Introduction

Evidence continues to show that open government affects people’s lives. But there are still skeptics who are not aware of all the benefits associated with this approach. Are you struggling to convince others to take an open government approach when implementing reforms? Are you looking for concrete evidence on the impact of making government more transparent, accountable, and participatory? Then, this guide is for you.

In this guide, you will find evidence on how open governments improve business efficiency, lower corruption, deliver better quality services, and much more. Unlike other reports, this guide is based on current, mostly peer-reviewed research, in each of the policy areas included. Experts around the world collaborated for rigorous reviews of current social science in and beyond Open Government Partnership (OGP) member countries.^[1] In each section, you will find a link to read the underlying papers. The OGP Support Unit and partners will continue to update this guide, with forthcoming chapters on additional open government policies. Visit this site often for new insights on the benefits of open government.

---

Organization

This guide translates what is often technical research language into points of evidence you can use. The report is divided into 6 sections. Each section features the following:

• DEFINITIONS of key terms;
• EVIDENCE OF IMPACTS of a specific open government policy on development outcomes in non-technical language;
• CASE STUDIES of instances where governments learned about how policies worked through specific reforms or experiments and that draw lessons than can be useful for other contexts; and
• LIMITS TO THE EVIDENCE that readers should consider. While this report aims to be easy to understand, it also needs to be clear about what we collectively do not know yet and what we cannot claim about open government results.

---

What’s Inside

This guide includes easy-to-grasp evidence on the impacts of some of the most popular and successful open government policies. This guide teaches how:

• Access to justice benefits the economy and restores the social contract;
• Fiscal openness can lower corruption and raise revenue;
• Grievance redress mechanisms in the public sector can improve governance and services;
• Social audits can improve public service accessibility and trust in service providers; and
• Open contracting can increase cost savings and competition.

You will find answers to questions such as:

• How does participatory budgeting help governments save money?
• How do grievance redress mechanisms empower women?
• How does a people-centered approach to justice help businesses?

….and much more.

---

How to Use This Guide

With evidence from this guide, you can:

• Create a value proposition for open government reforms in your agency, advocacy strategy, or organization;
• Inspire reforms and commitments;
• Include examples of results in a reformer’s speech;
• Use social media to share this evidence and reach a broader audience; and
• Foster synergies between government and academia to evaluate reforms and understand how to make open government work better for everyone.

What’s New in the Second Edition?

This guide is a follow-up to the first Skeptic’s Guide to Open Government, published in 2018. The 2018 edition looked at the general impacts of open government. Alongside an update to this evidence, this second guide takes a deeper look into the impacts of specific open government policies. Taking this approach, no skeptic’s guide could cover every policy in open government. For that reason, this edition focuses on a set of policies where there was clear priority from OGP’s leadership and a large body of peer-reviewed scholarship on impacts. Some areas, such as natural resource governance may deserve a chapter, but were not included to keep the project timely and manageable. Partner organizations inspired by the model can reach out to research@opengovpartnership.org where they have ideas for future chapters.

What is Open Government?

People often ask how OGP defines “open government.” The OGP Articles of Governance, OGP’s founding document, lays out the key values for the partnership. Open governments strive to embrace the following values to improve governance and solve public challenges:

ACCESS TO INFORMATION

Access to government-held information empowers citizens to exercise their rights, hold government accountable, and participate in decision-making. Examples in this guide that highlight how transparency advances progress in key policy areas include:

• Transparent fiscal management and public contracting discourage corruption and enable reformers to follow the money to ensure that public funds are spent efficiently and in ways that benefit citizens; and
• Providing access to justice information can help citizens better understand their legal problems and the resources available to solve them.

CIVIC PARTICIPATION

Civic participation empowers citizens to participate freely in governance, either through invited consultations or by creating space for free expression and assembly. Civic participation also helps make governments more responsive, innovative, and effective. This publication highlights several of the many ways civic participation can advance progress in key reform areas. For example:

• It is often difficult for governments—nationally or locally—to create mechanisms for the poor, women, youth, and other marginalized groups to participate in their community governance. Social audits are productive channels for this type of interaction in a collaborative rather than adversarial format.
• Evidence shows that participatory budgeting at the local level leads to more effective government spending on public services.

PUBLIC ACCOUNTABILITY

Public accountability refers to rules, regulations, and mechanisms that allow citizens to demand that government actors justify their actions, act upon criticisms or requirements made of them, and accept responsibility for failure to perform with respect to laws or commitments. Evidence in this guide highlights ways in which citizens can hold their government accountable. For example:

• Audits of public budget execution near election cycles give citizens valuable information to decide on whether to reelect incumbent officials.
• Grievance redress mechanisms allow citizens to correct government failures in fulfilling a legally mandated right.

Access to Justice

This chapter is based on a background paper by Mark Weston developed to support the Skeptic’s Guide with guidance from Pathfinders for Peaceful, Just, and Inclusive Societies.

Definitions

• ACCESS TO JUSTICE – the ability of people to resolve and prevent their justice problems and use justice as a platform to participate in their economies and societies.^[4]

The Evidence

Here is what the evidence shows on how inadequate access to justice causes different outcomes and how improving access to justice can avoid those impacts.

---

Limits to the Evidence^[41]

Fiscal Openness

This chapter is based on a background paper by Martin Haus, Joachim Wehner, and Paolo de Renzio developed to support the Skeptic’s Guide in partnership with the Global Initiative for Fiscal Transparency and the International Budget Partnership.

Definitions

• FISCAL OPENNESS OR BUDGET OPENNESS – government action that ensures transparency and citizen engagement at any stage of the budget cycle. This includes transparency and participation in spending and revenues as well as deficits and debt.^[43]
• BUDGET CYCLE – the budget cycle covers the formation, approval, execution (including procurement), and audit and evaluation stages of budgeting.^[44] (See Figure 2 for a simplified description of the budget cycle.)
• PARTICIPATORY BUDGETING (PB) – a deliberative process in which community members decide how to spend part of a public budget.^[45]

The Evidence

Fiscal openness can benefit individual citizens, societies, and governments. Here is what the evidence shows about how fiscal openness impacts areas such as health, control of corruption, and public security. This chapter includes case studies that support the evidence as well as future areas for research on fiscal openness.

---

Limits to the Evidence

Open Contracting

This chapter is based on a background paper by Andy McDevitt developed to support the Skeptic’s Guide with guidance from the Open Contracting Partnership.

Definitions

• OPEN CONTRACTING – an approach to reform public procurement policies and processes to increase transparency, participation, and inclusivity (from the planning to the implementation of public contracts) by using open, timely, and accessible data for decision-making, monitoring, and oversight.
• E-PROCUREMENT – the use of information technology in purchasing public goods, works, and services from suppliers.
• GENDER-RESPONSIVE PROCUREMENT – a government commitment to include gender-responsive objectives and actions, redress gender biases, and build internal capacity to implement gender-responsive actions. This may include both measures to support the inclusion of women-led businesses to participate as suppliers and measures to ensure goods, works, and services procured meet the needs of women and girls.^[71]

The Evidence

Here is what the evidence shows on the impact of open contracting on public spending, efficiency, quality of public works and services, fighting corruption, and social inclusion.

---

Limits to the Evidence

Social Audits

This chapter is based on a background paper by Suchi Pande developed to support the Skeptic’s Guide with guidance from the Accountability Research Center.

Definitions

• SOCIAL AUDITS – a tool where citizens can engage and hold government service providers accountable. For this guide, social audits have the following characteristics:^[108]
  • An information-gathering process to identify the standards or laws that determine citizens’ rights to, and quality of, public services;
  • A third-party convening (whether by government or civil society) to assess how policy goals, standards, or laws compare with public sector practice, with an emphasis on access to and quality of services; and
  • A public forum where findings from the assessment are discussed, and actionable recommendations are identified, prioritized, and agreed upon. See figure 4 below for an overview of the phases of social audits.

In this guide, we include evidence from two successful social audit approaches:

• Community Score Card (CSC): a type of social audit developed by CARE (Cooperative for Assistance and Relief Everywhere) in 2002. This approach includes: identifying the right to services; involving community members and service providers via in-person meetings for discussion and creation of joint action plans; and monitoring and evaluating plan implementation.
• Citizen Voice and Action (CVA): an evidence-based social accountability model developed by international humanitarian aid organization World Vision in 2008. While the practical implementation is flexible, there are four core elements: sharing information about public services; increasing opportunities for citizens to voice feedback and influence decisions; creating opportunities for dialogue and collaboration; and increasing accountability of service providers.

In addition to social audits led by local civil society or international nongovernmental organizations (INGOs), there are large-scale government-led social audits. Oftentimes, there are synergies between the different actors managing social audits. This section includes examples of social audits managed by local civil society, INGOs, and governments.

The Evidence

Here is what the evidence shows on how social audits impact public service delivery and how they can raise public awareness about public services and improve trust.^[109]

---

Limits to the Evidence

Grievance Redress Mechanisms

This chapter is based on a background paper by Suchi Pande and Naomi Hossain developed to support the Skeptic’s Guide with guidance from the Accountability Research Center.

Definitions

• Grievance redress mechanisms: Institutionalized, public sector processes that enable citizens to complain about and seek redress for services they should rightfully have received.^[129] Government agencies create GRMs to encourage individuals to voice their concerns.^[130] Elements of a successful GRM are:^[131]
  • Citizen capacity to complain: People should know their rights and what to expect from public programs. They should have institutional support to access benefits and to use complaint mechanisms to get a timely response without fear of reprisal.
  • State capacity to respond: State actors need independence, authority, and resources to register, track, and resolve complaints and enforce decisions. State actors must be able to redirect or escalate complaints when they cannot handle them themselves, and they must be committed to maintaining the anonymity of complainants.
  • Proactive transparency: GRMs must be proactively transparent about the nature and number of complaints, the rate of resolution, and the time taken to fully resolve them. Where complaints are dealt with in secrecy, there are few external pressures on government actors to resolve them. Proactive transparency is particularly crucial if GRMs are to achieve positive policy feedback, as it can highlight widespread or systematic problems within a sector that require transparency to rebuild institutional trust.

Experiences from GRM implementation in developing countries yield additional lessons learned. The following elements create strong GRMs:

• Civil society involvement in co-designing GRMs and in verifying that complaints are addressed;^[132] and
• Including safeguards against retaliation like ensuring anonymity even when complaints are transferred between departments. Citizens might decrease their use and trust in GRMs if their use leads to retaliation against them. This retaliation can mean losing their beneficiary status or can escalate to even more serious consequences such as harassment and murder.^[133]

Figure 5 highlights these elements.

GRMs in the public sector

GRMs are increasingly widespread in different sectors. Development aid agencies such as the World Bank have increasingly included them in their lending operations since 2008 and, since 2018, the World Bank has made them mandatory in all investment project financing.^[135] GRMs have also been implemented in the private sector, as they have proven to reduce “financial, construction, operational, reputational, and corporate risks.”^[136] Experiences from these sectors have strengthened the case for GRMs. Figure 6 lists some of the benefits of private-sector GRMs that can transfer to the public sector.

To date, much of the research on GRMs has been about their use in public-private partnerships, such as megaprojects, or major private investments such as mining or petroleum operations. This section focuses on a growing trend: putting GRMs into the day-to-day operations of public-sector administration.

Social audits and GRMs complement each other. Social audits are a tool to monitor public service delivery, while GRMs are an avenue to voice complaints when governments are not delivering those services as they should. In a sense, social audits are a preventative measure whereas GRMs are reactive ones.

The Evidence

Here is what the evidence shows on how GRMs work in the public sector and how they can be designed and implemented to be more effective.^[138]

---

Limits to the Evidence

Civic Space

This chapter is based on a background paper by Peter Noorlander developed to support the Skeptic’s Guide with guidance from the American Bar Association’s Rule of Law Initiative and Center for Human Rights.

The Benefits of Protecting Civic Space and Democratic Freedoms

Civic space and democratic freedoms are the cornerstone of an open and democratic society.^[1] An open civic space allows citizens and civil society organizations to freely and independently associate, express themselves, participate in public life, and seek information from the state.^[2]

Definitions

• Civic space and democratic freedoms. For the purposes of this chapter, this encompasses the fundamental freedoms of
  • expression,^[3]
  • assembly, and
  • association.

  These rights are protected in articles 19 and 20 of the Universal Declaration of Human Rights, as well as in Articles 19, 21, and 22 of the International Covenant on Civil and Political Rights, the main international binding human rights treaty ratified by 173 countries.

  Some organizations and experts in the field define civic space more broadly, to include interactions between the government and civil society, as well as the right to seek information. This chapter, however, will focus on the three rights listed in the definition. They are also often called “democratic freedoms,” “civic freedoms,” “civic rights,” or “civil liberties.” This review will use these terms interchangeably.

• Media freedom: The ability of journalists to report freely on matters of public interest.^[4] It is a principle within freedom of expression that, as the evidence in this chapter shows, is associated with positive societal benefits.

Evidence

The following sections present evidence of the positive impact protecting civic space has on society as well as the deleterious effects of measures that threaten or restrict any of the freedoms within it.^[5]

Economic Growth

• Improving democratic freedoms is closely associated with faster growth. Countries that move from 0 to 1 on the V-Dem Civil Liberties Index increase their per capita GDP by an average of 1.3 percent over a period of five years.^[6] This index includes indicators such as freedom of speech, media censorship, harassment of journalists, and civil society repression, covering data from 1900–2016 in 201 countries.

Social Inclusion

• Less civic space means that people with less power – the poor, disadvantaged women, rural communities, and indigenous groups–remain powerless. This results in more inequality. A 2019 report covering Bangladesh, Brazil, Cambodia, China, Colombia, Ethiopia, Mozambique, Myanmar/Burma, Nepal, Russia, Rwanda, and Zimbabwe concluded that restrictions on civic space are likely to reverse progress toward reducing inequality, ensuring inclusion, and improving sustainability.^[7] This is because closing civic space, with actions such as threatening, criminalizing, or being violent against civil society actors to delegitimize and stigmatize them, makes it difficult or impossible for marginalized groups to voice their concerns or to organize and empower their members.^[8]
• As civic space narrows, the most marginalized and disempowered groups face a growing risk of being left behind. A study based on research in Brazil, Ethiopia, Pakistan, and Zimbabwe found that legal restrictions to Civil Society Organizations’ (CSO) operations as well as intimidation, criminalization, and surveillance limit their ability to do the following:
  1. influence and critique policy to hold government accountable, and
  2. provide services to people facing poverty or hunger.

  This, in turn, obstructs civil society efforts to reach marginalized groups, widening their exclusion from policy processes and services.^[9]This is especially true when there are penalties for helping groups identified as criminal, such as drug addicts, sex workers, members of the LGBTQ+ community,^[10] asylum seekers, or migrants, to name a few.^[11]

• Restricting online civic space limits inclusive development. A study across ten African countries found that government measures such as internet shutdowns, the taking down of material, and surveillance make it impossible to reach the inclusive, responsive, participatory, and representative governance that SDG 16 aims at. Specifically, the report finds that digital surveillance, disinformation, internet shutdowns, anti-civil society legislation, and arrests for online speech serve to intimidate independent organizations and often amplify social cleavages, resulting in further exclusion.^[12]
• Local-level CSOs ensure inclusion in development. Nepal underwent a political transition in 2006 after ten years of civil war, the development of a new constitution, and rehabilitation after an earthquake in 2015. A case study on the role of civil society organizations during this period highlights that CSOs at the local level are particularly vital for ensuring inclusion.^[13] The study explains that civil society actors at the national level have played an important role in defining political and civil rights in the past three decades, but local and grassroots organizations are perceived to have higher legitimacy and independence from donor-driven agendas and national-level politics. This legitimacy fosters greater participation from communities at the local level and castes that might have been neglected after a conflict and increases the inclusion of their perspectives into decision-making.

Political Stability

• The media contributes to political stability by exposing constituents to different viewpoints. A case study conducted during Uganda’s 2016 parliamentary elections demonstrated that voters who were exposed to electoral messages from multiple candidates were less likely to vote for ruling-party candidates.^[14] This strengthens democratic values, such as electoral competition, reducing the possibility of a dominant-party regime.
• Free debate in media is associated with a more active citizenry. A study across seven Asian and African countries with data from more than 23,000 respondents found increased political participation for the audiences of political debate programs.^[15] See the “Case Study: Media Fosters Political Participation in Asian and African Countries” for more details.
• Crackdowns on the press precede political democratic breakdown:
• • Following an in-country investigation in Turkey, Freedom House analyzed that pressuring the independent media has been part of an overall crackdown on democratic values in the country.^[16]
  • In countries such as those in the Western Balkans, the weakness of the independent media has been identified as one of the principal factors in the overall deterioration of democracy.^[17]

CASE STUDY The Media Fosters Political Participation in Asian and African Countries Action: Governance programs of the British Broadcasting Corporation’s charity BBC Media Action used TV, radio, online, and mobile platforms to produce debate and discussion programs aimed at increasing citizens’ political knowledge and encouraging civic participation. Impact: A study found that the audiences from these programs participate more and have higher political knowledge than those people who do not listen or watch the programs.[18] This study also found that the audiences of these programs were more likely to vote in elections and more likely to participate politically by doing the following: collaborating to solve a community problem, attending a town hall meeting, contacting a local official, contacting a national-elected official, contacting a local chief or traditional leader, or taking part in a protest, march, or demonstration.[19] The study analyzed data from more than 23,000 respondents from Bangladesh, Myanmar, Nepal, Kenya, Nigeria, Sierra Leone, and Tanzania.

 

Literacy

• A free press contributes to higher levels of literacy. A wide-ranging study of sub-Saharan African countries found that in free-press countries, the literacy rate is higher than in countries that do not have a free press.^[20] This study relies on figures from the UNESCO Institute for Statistics finding that the literacy rate for free press countries had risen from 60% in 2002 to 77% by 2009. The authors concluded that “literacy, a free press, and democracy can create a virtuous cycle that lays the ground for improvements in governance.”^[21]

Civic Participation

• Journalism fosters civic participation:
• • UNESCO’s 2022 WorldTrends report describes journalism as a “public good.” It further explains that journalism provides citizens with trusted and fact-based information necessary to participate in a free and open society, and, thus, its role is critical in promoting a healthy civic space.^[22] UNESCO explains that journalism needs a politically and economically favorable environment to increase its independence and quality analysis and information.
• Digital media employees believe that their work fosters civic participation. A 2021 study interviewed over 200 staff members of digital media start-ups across Africa, Asia, and the Americas. It found that 82% of them believed their work had led to positive impacts on society, and 76% answered that their work has increased civic engagement.^[23]

Control of Corruption

• Media Freedom and Corruption

• Press freedom matters in the fight against corruption. The following examples illustrate some of the areas in which this has been the case:
• • Bribery: A 2021 study examining the impact of press freedom on corruption in business found that countries with greater press freedom have significantly fewer incidences of bribery involving public officials and that a free press is associated with a reduction in reported incidents of corruption.^[24]
  • Asset recovery: Multi-country collaborative reporting projects, such as the Panama, Pandora, and Paradise Papers, have exposed corruption on a grand scale, leading to structural reform, as well as the removal of corrupt senior government leaders around the world.^[25] The Organised Crime and Corruption Reporting Project’s latest impact data claims US$7.4bn in fines levied and monies seized as a result of its journalism.^[26]
  • Accountability: The Media Development Investment Fund, which invests in media across the world, has surveyed its investees, finding that 78% of their corruption and accountability reporting had a measurable impact: “media played a central role in uprooting corruption and holding those in power accountable”.^[27]
• Press freedom has a reputational premium. A 2020 study that used a survey of high-level officials in private companies identifies a reputational premium associated with press freedom: corruption perceptions are improved by greater press freedom. This is most evident in countries with low to moderate levels of corruption by global standards.^[28]

• Civil Society Participation and Corruption

• Nongovernmental organizations (NGOs) are an essential element in successful anti-corruption drives. A study analyzed what makes anti-corruption efforts successful, finding that control of corruption requires both the state and society.^[29] The reason that civil society is so important is because NGOs are among the main users of anti-corruption tools, such as transparency laws.^[30]
• Still, civil society on its own cannot successfully fight corruption; it needs an effective enabling environment. A study that reviewed data from 200 countries covering the period of1789 to 2018 points out that civil society on its own is not sufficient: transparency of laws and predictability of enforcement, rigorousness, and impartiality of public administration are also necessary.^[31]

Trust in Government

• Respect for freedoms underpinning civic space increases trust in government.
• • Freedom of expression:
  • • An Indonesian study from 2010 indicated that trust in government had increased with greater democratization and respect for freedom of expression.^[32] This is because freedom of expression provides the opportunity for increased transparency and accountability for government practices.^[33]
    • A study of data from 29 European countries found that respect for freedom of expression improves trust in government.^[34] This paper also studied whether economic prosperity was a good predictor of increased trust in government, finding that it is not significantly related. Thus, in light of the results from this study, protecting freedom of expression plays a more important role in increasing trust in government than economic growth.
  • Civic participation: A study in South Korea found that as participatory political culture grew, confidence in public institutions improved.^[35]

• Nonetheless, trust in government depends on actual performance in its service delivery. A 2022 study found there is a positive impact of promoting civic space if the government actually performs in a way that inspires trust.^[36] The study found that citizens generally trust their leaders if they believe them to be competent, honest, and disinterested.^[37] If there is media freedom and the media report on an ineffective government, then trust will be diminished.^[38]

Public Service Delivery

• Citizen involvement helps the government deliver better services. OECD research has shown that public engagement makes public services more accessible, cost-effective, and efficient.^[39] Engaging individuals and CSOs in the design, production, and delivery of services leads to higher user satisfaction and better value for money.^[40] This is because citizen input helps the government identify the main problems citizens are facing, leads to innovative solutions, and creates shared ownership of public service projects.^[41]
• However, there is mixed evidence in this realm. A 2019 study of 20 lower- and middle-income countries found that interventions promoting direct engagement between service users and service providers—such as in health care—are often effective in achieving improvements in access and quality of service provision.^[42] Nonetheless, if there is a lack of interventions to address bottlenecks in service delivery, citizen engagement alone might not help to improve services. Furthermore, citizen pressures on the government for services where there is no direct interaction between users and providers—such as infrastructure—are not usually able to influence service delivery.^[43]
• Less open societies struggle to deliver high-quality services. Research suggests that countries with more restrictive controls on civic space, such as Rwanda, achieve good ratings on indicators like overall economic growth but struggle to deliver complex public services. This often happens because low-level public officials are instructed to reach specific targets without aiming to meet specific standards for the quality of the services provided.^[44] Research in countries including Pakistan, Zimbabwe, and Cambodia similarly concludes that countries with restrictions on civic space and the media are not likely to deliver high-quality public services because they lack the demands and accountability that civil society provides.^[45]

Health

• Countries’ restrictions on civic space harm their efforts to deal with epidemics. A 2018 study found that restrictions on civic space introduced in Ethiopia, Kenya, and Uganda had a negative impact on their response to HIV/AIDS.^[46] CSOs could play a key role in helping countries achieve the commitment they made in the United Nations to end the AIDS epidemic by 2030. Nonetheless, CSOs’ contribution is currently being obstructed by criminalization and stigmatization, as well as by government barriers to formally registering, accessing financing, and freely operating and assembling.

CASE STUDY South African CSO Helps Save Thousands of Lives Action: Until the early 2000s, the Mbeki administration in South Africa refused to make antiretroviral drugs available.[47] The activist organization Treatment and Action Campaign (TAC) confronted the government by organizing a march and a civil disobedience campaign demanding universal access to HIV treatment through the public health system.[48] Impact: Following TAC’s pressure, South Africa launched a program in 2003 to prevent mother-to-child transmission of HIV and a national antiretroviral drug program in 2004.[49] To date, TAC continues to advocate for access to and quality of health care and has extended its scope beyond HIV to issues such as tuberculosis, gender-based violence, and COVID-19.

Public Safety and Peace

• CSOs have a positive role to play during and after conflict. During conflict, CSOs contribute to peacebuilding in at least three ways:
• • by engaging with most-affected communities and building an evidence base of the consequences of conflict for civilians,
  • by drawing national and international attention to the conflict, and
  • by pushing for a negotiated settlement.^[50]
• In the aftermath of conflict, there is a role for peacebuilding CSOs:
• • to support a nonsecuritized space and
  • to build the capacity of civil society.^[51]
• When CSOs are restricted, there is higher risk of conflict. The UN Special Rapporteur on Freedom of Peaceful Assembly and Association warns that restricting civic space increases the risk of conflict due to tensions that arise from the unaddressed grievances of individuals and the lack of channels to voice them.^[52] Civil liberties scholars, such as Thomas I. Emerson, advocated that freedom of speech—excluding hate speech—acts as a “safety valve,” allowing a civilized way to let off steam.^[53]
• Despite common belief, restricting CSOs does not protect against terrorism. A 2022 study of the effectiveness of 158 laws and regulations over a period of 8 years that restricted NGO activities with the aim of countering terrorism found no evidence whatsoever of decreased terrorist incidents as a result of any of these laws.^[54]
• In fact, discussion freedom can prevent terrorism in democracies. A 2020 study using data for 162 countries found that improving discussion freedom can reduce the risk of attacks on the military and police by almost 50%.^[55] The authors conclude that freedom of expression and discussion can help security and intelligence services get more information about potential threats than they could if public and private discussion was restricted.^[56]
• More media freedom means less state violence. A 2021 multi-country study found that in contexts where the media is restricted, the state can hide and deny state violence. In contrast, where the media is free, the political costs of repression are higher. Thus, media freedom reduces the risk of escalating state violence.
• Freedom of association and assembly play an essential role in nonviolent democratic transitions. A study found that democratic transition through nonviolent resistance increases democratic quality. The main drivers of this democratization are the improvements in freedom of association and assembly.^[57]

   

Debunking the Myth: Autocracies and Development Countries such as China and Vietnam, where CIVICUS rates civic space as “closed,” seem to have achieved positive development outcomes. Rwanda and Ethiopia are named as examples of countries where civic space is rated as “repressed” but still appear to have achieved significant development outcomes.[58] Academics studying this topic have theorized that this might be because ruling elites need to silence critics and repress dissent to buy the longtime horizons needed to achieve economic and social change.[59] Nonetheless, autocracy doesn’t always translate into economic gains and human development benefits. A study based on research in Brazil, Ethiopia, Pakistan, and Zimbabwe points out that restrictions on civil society adversely affected civil society efforts to tackle poverty and hunger and reach marginalized groups. As a result, their exclusion from policy processes and services will deepen, and as civic space narrows, the most marginalized and disempowered groups face a growing risk of being left behind.[60] Additionally, strong economic development doesn’t equal stable leadership in an autocracy. Amartya Sen notes that China has had strong economic development, but it has been fragile and subject to power politics at the top.[61] For example, when universal health care was withdrawn, life expectancy quickly went down.[62] Also, there is an inability to trust growth data from autocracies. A 2017 study has highlightedthe need to cross-check macro-level statistical analysis with household data, because of the high political stakes of data collection and analysis and suspicions that the data may be incorrect or that it is interpreted in ways that are potentially misleading.[63]

 

Limits to the Evidence Although there is growing evidence on the benefits of protecting civic space and the negative impacts of attacking it, there are still challenges to studying it.[64] Conceptual ambiguity: First, “civic space” is a very abstract concept, and despite efforts to define it, there is still disagreement among experts and organizations in the field.[65] Difficult measurement: Also, given that it is an abstract concept and an umbrella term for several rights, academia needs to study it by using proxies. For instance, to know how “strong” civic space is in a country, researchers often study the number of registered civil society organizations and media outlets. This is an imperfect measure, since it does not capture the type of work and impact of these organizations and media outlets. Authoritarian civil society: Often, civic space might be used for the “wrong” reasons. CSOs might advocate for anti-democratic causes and protest for anti-democratic demands. There have been cases where public mobilization has led to authoritarian coups. Also, media freedom can be subject to the spread of disinformation and hate speech, and media outlets can be captured by the state and broadcast government-controlled content, also defeating the purpose of media freedom. Therefore, evidence on civic space must be carefully interpreted, taking all important nuances into account. Linking cause and effect: Research on civic space should account for political context and time lags. A country’s regime history plays a role in how current contexts play out. Also, the benefits of preventive—as opposed to reactive—initiatives to protect democratic freedoms rarely yield immediate effects. Thus, consumers of research on civic space should steer away from simplistic assumptions and interpretations. Public access to media: Countries that aim to protect media freedom should couple these efforts with improving media access. As some of the evidence in this chapter shows, media freedom can improve political knowledge and citizen participation. Nonetheless, this benefit is contingent on citizen access to media outlets.

 

Debt Transparency

This chapter is based on a background paper by Jessica Hickle developed to support the Skeptic’s Guide.

The Benefits of Debt Transparency

Public debt allows governments, especially in developing countries, to invest in critical infrastructure, improve service delivery, and contribute to their path to development. Nonetheless, unsustainable debts and conditions can harm more than they help. There is growing evidence that debt transparency can help countries better manage their debt, increase their creditworthiness, and allocate borrowed funds to their intended purposes.^[1]

Definitions^[2]

• Debt transparency: Debt transparency is the availability and publication of debt data and information about borrowing processes, which allows borrowers and creditors to make informed financial decisions and allows citizens and civil society to hold governments accountable.^[3] This includes important documents such as public sector balance sheets, medium-term debt management strategies, a breakout of creditors, and intended purpose and use of resources, to name a few.^[4]

  Debt sustainability: When a government is able to meet all its current and future payment obligations without exceptional financial assistance or going into default.^[5]

  Debt distress: When a government is finding difficulties in fulfilling its financial obligations and might need debt restructuring.^[6] Debt distress is often a warning sign of potential default.

  Debt default: When a government is unable to pay back a loan and its associated interest.^[7]

  Debt restructuring: When a borrower and a creditor negotiate the terms of a loan to help the borrower meet their debt obligations. Debt restructuring can involve a variety of mechanisms, including reducing the percentage of a loan a country is required to pay back and extending debt maturity (delaying the final repayment date).^[8]

  Debt servicing: When a borrower makes payments to meet their debt obligations.

  Sovereign debt/government debt: The money that countries borrow. A negative financing balance is called a “public deficit.”^[9]

  Risk premium: An additional cost beyond the amount of a loan that a creditor can charge for the risk they incur when lending to a risky or unstable government. Creditors tend to charge higher premia when there is a greater risk that the borrowing country will be unable to repay the debt in full.^[10]

Evidence

Here is what the evidence shows on the benefits of debt transparency on the fight against corruption, public accountability, countries’ creditworthiness, debt sustainability, and development investment.

Control of Corruption

• Countries that are more transparent are perceived as less corrupt.
  • A study analyzed 95 countries between 2006 and 2014 and found that fiscally transparent countries are perceived as less corrupt.^[11] To measure fiscal transparency, authors used scores from the Open Budget Index (OBI), which includes measures of debt transparency.^[12] The study also found that transparency in the latter stages of the budget process has the most significant effect on perceived corruption.^[13] The authors argue that information on the final stages of the budget process allows citizens to compare actual spending to the budgeted allocation, thus creating greater accountability.
  • A 2020 study used a sample of 82 countries and found similar results: fiscal transparency reduces the perception of petty and grand corruption as well as of state capture by elites and private interests.^[14] The authors of this study argue that fiscal transparency reduces information asymmetries between citizens and public officials and politicians. The more citizens know, the harder it is for public officials to abuse their power and get away with committing corruption. Thus, transparency has a deterrent effect.^[15]

Government Effectiveness

• Transparent and democratic countries have more effective policies. A study analyzed data for 82 countries and found that 80 percent of them improved their fiscal transparency between 2006 and 2014.^[16] The authors found that this increase in transparency is associated with higher levels of efficiency, measured using the World Bank’s World Governance Indicators.^[17] Nonetheless, this effect might be limited to democratic countries. Another study built on these findings and confirmed that fiscal transparency improves government efficiency, but the effect is only observed in highly democratic countries.^[18] In non-democratic countries, the public is rarely able to sanction politicians for their conduct. As a result, transparency does not increase leaders’ accountability and therefore does not create incentives for state actors to spend more efficiently.^[19]

Citizen Satisfaction

• A lack of fiscal transparency contributes to public dissatisfaction with the government. A study in Portugal used evidence from public opinion survey data to show that the poor management of the mid-2000s sovereign debt crisis contributed to rising discontent with the government and the performance of democracy more generally.^[20] While Portuguese citizens’ satisfaction with the government has fluctuated for the last several decades, aggregate satisfaction with democracy fell to its lowest levels ever recorded following the debt crisis.^[21]

Creditworthiness

• Transparency improves a country’s creditworthiness. A 2015 study found that fiscal transparency can increase credit ratings by between 0.7 and 1 notches (or steps in the credit rating scale).^[22] This result holds in both developed and developing countries, although the mechanisms are different. In developing countries, fiscal transparency affects credit ratings directly by reducing lenders’ uncertainty about a country’s fiscal and financial policies. In developed economies, this effect is indirect: fiscal transparency deters corruption in borrowing, which reduces the risk of default and improves credit ratings.^[23]
• This is also true for local governments. A 2021 study found that increased fiscal transparency improves the credit ratings of state governments in the United States. The findings show that as the level of transparency increases from “low” to “high,” the predicted probability of a state receiving the highest possible credit rating increases from 20 to 80 percent.^[24]
• Debt transparency is a better predictor of creditworthiness than regime type. While conventional wisdom suggests that democracies are more creditworthy than nondemocracies, evidence from a survey of 63 developing countries indicates that debt transparency—encompassing transparent debt data, forward-looking debt management strategies, and public financial processes—improves sovereign creditworthiness.^[25] Debt transparency policies are not exclusive to democratic regimes and could potentially outweigh the negative effects that non-democratic governance has on creditworthiness.^[26]
• Fiscal transparency can lower interest rates. A 2008 study of 103 countries found that better fiscal transparency is associated with lower sovereign interest rates while higher levels of corruption are associated with higher interest rates.^[27] The authors of the study argue that transparency reduces information asymmetry between governments and financial markets, which in turn reduces lenders’ uncertainty. Nonetheless, another study found that this pattern in which transparency is associated with lower interest rates is conditional on other factors.^[28] During times of economic crisis, countries’ fiscal situations, rather than transparency, explained decreased interest rates. The authors hypothesize that when the economy is strong, lenders are willing to take more risks and prioritize good governance in lending decisions. But in times of crisis, lenders are less willing to take risks, and therefore prioritize economic factors in their decisions.^[29]
• Data transparency policies beyond fiscal transparency also lower the cost of borrowing. A 2017 study analyzed data for 52 emerging market economies and found that a country’s subscription to the IMF’s Data Standards Initiatives^[30] leads to a 15 percent reduction of its risk premia.^[31] These findings demonstrate that countries’ willingness to increase transparency is also a positive signal to lenders.

Debt transparency makes development cheaper Debt can help governments invest in critical goods and services, particularly in developing countries. National and subnational governments can borrow from an array of lenders, including domestic sources, such as banks and private funds. However, most governments borrow from international lenders, such as international financial institutions (IFIs) or other governments.[32] Governments incur a cost for borrowing that debt transparency can lower. The cost of borrowing refers to the additional amount a borrower is required to pay a creditor for the amount of the loan. This cost for borrowing varies according to lenders and can include interest rates and risk premia, among other conditions. Lenders usually charge higher premia when there is a greater risk that the borrowing government will be unable to repay the debt in full. Increased information about a country’s or local government’s finances reduces this risk. The evidence supports this assertion. Since 1998, the International Monetary Fund has argued that international financial markets are likely to demand lower premia from governments that are forthcoming about their fiscal position.[33] Thus, debt transparency, and fiscal transparency in general, can lower the cost of borrowing. Figure 1 illustrates this relationship.

 

Figure 1. An opaque vs. transparent supply curve

Debt Sustainability and Financial Stability

While sovereign borrowing has many benefits and is not inherently harmful, taking on high amounts of unsustainable debt presents many risks. As such, governments should seek to keep their total debt burden at manageable levels. The evidence shows that debt transparency improves debt sustainability and reduces the risk of a financial crisis as a result of debt.^[34]

• Debt signals competence when it is transparent and correctly managed. A study of OECD governments found that governments with more fiscal transparency have a lower public debt and deficit.^[35] The authors of this study argue that governments might take on debt to signal that they are competent. Nonetheless, when debt is transparent, citizens can monitor that the resources are correctly allocated and that debt is not simply used as a strategy to appear competent.
• Institutional quality is associated with debt distress. In low-income countries, the quality of policies, including public sector transparency and debt and fiscal policy, is one determinant of debt distress.^[36] Researchers assert that these findings suggest that policymakers should consider institutional quality as a factor in determining how much they can sustainably borrow. A country with strong and transparent institutions may be able to sustain more debt than if the same country had weaker institutions. Countries seeking to improve public service delivery and access to resources through borrowing should therefore consider transparency as a vehicle for securing additional funding.^[37]
• Debt transparency can help local governments sustain their debt. A case study of subnational governments in Pakistan shows that publishing a province’s debt acquisition can help the national government assess if it would be able to bail out the provincial governments if needed.^[38] This transparency also helps shield both the national and subnational governments from a financial crisis that could occur if their debt increases with a drastic change in their currency’s exchange rate.

Development Investment

• Debt transparency can encourage domestic and foreign investment. Research has found that investors are more likely to invest where institutions are transparent and predictable.^[39] A study found that countries that default on their debt experience significantly reduced FDI in the years following a default. This effect increases for countries with a history of recurrent default.^[40]
• Fiscal opacity might decrease economic growth expectations. A study of more than a decade’s worth of data from Brazil’s budget deficit and economic growth found that fiscal opacity creates uncertainty in the private sector’s investment decisions.^[41] The authors of this study measured opacity by looking at the difference between the forecasted deficit, which is estimated by private institutions, and the actual deficit, collected by Brazil’s Central Bank. If the difference between these two figures is large, researchers assume there is fiscal opacity.

Limits to the Evidence More research is needed on the effects of debt transparency in isolation. The evidence presented in this chapter makes a solid argument to support debt transparency. Nonetheless, most of the cited studies rely on measures of fiscal transparency that aggregate a variety of indicators, including debt. Since these studies bundle additional fiscal transparency measures, it is hard to assess whether debt transparency or other aspects of fiscal transparency are responsible for the observed findings. Future research could focus on studying the impacts of debt transparency more particularly. Researchers could expand their studies to other policy areas. Studies on the effect of debt transparency have traditionally focused on its impact on other economic measures, such as the risk of default and the cost of borrowing. In the future, research can also explore the effect of debt transparency on other governance areas, such as its effect on trust in government or public participation. Micro-level research can complement the existing macro-level literature. Most studies analyzing debt transparency use data from dozens of countries to assess global trends. Nonetheless, future research could focus on specific national or subnational governments and their fiscal performance over time to illuminate the mechanisms by which transparency changes outcomes.[42]

Algorithmic Transparency

This chapter is based on a background paper by Matías Valderrama, María Paz Hermosilla, and Romina Garrido with the GobLab at the Universidad Adolfo Ibáñez.

The Benefits of Algorithmic Transparency

Governments are increasingly using algorithmic decision-making (ADM) to aid in processes. Recent examples around the world show how governments apply ADM across a range of public decisions, from scoring bidders for a public contract to assessing if a family is eligible for social benefits or calculating the amount of bail for a person who has been arrested. The widespread adoption of ADM on such sensitive issues has raised concerns about how algorithms inform decisions and whether they are potentially biased. Recent evidence shows that different methods of algorithmic transparency can reduce potential biases, make bureaucratic processes more efficient, and increase the public’s trust in government decisions that rely on ADM.^[1]

Definitions

• Artificial Intelligence (AI): Generally refers to “a branch of computer science concerned with creating machines that can think and make decisions independently of human intervention.”^[2] AI software systems, therefore, “make decisions which normally require [a] human level of expertise.”^[3]
• Artificial Intelligence algorithm: The set of instructions or programming that tells a computer how to learn to operate on its own.^[4]
• Algorithmic decision-making: The use of AI algorithms to process data and conduct statistical analyses to automate or support a decision-making process.
• Algorithmic transparency: The ability of actors internal or external to the development of an algorithm to obtain information, monitor, test, critique, or evaluate the logic, procedures, and performance of an algorithmic system in order to foster trust and increase accountability of the developers or controllers of the system.^[5][6]
• Reactive transparency (or reactive disclosure): “When individual members of the public file requests for and receive information.”^[7]
• Proactive transparency (or proactive disclosure): “When information is made public at the initiative of a public body, without a request being filed.”^[8]

Why should algorithms be transparent? In recent years, international organizations, governments, and relevant specialized institutions have developed principles and standards for ethical artificial intelligence (AI). While these principles differ in their content and scope, almost all of them agree that transparency is a key component of ethical AI.[9] These principles and academic studies have identified several reasons why algorithms should be made transparent. Below is a non-exhaustive list, synthesized from the various principles. This list will evolve as more institutions adopt ADM and more evidence on the benefits of algorithmic transparency becomes available. Algorithms should be transparent: because they affect the lives of millions of people on sensitive issues, such as child protection and employment; to ensure that they do not harm anyone; to ensure that they operate with fairness and equality; to increase people’s trust in them; to ensure governments use them for their intended purpose; to foster innovation by allowing other algorithm developers to replicate what worked; to ensure algorithms and the private providers selling them to the government are not escaping legality; and to give the public a chance to challenge their results.

Algorithmic Transparency Mechanisms

There are many mechanisms that promote algorithmic transparency. These mechanisms vary in terms of who is responsible for the algorithms, when information is made public (before, during, or after a system is put into use), and the intended audience. Below is a summary of three broad categories of methods for algorithmic transparency, along with their subcategories and a few examples of their implementation.

Disclosures

These mechanisms entail the direct disclosure of information about algorithmic systems. Each subcategory differs in terms of who has to make the disclosures, what kind of information is disclosed, and to which audiences disclosures are made.

1.1 Information Requests

What is this mechanism?	Disclosures through requests of information, often referred to as “reactive transparency”
Who makes the information transparent?	Government agencies that are subjects compelled to disclose under freedom of information legislation
What information is made transparent?	This aspect is unclear, as several research studies have shown that current legislation on freedom of information in multiple countries does not specify if government algorithms should be publicly disclosed.[10] Case studies have shown that freedom of information requests for algorithms have been denied based on trade secret grounds.[11] This happens because government software is often outsourced, and disclosing this information could compromise vendors’ trade secrets.
When?	A member of the public submits a freedom of information request
To whom?	The requested information is disclosed directly to petitioners, but in most jurisdictions, the response to requests is also made publicly available.

1.2 Model Cards

What is this mechanism?	Proactive disclosure of basic information about an algorithmic model to increase transparency about how it works
Who makes the information transparent?	The developers of the algorithmic systems: When government systems are outsourced, the government could require vendors to submit a model card.
What information is made transparent?	Academics have proposed a model card template that includes: basic information about the model the intended uses during its development the most relevant factors informing the model, including data on cultural, demographic, or phenotypic groups metrics of the real-world impacts of the model (performance measures, decision thresholds, variability, and so on) or how the model performs differently when considering these groups details of the datasets used in the evaluation and construction of the card information from training data (if possible) results of quantitative analyses ethical considerations recommendations
When?	At the launch of an algorithmic system
To whom?	The general public

1.3 Source Code

What is this mechanism?	The actual publication of the underlying source code of an algorithmic system
Who makes the information transparent?	The developers of the algorithmic systems: When government systems are outsourced, the government could require vendors to submit the source code.
What information is made transparent?	This mechanism entails the publication of the code, but not necessarily the datasets on which this code has been applied. Thus, there are limitations to this mechanism.
When?	At different stages of the algorithmic life cycle
To whom?	This information is mostly useful to other algorithm developers, so the general public might not benefit from knowing the underlying code.

1.4 Algorithm or Artificial Intelligence Registers

What is this mechanism?	Registers are consolidated repositories or directories of information on the algorithmic systems used by governments,[12] often including standardized information about different systems for ease of comparison. Registers are considered a mechanism of proactive transparency.
Who makes the information transparent?	Government agencies or external parties
What information is made transparent?	Currently, there is no consensus on the information that registers should include. Based on case studies, governments can include information that ranges from a list of the algorithms it uses to their main characteristics, their purpose, and their source code. Nonetheless, the International Organization for Standardization (ISO) is preparing a taxonomy of information that should be disclosed to help stakeholders identify and address the transparency needs of AI systems.[13]
When?	At different stages of the algorithms’ life cycle
To whom?	The general public

Explanations

What is this mechanism?	Information that explains how an algorithmic system arrived at a particular decision or outcome to increase the public’s trust in it. For example, in the EU, the General Data Protection Regulation (GDPR) protects individuals from automated decision-making systems that may be opaque or discriminatory and establishes the right to obtain an explanation about such decision-making.
Who makes the information transparent?	The developers of the algorithmic systems: When government systems are outsourced, the government could require vendors to submit the source code.
What information is made transparent?	While there is no consensus on what should be part of an explanation, explanations can include model explanations, outcome explanations, decision and prediction models, scoring processes, etc.
When?	At different stages of the algorithm’s life cycle
To whom?	Though mostly aimed at an audience with a certain level of understanding of algorithms, it can also be used to explain an algorithm’s decision model to the general public.

Evaluations

3.1 Audits

What is this mechanism?	Audits are an analysis of the outcomes of an algorithm to evaluate patterns of performance that negatively affect certain groups or individuals. There are three main types of audits: Bias audits: Actors external to the developers of the algorithm test its outputs to certain inputs.[14]  Regulatory inspection: Regulators or audit professionals assess whether algorithmic systems comply with regulations or standards.[15] Ethical algorithm audits: “Assessments of the algorithm’s negative impact on the rights and interests of stakeholders, with a corresponding identification of situations and/or features of the algorithm that give rise to these negative impacts.”[16]
Who makes the information transparent?	Internal or external parties
What information is made transparent?	Depending on the type of audit, these mechanisms can shed light on the algorithm’s performance and how such performance affects stakeholders.[17]
When?	Once the algorithm is in place
To whom?	Depending on the type of audit, this information can be revealed to the developing company, a competent authority, a specific group, or the general public.

3.2 Impact Assessments

What is this mechanism?	These assessments measure the impacts of an algorithm on vulnerable groups and demonstrate them to a forum that allows the developers or controllers of such a system to be held accountable. These types of assessments are similar to environmental, human rights, and fiscal impact assessments.
Who makes the information transparent?	This might differ depending on the regulation that mandates the assessment. It can be the developer company, the company or agency that uses it, or the public institution that chose it.
What information is made transparent?	This might differ in every context, but it might include a risk assessment of the algorithm, the potential or actual impacts it has, the harmed communities, its unintended consequences, etc.
When?	Before or after the algorithm is deployed (depending on the context)
To whom?	The developing company, a competent authority, the purchasing or user agency, a specific group, or the general public (depending on the context)

Evidence

Although ADM is a technology that has just recently been deployed, there is emerging evidence of its impacts. Here is what the evidence says on the benefits of algorithmic transparency.

Equality

• Algorithmic transparency helps identify biases in algorithms. A bias-focused audit called the “Gender Shades” project evaluated three application programming interfaces (APIs) of facial recognition in classifying faces by gender and race to determine whether there were accuracy disparities. They found that darker-skinned women were systematically misclassified.^[18] These types of audits are the first step toward addressing the potential biases that ADM might have.

Trust

• People trust government decision-making using algorithms more when they understand it. In a 2019 experiment in the Netherlands, a researcher showed citizens two fictional decisions made by ADM: one decision was related to processing a visa application, and the other focused on identifying potential welfare fraud. The characteristics of each of the two scenarios were different in terms of the level of discretion public officials had in each and, more importantly, in the level of information that each scenario provided on its decision-making process. The experiment showed that providing people with access to information on the algorithm had positive effects, but what increased people’s trust in ADM was giving them explanations on how the decision-making processes worked.^[19]

• Information shared through algorithmic transparency should be tailored to its audience for increased trustworthiness. A 2021 study in the United Kingdom gathered a focus group and confronted its members with ADM on police officer allocation, job recruitment, and automatic recognition of vehicle license plates for parked cars. Throughout the experiment, researchers exposed participants to different mechanisms of algorithmic transparency. This study found that as people gained access to more information, they gained a better understanding of how ADM worked and increased their trust in the algorithmic system. Participants expressed a tension between wanting to have a lot of information about the decision-making process and wanting the information to be simple and easy to understand. Thus, researchers and participants suggested a two-tier approach, in which developers share enough information in simple terms so that the public understands their decision-making process. When a more technical audience asks for additional information under this approach, the developers must share it with them. These suggestions combine proactive and reactive transparency methods.^[20]

• When the government adopts ADM without safeguarding public values or taking potential impacts into account, people rate governments worse. An experiment in the United States showed that when governments adopt ADM for the sake of efficiency without considering how to safeguard public values in the decision-making process (such as fairness, transparency, and human responsiveness), people tend to have more negative perceptions about the government. For this experiment, researchers presented a few fictional scenarios in which algorithms had made a decision that breached some of these public values. The results of the experiment show that when people saw that ADM lacked fairness or transparency, their perception of the government worsened.^[21] Another study of New York City’s Algorithmic Accountability Task Force similarly found that the “technology-centric” view of accountability taken by the task force–which focused on listing the technologies used by city agencies–did not result in greater public trust. The author suggests that the lack of trust in this case is likely due to the absence of public engagement in determining whether the tools on the list should be used at all, and the inability of the information provided to identify or address the impact of the technology.

Control of Corruption

• Flagging potential conflicts of interest prevents possible discrimination and helps public officials make better decisions. A 2020 study analyzed SALER, an early warning system for procurement implemented by the government in Valencia, Spain. SALER’s objective is to help public inspectors analyze data on contracts to detect and prevent conflicts of interest and corrupt practices. The study found that transparency helps control possible cases of discrimination in procurement and helps civil servants make better decisions in their detection of corruption. This happens because SALER is meant to be a complementary tool that provides public officials with descriptive statistics and social network analysis—it is not meant to reduce public officials’ discretion or autonomy. Thus, it is both useful for public officials and easy to audit.^[22]

CASE STUDY Algorithm Registers in Local Governments Action: Different cities across the world have implemented algorithm registers, such as New York, Ontario, Amsterdam, Helsinki, Antibes, and Nantes. Impact: These registers vary in their format as well as in the information they include. While some only highlight basic information, such as the name of the algorithmic systems, these registers have allowed the public to access other relevant information, such as their purpose, uses, and implementation, even including the actual code of the systems.

Limits to the Evidence New area of study. The public sector is rapidly adopting ADM. Nonetheless, its adoption is still recent, which makes it difficult to assess the impacts of ADM. Most of the existing academic literature on the impacts of ADM is experimental and scenario-based. As more public agencies adopt ADM for a sustained period of time, there will be more information to assess its impacts. Lack of specificity on transparency mechanisms. Although the tech policy community has increasingly advocated for and addressed algorithmic transparency, there is a lack of a uniform vocabulary for algorithmic transparency and how it is operationalized in practice. A lack of agreement in the definition of algorithmic transparency might pose an obstacle for its study. Future evidence may evaluate the comparative effectiveness of different transparency and oversight mechanisms. Need for a greater body of evidence. Algorithmic transparency might allow journalists and the academic community to gather more data for analysis. As more algorithm developers and public agencies make information about ADM transparent, the research community will be able to assess what works and what does not, and might help developers and users improve their ADM. Ethical implications of experimenting with ADM. Given that public authorities are using ADM for public decision-making in sensitive issues, there are ethical implications to conducting experiments to assess the impact of ADM. While trying to prove algorithmic bias, researchers might expose some groups to the same discrimination they are seeking to prevent. Thus, simulations and scenario-based experiments might be an avenue that the research community can continue to explore.

Parliamentary Transparency

This chapter is based on a background paper by Laura Neuman, developed with support from the National Democratic Institute (NDI).

A Special Approach for Parliaments

OGP’s parliamentary strategy has four key roles for parliaments. These roles are legislation, oversight, opening parliament, and public dialogue.

Opening parliament requires a special approach. Compared with the executive branch, parliaments are more representative. Each parliamentarian is also accountable to smaller constituencies (such as regions or districts with multi-member districts in party-list voting systems). Yet, relative to the executive, they are often weak. Parliaments are under-resourced, intimidated, or compromised by executive branch actors. Because of this weakness, they are less able to check abuses of power. Thus, efforts to make parliaments more transparent should also strengthen their position relative to the executive.

To that end, this chapter of the Skeptic’s Guide has a more mixed set of evidence. Supporting transparency in the operations of the most representative body of government means understanding where it will be most effective.

The Evidence

The research here is conclusive about a few things.

• Without equal access to decision-making, transparency can contribute to cynicism and undue influence.
• Voters expect transparency and that they will have a voice in legislative processes.
• Transparency skeptics may overstate how openness limits negotiation.
• Transparency by itself may change behavior, but it only works when the information is good and the consequences are clear.
• Transparency functions best when there are additional actors, such as parliamentary monitoring organizations.

See Table 1 for an overview. As a note, because this is an understudied area, all relationships are single studies except where noted.

---

Limits to the Evidence

Beyond Policies

The Growing Case for Open Government

The previous chapters focus on the impacts of specific open government policies. This chapter includes additional, more general, evidence that strengthens the case for open government.

The section is broken into three parts:

1. Growing evidence for the importance of free media and civic space
2. Open government, COVID, and public health
3. Democratic dividend: The role of open government

1. The Growing Case for Public Oversight and Civic Space

Open government matters for development outcomes. In 2018, OGP worked with researchers at the Varieties of Democracy (V-Dem) Institute to look at “diagonal accountability.”^[156] V-Dem measures diagonal accountability in terms of media freedom, freedom of expression, independence of public deliberations, and strength of civil society organizations. The researchers found that stronger diagonal accountability is predictive of the following development indicators:

• Lower infant mortality: As countries become more accountable and open through time, their rates of infant mortality decrease. With maximum diagonal accountability, infant mortality is reduced by 10 deaths per 1000.
• Higher average years of adult education: On average, for every 10 years a country has a high level of diagonal accountability, adult education increases by 1 year.
• Inequality and growth: Higher levels of diagonal accountability in a country are correlated with more even income distribution (a lower Gini coefficient) and higher economic growth.

2. Spotlight on COVID-19 and Public Health

A growing body of evidence^[157] shows that democracies do better in pandemics and in public health generally. Articles published in leading journals like The Lancet^[158] and The British Medical Journal,^[159] and prestigious think tanks such as Chatham House^[160] and Our World in Data,^[161] predicted that democracies would do better at dealing with the COVID-19 pandemic. But what is the role of transparency?

• Surprisingly, official COVID-19 casualties are higher for more transparent countries.^[162]
• But, less transparent governments are likely underreporting COVID-19 fatality. When researchers looked at “excess mortality”—-a metric that compares all deaths recorded with those expected to occur—they found that transparency is not correlated with excess mortality. This suggests that less transparent countries may not be accurately reporting deaths due to weaker national statistics systems or potential interference in data collection and publication.^[163]

3. The Democratic Dividend: The Role of Open Government

A growing body of scholarship is looking at the relationship between democracy and development outcomes beyond health. Much of this work has been centered around the data provided by the V-Dem Institute. Highlights from the researchers around this work have been summarized in the Case for Democracy Conference Report. This section highlights findings that show particular relevance to the open government values of transparency, participation, and accountability.

Openness is good for economic growth

• Freedom of expression plays an important role in democracy’s dividends on economic growth. Recent research establishes a positive long-run relationship between democracy and economic growth.^[164] A 2022 study analyzed which specific factors of democracy lead to economic growth and found that it is primarily clean elections and freedom of expression that drive the long-run effect of democracy on economic growth.^[165]

Protecting civic space enhances environmental protection

• Allowing freedom of expression and association is good for the environment. Respect for civil liberties provides environmental NGOs with the necessary conditions to influence environmental policy. A 2020 study found that a 1% increase in civil liberties generates a 0.05% reduction in national CO₂ emissions.^[166]

Democracy supports openness

• Democracies tend to provide more data to international organizations. A study found that when countries transition from autocracy to democracy, they increase their supply of data to the World Bank by 13 percentage points.^[167]
• Democracies are more transparent about their growth. A novel study compared the statistics provided to the World Bank with another measure of economic growth: nighttime lights. They found that measured by an increase in light, indeed, autocratic regimes overstate their annual GDP growth rates by 0.5–1.5 percentage points on average in the statistics submitted to the World Bank.^[168]
• Democratization also leads to higher fiscal transparency. A study found that transitioning from autocracy to democracy improves fiscal transparency by 18 percentage points. The authors explain that this increase is very likely a consequence of free, fair, and competitive elections.^[169]

What Can OGP Do for You?

Participating in OGP provides a unique space to explore solutions for today’s challenges and make governments more open, inclusive, participatory, and accountable to their citizens.

OGP is a proven platform for collaboration between government and civil society

OGP members co-create action plans and engage in regular dialogue with civil society through platforms like multistakeholder forums on open government. The recently released OGP Vital Signs report, which reviews the last 10 years of OGP data, shows statistical evidence that the strength of this civil society engagement predicts positive outcomes.^[170] When civil society participates in iterative dialogue and agenda-setting with the government, commitments tend to be more ambitious. The OGP platform brings together a diverse set of stakeholders. Over the last decade, OGP has not only increased its membership, but also expanded the number of branches and levels of government involved in its process and participation is becoming more inclusive and equitable.^[171]

OGP is a platform for advancing policy areas

OGP members have shown “real-world” improvements in specific policy areas such as fiscal openness and open contracting. OGP countries that have made ambitious commitments across multiple action plans to disclose budgetary information have become more open than other OGP countries in these areas, based on third-party data.^[172] Commitments related to anti-corruption topics, such as open contracting, regularly translate into early results.^[173] Justice-related commitments have increased in frequency dramatically over the last several years. Specifically, around access to justice, a number of OGP members have put forward people-centered justice commitments to address barriers and systemic problems.^[174] These efforts demonstrate the power of OGP’s global platform and domestically-owned action plans to improve outcomes for citizens.

OGP supports learning and results

OGP is not a talk shop. OGP membership means being accountable for working toward real results. OGP’s Independent Reporting Mechanism (IRM) holds governments accountable by assessing members’ action plans and allows stakeholders to monitor progress in participating members. The IRM also provides key recommendations for improvement and learning. Implementing these recommendations is associated with higher-quality action plans and processes.^[175] Over the past decade, there have been hundreds of completed commitments and rates of early results (concrete changes in government practice related to access to information, civic participation, and/or public accountability as a result of a commitment’s implementation) remain steady.
 

Acknowledgements

The Open Government Partnership Support Unit developed this publication in 2022. Paulina Ornelas was the lead author with support from Sandy Arce, Joseph Foti, and Chathan Raj, and Christina Socci.

The OGP Support Unit commissioned several background papers that form the backbone of this guide. The Support Unit thanks all of the authors of the background papers. All errors of fact and interpretation in the Skeptic’s Guide remain those of the OGP Support Unit alone. Please find each of the input papers linked below.

Access to Justice
Weston, Mark. The Benefits of Access to Justice for Economies, Societies and the Social Contract: A Literature Review. Open Government Partnership and Pathfinders for Peaceful, Just, and Inclusive Societies, 2022. https://www.opengovpartnership.org/documents/the-benefits-of-access-to-justice-for-economies-societies-and-the-social-contract-a-literature-review/

Fiscal Openness
Haus, Martin, Joachim Wehner, and Paolo de Renzio. (When) Do Open Budgets Transform Lives? Progress and Next Steps in Fiscal Openness Research. International Budget Partnership, Global Initiative for Fiscal Transparency and Open Government Partnership, 2022. https://www.opengovpartnership.org/documents/when-do-open-budgets-transform-lives-progress-and-next-steps-in-fiscal-openness-research/

Open Contracting
McDevitt, Andrew. State of Evidence: Open Contracting. Open Government Partnership and Open Contracting Partnership, 2022. https://www.opengovpartnership.org/documents/state-of-the-evidence-open-contracting/

Social Audits
Pande, Suchi. Social audits in service delivery: An annotated bibliography. Open Government Partnership and Accountability Research Center, 2022. https://www.opengovpartnership.org/documents/social-audits-in-service-delivery-an-annotated-bibliography/

Grievance Redress Mechanisms
Pande, Suchi and Naomi Hossain. Grievance redress mechanisms in the public sector: a literature review. Open Government Partnership and Accountability Research Center, 2022. https://www.opengovpartnership.org/documents/grievance-redress-mechanisms-in-the-public-sector-a-literature-review/

Civic Space
Noorlander, Peter. The Impact of Protecting Civic Space: Review of the State of the Evidence. Open Government Partnership and the American Bar Association, 2022. https://www.opengovpartnership.org/documents/the-impact-of-protecting-civic-space-review-of-the-state-of-the-evidence/

Debt Transparency
Hickle, Jessica. State of the Evidence: Debt Transparency. Open Government Partnership, 2023.

Algorithmic Transparency
Valderrama, Matías, María Paz Hermosilla, and Romina Garrido. State of the Evidence: Algorithmic Transparency. Open Government Partnership and GobLab (Universidad Adolfo Ibáñez), 2023.

Design
Stacey Simmons

Copy Editing
Amalia Pleake-Tamm
Lora Bolton

Reviewers
Sandy Arce, Shreya Basu, Tonu Basu, Maaike de Langen, Paolo de Renzio, Joseph Foti, Jonathan Fox, Juan Pablo Guerrero, Munyema Hasan, Martin Haus, Naomi Hossain, Maha Jweied, Adna Karamehic-Oates, Paul Maassen, Lindsey Marchessault, Jose Maria Marin, Andy McDevitt, Allison Merchant, Georg Neumann, Marissa O’Neill, Ivy Ong, Suchi Pande, Rosario Pavese, Paula Pérez, Vivek Ramkumar, Camila Salazar, Christina Socci, Helen Turek, Joachim Wehner, Mark Weston, and Katherine Wikrent.

License
This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Quote Box Photos
Bon Moya: Maria Teresa Magno-Garcia
President Abinader: Wikimedia Commons
Secretary Madeleine Albright: Wikimedia Commons – Public Domain
Victoria Alsina-Burgués: Generalitat de Catalunya
Steve Potter: Open Government Partnership
Isaac Aidoo: Isaac Aidoo
Lysa John: CIVICUS

ENGAGE WITH US

If you have any questions about this product or if your organization is staying on top of the state of evidence on open government policies and you would like to collaborate with the Support Unit, please reach out to: research@opengovpartnership.org.

Downloads

Download a PDF – English

Download a PDF – Spanish

Download a PDF – French

Download a PDF – Portuguese