one or two motivated political officials to set the above reforms in place. COVID‐19 may offer a unique window of opportunity to spur significant reforms in the political and administrative systems. The alternative would be a further deterioration in many late democracies of the already poor quality of the public services, an overstretched and inefficient public administration, and an intractable conflictual political system.

Conclusions

Many late democracies grapple with challenges that are not uncommon to other countries. Still, these occur against a backdrop of limited state capacities and low or stagnated incomes. In addition, they are dealing with unfinished business related to their constitution and populations that, while passionate about their hard‐won freedoms and rights, are apathetic in their expectations of public sector leadership to deliver on the initial promises of democracy. Based on the flaws of the COVID‐19 response in Brazil, we made several suggestions to improve state capacities that emphasize increased performance by rebalancing the roles and accountabilities of politicians and the bureaucratic apparatus. In addition, responses should lay out the different paths, whether through managerial, political, and/or constitutional processes, to fix some of the issues left unfinished by state‐building.

There exists a need for late democracies to chart new reform paths that address the issues confronting them, including future issues like climate change and other pandemics. While they can and should learn from developed democracies, their challenges run deep and often call for deeper reforms, too, such as involving constitutional review and heightening performance standards and review of political appointees. Perhaps this is not surprising; several decades into democratization, it should be normal to take stock and make course corrections, even where some involve major recalibrations for which there is little modern precedent. The likely results are somewhat different experiences and emphases across late democracies, when taken together, may offer exciting new avenues for mature democracies to consider as well.