Open Data in Developing Economies: Toward Building an Evidence Base on What Works and How

Front matter

Open Data in Developing Economies Toward Building an Evidence Base on What Works and How AFRICAN MINDS 9 781928 331599 U User Research C Causes and Context Di Data Infrastructure Dh Data Holders Ds Data Security Pr Privacy Concerns Rf Refinement Pu Public Infrastructure Se Skills & Expertise I Intermediaries Dm Poor decision-making due to faulty information Bg Benefit and Goals Lp Tech Literacy & Internet Penetration Fl Feedback Loops De Domain Experts Pa Entrenching power asymmetries Da Data Audit and Inventory Rb Cultural/ Institutional Roadblocks Rs Resource Availability & Sustainability Co Collaborators Ow Open washing Rm Risk Mitigation M Performance Metrics Problem and Demand Definition Capacity and Culture P artnerships RisksGovernance Fi Freedom of Information Od Open by Default Dq Data Quality R Responsiveness Stefaan G. Verhulst & Andrew Young Recent years have witnessed considerable speculation about the potential of open data to bring about wide-scale transformation. The bulk of existing evidence about the impact of open data, however, focuses on high-income countries. Much less is known about open data’s role and value in low- and middle-income countries, and more generally about its possible contributions to economic and social development.

Open Data for Developing Economies features in-depth case studies on how open data is having an impact across the developing world—from an agriculture initiative in Colombia to data-driven healthcare projects in Uganda and South Africa to crisis response in Nepal. The analysis built on these case studies aims to create actionable intelligence regarding: • the c onditions under which open data is most (and least) effective in the development process – presented in the form of a new Periodic Table of Open Data; • str ategies to maximize the positive contributions of open data to development; • and means f or limiting open data’s harms on developing countries. “ A n empirically grounded assessment that helps us move beyond the hype that greater access to information can improve the lives of people and outlines the enabling factors for open data to be leveraged for development.”—Ania Calderon, Executive Director, International Open Data Charter “ T his book is compulsory reading for practitioners, researchers and decision-makers exploring how to harness open data for achieving development outcomes. In an intuitive and compelling way, it provides valuable recommendations and critical reflections to anyone working to share the benefits of an increasingly networked and data-driven society.” —Fernando Perini, Coordinator of the Open Data for Development (OD4D) Network, International Development Research Centre, Canada

Open Data in Developing Economies Stefaan G. Verhulst and Andrew Young AFRICAN MINDS Toward Building an Evidence Base on What Works and How

CHAPTER 2 The Relevance of Open Data for Developing Economies What is Open Data?

In this book, open data is defined as follows: Open data is publicly available data that can be universally and readily accessed, used and redistributed free of charge. It is structured for usability and computability.5 Not all forms of data shared actually possess all the attributes included in this definition, nor do they necessarily conform to all the principles found in the Open Data Charter.6 In many ways, this is a gold-standard definition of open data, an important target to work toward. In fact, the openness of data exists on a continuum, and many forms of data that are not strictly “open” in the sense defined above are nonetheless shareable and usable by third parties. It is this broader sense of “open” that is used in this paper. Open data exists in a wide variety of fields and domains. Three sectors in particular are responsible for producing the bulk of open data: governments, scientists, and corporations. In this paper, we focus mainly on the release and use of government data. We acknowledge, however, the importance and often untapped potential of more open access to science data and corporate data. Those other data sources, as well as crowdsourced data collection are also often mashed up with open government data, supplementing official public datasets to create new insights, opportunities, and impacts as a result. In what follows, we deconstruct the main reasons why open government data matters to developing economies. 5 GovLab, “Open Data: What’s in a Name,” January 16, 2014, GovLab Blog, http://thegovlab.org/opendata-whats-in-a-name/. 6 Open by Default; Timely and Comprehensive; Accessible and Usable; Comparable and Interoperable; For Improved Governance and Citizen Engagement; For Inclusive Development and Innovation. http:// opendatacharter.net.

The literature on open data reflects considerable enthusiasm about the potential for open government data in development. For example, a recent report published by the Open Data for Development Network suggests that open data is central to the development community’s goals of “enabling widespread economic value, fostering greater civic engagement and enhancing government transparency and accountability to citizens.” The report goes on to argue that “open data is increasingly recognized as a new form of infrastructure that is transforming how governments, businesses, and citizens are organized in an increasingly networked society.”7 We do find some evidence to support such enthusiasm: across sectors, we see signs that open data can indeed spur positive economic, political, and social change. On the other hand, we also find grounds for caution; the impacts of many of the projects we examined remain largely aspirational or speculative, and some cases even led to harms (or potential harms). Although the real-world impacts of open data in developing economies remain emergent, it is important to distil these early lessons and develop a frame of analysis to support the current window of opportunity to increase access to data sources—a window that is likely to close absent any further evidence of open data’s impact or a better, more targeted description of the value proposition and change theory driving the field. What makes open data uniquely relevant to developing economies?

We live in an era of big data. Every day, an unprecedented amount of information is being generated by an ever-increasing diversity of devices and appliances. Today, a growing consensus exists that this data, if applied correctly, and with attention to the attendant risks, can help spur positive social change. Sometimes called the “data revolution,” 8 this new paradigm often fails to distinguish between the benefits of data per se and the complementing benefits of unlocking government data.9 7 Open Data for Development Network, Open Data for Development: Building an inclusive data revolution, Annual Report, 2015, 11, http://od4d.com/wp-content/uploads/2016/06/OD4D_annual_report_2015.pdf. 8 See United Nations Independent Expert Advisory Group on a Data Revolution for Sustainable Development, “A World That Counts, Mobilizing the Data Revolution,” 2014, http://bit.ly/2am5K28. 9 At the same time, there is a growing recognition of open data’s potential to meet the Sustainable Development Goals (SDGs). The World Bank, for instance, has explored the various ways in which open data could help to make progress toward the SDGs. Similarly, a recent White Paper by the Open Data Institute (ODI) concludes: “Open data can make an impact across the globe. Its role in combating development challenges of the next 15 years, both as a tool for measuring progress and in finding solutions, is becoming more clear.” See The World Bank, “Open Data for Sustainable Development,” Policy Note, August 2015, http://bit.ly/2aGjaJ4; and Open Data Institute, Supporting Sustainable Development with Open Data, 2015, 3, http://theodi.org/supporting-sustainable-development-with-open-data.

Chapter 2 – The Relevance of Open Data for Developing Economies THE UNIQUE FEATURES OF OPEN DATA EQUALITY SCRUTINY TRUST PARTICIPATORYVALUE AMPLIFIER FLEXIBILITY OPEN GOVERNANCE & OPEN DEVELOPMENT Figure 1. The Unique Features of Open Data Based on our examination of the narratives and evidence provided in the existing literature, six distinguishing features seem to be credited to open data. Although these characteristics are unique to open data, in many cases, they would not be possible without a broader data, technology, and innovation ecosystem. Across our case studies, we’ve seen that the existence of a strong information and communication technology for development (ICT4D) sector in a country, for example, tends to result in higher impact, more quickly developing open data efforts.

With the understanding in mind that open data must exist in a strong ecosystem, the six distinguishing features that are most quoted with regard to open data in a development context include: • Scrutiny: Because open data is subject to greater scrutiny and exposure than inaccessible institutional data, there is potential for enhanced review and improvement of government data quality (e.g., by dataliterate civil society groups or other crowdsourced methods). This can result in more useful data—again, a benefit that is relevant in less developed countries and societies, where data is scarce, and of limited quality and usefulness.10 10 Tim Davies, for instance, notes that, “researchers and other users outside of government may highlight inaccuracies and inconsistencies between datasets” as a result of access to open data, and thereby improving data quality and usefulness. Tim Davies, “Open Data in Developing Countries: Emerging

• Equality: Open data can lead to an inherently more equitable and democratic distribution of information and knowledge. This is a key intended benefit in all countries, but particularly salient in many developing economies that struggle with large socioeconomic and digital divides.11 It is important to keep in mind, however, that the lack of Internet penetration and access to tools for using and accessing open data still present challenges in many contexts—and, indeed, such technological inequities can be further entrenched through open data in some cases. • Flexibility: Open data is open with regard not only to the information it contains, but also to its format. This means that, when released in a usable manner, open data can be easier to repurpose and combine with other pieces of information than data institutions fail to make accessible, which in turn means that it is more flexible, with secondary uses that are likely to yield innovative insights. This is true of data from all sectors, but perhaps especially of government data, which often exists in vast, untapped silos; opening that data (turning it “liquid”) can play a key role in generating new insights and policies. 12 Such liquidity can only become a reality if data, and the tools used to manipulate it, are interoperable and adhere to agreed upon standards. Creating such technical capacity can, however, lead to opportunity costs and require significant upfront resource allocation on the supply side, potentially slowing progress at the outset. • Participation: By facilitating citizen participation and mobilization, open data can allow a wider range of expertise and knowledge to address and potentially solve complex problems.

This quality of “open innovation” can allow resource-starved developing economies to access and benefit from the best global minds and expertise. It can offer a more participatory way of solving complex public dilemmas, with pathways toward more easily tapping into previously inaccessible knowledge (e.g., those related to social and economic development).13 insights from Phase I,” Open Data Research, July 15, 2014, http://opendataresearch.org/sites/default/files/ publications/Phase 1-Synthesis-Full Report-print.pdf. 11 Aspasia Papaloi and Dimitris Gouscos, “Parliamentary Information Visualization as a Means for Legislative Transparency and Citizen Empowerment,” Journal of Democracy and Open Government, 2013, http://www.jedem.org/index.php/jedem/article/view/222/183. 12 “By using common open repositories, public administrations can save time and money from the automisation of internal data exchange,” Raimondo Lemma, Federico Morando and Michele Osella, “Breaking Public Administrations’ Data Silos: The case of Open-DAI, and a comparison between open data platforms,” Journal of Democracy & Open Government, 2014, http://www.jedem.org/index.php/jedem/ article/view/304. 13 Jae-Nam Lee, Juyeon Ham and Byounggu Choi, “Effect of Government Data Openness on a Knowledge-based Economy,” Procedia Computer Science, 91, 2016. http://www.sciencedirect.com/science/ article/pii/S1877050916312364.

Chapter 2 – The Relevance of Open Data for Developing Economies • Trust: Because it increases transparency and avenues for citizen oversight, unlocking data can lead to higher levels of accountability and trust throughout societies and countries. 14 This “sunlight” or “trust” quality of open data can have powerful ripple effects, including incentives for better government practice, and the enhancement of the quality of public life and citizenship. Such increases to trust and accountability rely on meaningful data being made open, however, rather than governments participating in “open washing” where largely useless datasets are made accessible toward boosting institutional reputation. • Value amplifier: Finally, it is now widely recognized that data is a new kind of asset or knowledge is a form of wealth. The opening of government datasets in a flexible and equitable manner can amplify the value of data thanks to data filling important data gaps felt in society. Though this attribute is important across the world, it may have a particularly important role to play in developing economies. In its 2016 World Development Report, The World Bank pointed out that technology can play an “accelerator” role in developing countries. 15 But while the inherent scarcity of resources (data and otherwise) in the developing world increases the apparent value and potential impact of open data, cultural and political barriers to timely and well-targeted open data provision efforts could slow progress.

These narratives surrounding the open data movement reflect those associated with the cross-sector paradigm shift from closed processes to open ones, and how it applies to governance and development. Software, for example, is increasingly developed in an open source manner. With the rise of the collaborative coding platform GitHub, a notable driver, 16 the open source movement, similar to open data, is seen to be providing for more equal and flexible ways to create and access code—resulting in distributed coders, not just tech company employees, creating and improving exciting new products. Similarly, businesses and governments alike are embracing open innovation techniques, posing opportunities to the crowd to provide input on important challenges and absorbing the best ideas—providing for enhanced participation 14 Ali Clare, David Sangokoya, Stefaan Verhulst and Andrew Young, “Open Contracting and Procurement In Slovakia,” GovLab, Accessed October 25, 2016, http://odimpact.org/case-open-contracting-andprocurement-in-slovakia.html. 15 World Bank, World Development Report 2016: Digital Dividends, World Bank, 2016, 4, http://documents. worldbank.org/curated/en/896971468194972881/pdf/102725-PUB-Replacement-PUBLIC.pdf. 16 Laura Dabbish, et al., “Social Coding in GitHub: Transparency and collaboration in an open software repository,” Proceedings of the ACM 2012 Conference on Computer Supported Cooperative Work, February 11, 2012, http://dl.acm.org/citation.cfm?id=2145396.

and scrutiny, other features of open data.17 The emerging fields of open governance and open development are also built on similar principles and techniques (see text box below).

OPEN GOVERNMENT AND OPEN DEVELOPMENT Open Governance/Government18 Definitions of open governance or open government vary not only across sectors but within them. Definitions focus to varying degrees on the key elements of transparency, citizen participation, and collaboration, among others, depending on the context. Some illustrative examples of “open government” definitions19 include: • Wallace Parks, “Open Government Principle: Applying the right to know under the Constitution” (1957)—“The general availability of government information is the fundamental basis upon which popular sovereignty and the consent of the governed rest, subject to several important restrictions on this general rule (i.e., to allow for the carrying out of the constitutional powers of the Congress and the President; to protect the personal and property rights of individuals, corporations and associations; to acknowledge administrative complications as to whether to release, to withhold, or to partially release particular types of information under particular conditions; to protect confidentiality of communications internal to government; to acknowledge the difficulty of segregating information when parts of a document should be released and parts withheld).”20 • White House, “Transparency and Open Government: Memorandum for the Heads of Executive Departments and Agencies” (2009)—“Open government is defined as a system of transparency (information disclosure; solicit public feedback), public participation (increased opportunities to participate in policymaking), and collaboration (the use of innovative tools, methods, and systems to facilitate cooperation among Government departments, and with nonprofit organizations, businesses, and individuals in the private sector).”21 17 Dietmar Harhoff and Karim R. Lakhani (eds.), Revolutionizing Innovation: Users, communities, and open innovation, Boston: MIT Press, 2016. 18 See GovLab, “Open Data: What’s in a Name,” January 16, 2014, GovLab Blog, http://thegovlab.org/opendata-whats-in-a-name/. 19 See also (in Spanish), Silvana Fumega, “Algunas Ideas para ‘Debates Conceptuales sobre el Gobierno Abierto,’” September 25, 2016, http://silvanafumega.blogspot.fr/2016/09/algunas-ideas-para-debatesconceptuales.html. 20 George Washington Law Review 26, no. 1 (October 1957), http://heinonline.org/HOL/Page?handle=hein. journals/gwlr26&div=10&id=&page=&collection=journals. 21 At: https://www.whitehouse.gov/the_press_office/TransparencyandOpenGovernment.

Chapter 2 – The Relevance of Open Data for Developing Economies • Open Government Partnership, Open Government Partnership Declaration (2011)—“Open government involves: Increasing the availability of information about governmental activities; Supporting civic participation; Implementing the highest standards of professional integrity through: Increasing access to new technologies for openness and accountability, information sharing, public participation, and collaboration.”22 Open Development23 In the wake of open government taking hold as an organizing concept for improving and innovating governance, open development has evolved as a more networked and innovative pathway to improving international aid and development efforts. In a book on the topic, the International Development Research Centre (IDRC) seeks to gain clarity on the contours of the field of open development. 24 Their assessment of the theory and practice involves a number of key elements present in open development work, including: • The power of human cooperation • Sharing ideas and knowledge • The ability to reuse, revise and repurpose content • Increasing transparency of processes • Expanding participation • Collaborative production Based on the examination of these strands of openness in development efforts from the World Bank, ONE, African Development Bank, and others, the IDRC authors conclude that the central idea behind open development is: “harnessing the increased penetration of information and communications technologies to create new organizational forms that improve the lives of people.”25 22 Open Government Partnership, “Open Government Declaration,” http://www.opengovpartnership.org/ about/open-government-declaration. 23 See Matthew L. Smith and Katherine M.A. Reilly, eds., Open Development: Networked innovations in international development, MIT Press, 2013, http://idl-bnc.idrc.ca/dspace/bitstream/10625/52348/1/IDL- 52348.pdf. 24 Ibid. 25 Ibid.

The Data Life Cycle An important insight from the emerging research and practice is that data is not “a thing” but involves a “process”—what we call a “data life cycle.” COLLECTION PROCESSING SHARING ANAL YZING USING Figure 2. Data Life Cycle How each stage of the data life cycle is implemented—from collection to processing and sharing; to analysis and using; and back to the start of the cycle again—will determine the value of data and who ultimately benefits. Disparities among those who collect and have access to data or have the capacity to make sense of the data can reinforce existing imbalances in power or influence. This is especially true in a developing economies context where the number of data holders and data scientists is more concentrated, and this smaller group is disproportionately empowered to make meaningful use of data. Within that context, opening datasets is often characterized as a force for democratization—engaging private and civil society actors, and, often indirectly as a result of intermediation, citizens themselves in analyzing and using data.26 As such, open data provides for unique efficiencies by leveraging civic-minded technologists (government and nongovernment), as well as entrepreneurs, to analyze, disseminate, and/or use data in a new, sometimes profitable way, as discussed more below.

On the other hand, each stage of the data life cycle contains risks. Risks are often the result of technological weaknesses (e.g., security flaws); individual and institutional norms and standards of quality (e.g., weak scientific rigor in analysis); legal confusion or gaps; or misaligned business and other incentives (e.g., companies seeking to push the boundaries of what is socially appropriate). Although there are common elements across these risks, it is useful to examine them by separately considering each stage of the data value cycle. 26 Jonathan Gray, “Five Ways Open Data Can Boost Democracy Around the World,” The Guardian, February 20, 2015, https://www.theguardian.com/public-leaders-network/2015/feb/20/open-data-dayfairer-taxes.

Chapter 2 – The Relevance of Open Data for Developing Economies When risks are not addressed at the initial stages of the value cycle (e.g., when dirty data is not cleaned at the collection stage) they may accumulate and lead to additional risks downstream (e.g., making flawed inferences from the data analysis due to inaccurate data). Therefore, it is important to consider potential risks not just at the points of opening data, but also at the data collection stage and evaluate those risks vis-à-vis the (potential) value of releasing the data. As such, to prevent harm, there may be a legitimate case— especially when there is a clear understanding of the purpose of the use and user—to share certain government datasets with those targeted audiences in a more protected manner to generate necessary insights while limiting the risks. We examine the risks introduced by open data efforts in developing economies in more detail in Part II.

CHAPTER 3 The Impact of Open Data on Developing Economies Capturing impact and developing evidence Many studies of open data are concerned with proving a case—either that open data can spur rapid social transformation (the positive case), or that it has a negligible or harmful effect (the negative case). In truth, the evidence is mixed and emergent; the impact of open data is, in fact, far more ambiguous. Rather than just asking Does open data spur development? we seek in this paper to also ask, How and under what conditions can it work? To answer this question, we’ve examined a wide and variable range of attempts to provide evidence to develop a plausible “theory of change” that would explain the role of open data in development. Theories of change are important. A recent report from the United Kingdom’s Institute of Development Studies points to “the persistence of poorly articulated theories of change that fail to specify realistic causal pathways at the outset” in relation to transparency initiatives.27 Weak theories of change can lead to a variety of false assumptions and misconceptions when it comes to understanding how open data works; these, in turn, can lead to missed opportunities, spending inefficiencies, and a general failure to live up to open data’s potential. A review of the literature shows that numerous pathways and theories of change have in fact been proposed. For example, a recent study conducted by IDRC, the World Wide Web Foundation, and the Berkman Klein Center for Internet and Society at Harvard University cites at least thirteen “theories of change,” including open data’s ability to reduce transaction costs, generate new forms of economic growth and prosperity, generate new revenue models, and disrupt traditional business models. 28 Others point to the social and environmental benefits of open data. For example, Martin Hilbert draws 27 Duncan Edwards and Rosie McGee, eds., “Opening Governance,” IDS Bulletin 47, no. 1 (January 2016), https://opendocs.ids.ac.uk/opendocs/bitstream/handle/123456789/7686/IDSB_47_1_10.190881968- 2016.103.pdf?sequence=1. 28 José M. Alonso, “Measuring Impact of Open Government Data—Open Data Research (South) Meeting Report,” May 24, 2012, http://webfoundation.org/2012/05/odrs-meeting1-report-available/.

Chapter 3 – The Impact of Open Data on Developing Economies attention to the potential of opening geospatial-, education-, and housingrelated information. Based on a review of 180 pieces of literature related to Big and Open Data, he concludes—with caveats we discuss further below—that open data does in fact contain true opportunities for development.29 Open Data in Developing Economies Logic Model In what follows, we describe a change theory for open data using the Open Data in Developing Economies Logic Model (Figure 3). 30 This model suggests that: Open data (supply), when analyzed and leveraged by both governmental and nongovernmental actors (demand), can be used in a variety of ways (actions and outputs), within the parameters established by certain enabling conditions (and disabling factors), to improve government, empower citizens and users, create economic opportunity and/or solve societal problems (impact). It is important to reiterate, however, that these positive impacts are always subject to certain local, context-sensitive enabling conditions and disabling factors. While this logic model presents a general outline of how open data can work, Table 1, below, presents a more detailed explanation of how it interacts with various sector-specific opportunities and challenges to create genuine impact on the ground.

The logic model is built around the premise (informed by the case studies and primed for further experimentation and research) that higher impact open data projects are the result of matching the supply to the demand of data actors who can operationalize open data toward specific activities and outputs. These outputs and activities can, in turn, serve a broader and more diverse group of users and objectives.

Having a logic model allows for a more detailed analysis within and across sectors of open data toward providing a number of highly specific lessons about actors and conditions, opportunities and challenges. 29 Martin Hilbert, “Big Data for Development: A review of promises and challenges,” Development Policy Review 34, no. 1 (January 2016), http://onlinelibrary.wiley.com/doi/10.1111/dpr.12142/full. 30 See also Andrew Young, Stefaan Verhulst, and Juliet McMurren, “The GovLab Selected Readings on Open Data for Developing Economies,” August 1, 2016, http://thegovlab.org/open-data-for-developingeconomies/.

ENABLING/DISABLING CONDITIONS OPEN DATA SUPPLY ACTORS DEMAND OUTPUT Collaboratives Intermediaries Infrastructure Policies Resource Availability Problem Definition GOVERNMENT NON-GOVERNMENT Examples • Analysis • Presentation • Dissemination Examples • Apps • Data-Driven Journalism Policies FEEDBACK Examples • Court Cases on Corruption • Data-Driven Startups Created INDICATORS Improving Governance Empowering Citizens Creating Economic Opportunity Solving Public Problems IMPACT USE / USERS ACTIONS GOVERNMENT NON-GOVERNMENT Figure 3. Open Data in Developing Economies Logic Model Examining the Open Data in Developing Economies Logic Framework The logic model presented in Figure 3 above describes the elements in place across the lifespan of an open data initiative, from the initial input or supply of data through its use and impact, with several enabling conditions and disabling factors influencing impact. We explore each of the elements introduced in the table in more detail below, with attention to the types of impact and the enabling conditions and disabling factors that inform the Periodic Table of Open Data Elements.

Chapter 3 – The Impact of Open Data on Developing Economies Table 1. In-Depth Logic Framework Input (Supply) Actors (Demand) Activity Output Use/Users Indicators Impact Open Government Data “Open data is publicly available data that can be universally and readily accessed, used and redistributed free of charge. It is structured for usability and computability.”

Including, for instance: • GIS Data • Observed Data (e.g., weather) • Financial Data • Administrative Data • Statistical Data • Audit Data NGOs & Interest Groups Researchers and Academia Journalists and Media Outlets Donor Organizations Private Sector— Entrepreneurs and Corporations Government Officials Data Analysis (Methods) Presentation (Visualization) Aggregation and Commingling (Mashups) Dissemination— toward, for instance: • Benchmarking • Hotspotting • Comparing services and performance • Resource allocation • Diagnosis and assessment Decision Trees Maps Apps and Platforms Dashboards Process improvements Data-driven journalism Infographics Searchable databases Policies Advocacy Alerts Other Government Agencies and Officials Nongovernment Actors including, for instance: • Citizens • Researchers • Entrepreneurs • Corporations • Service providers • NGOs & interest groups • Media outlets • Donor organizations The outputs of open data have had several real-world effects, which can be assessed according to a number of indicators, for instance: Accountability • Court cases on corruption Improved service delivery: • Increase in citizens receiving services Increased information sharing: • Collaborations between public entities Enhanced decision-making capacity and choice: • Satisfaction among citizens Social mobilization: • Data-driven advocacy efforts Job creation: • Employment in data-driven fields Frugal innovation: • Data-driven startups Economic growth: • Industry-level growth figures Improved situational awareness: • Time saved responding to emerging situations and crises More expertise and knowledge brought to bear: • Number of disciplines engaged in decision-making processes Targeting interventions and tracking impact: • Assessment of change against baseline indicators Improving Government • Accountability • Improved service delivery • Increased information sharing Empowering Citizens • Enhanced decision making capacity and choice • Social mobilization Innovation and Creating Economic Opportunity • Job creation • Frugal innovation • Economic growth Solving Public Problems • Improved situational awareness • More expertise and knowledge brought to bear • Targeting interventions and tracking impact Enabling Conditions and Disabling Factors • Problem and Demand Definition: whether and how the problem to be addressed and/or the demand for open data are clearly defined and understood • Capacity and Culture: whether and how resources, human capital and technological capabilities are sufficiently available and leveraged meaningfully • Partnerships: whether collaboration within and, especially, across sectors using open data exist • Risks: whether and how the risks associated with open data are assessed and mitigated; • Governance: whether and how decisions affecting the use of open data are made in a responsive manner

Input (Supply) A diversity of data types make up the supply side of open data in developing economies and, as the input, plays a key role in determining the ultimate impact. Data types being made available in developing countries range from information about the planet, such as geospatial and weather information, to information about the workings of government itself, like financial and administrative data. Of importance within a developing country context, involves data that is collected and potentially supplied by international (donor) organizations and civil society (often through crowd-sourced means) that may complement the supply of domestic government data. Much of the focus in the early days of open data (in both developed and developing economies) has fallen on improving the supply side of open data, with the Open Data Charter and Open Government Partnership, for example, pushing government data holders to make certain types of data accessible according to a number of principles and in a standardized way.

Actors (Demand) As the global open data ecosystem matures, a greater focus is being placed on understanding the demand side of open data—the actors who will make use of the information the governments released. As the supply side of open data continues to improve thanks to international standardization efforts, including the Open Data Charter and advocacy at the national and regional level, the demand side stands to benefit through greater engagement of existing demand-side actors, and the identification of additional stakeholders who could make use of that information. Some of the key yet distinctive segments or constituencies (often with different interests and needs) on the demand side of open data include nongovernmental organizations (NGOs), including not only government watchdog groups but also service providers, researchers and scholars, data-driven journalists, entrepreneurs and businesses, and government officials themselves who benefit from more liquid data that has escaped internal silos.

Activities The activities enabled by access to open data are in many ways only limited by the imagination and skills of the actors on the demand side of the equation. Some of the most common ways that open data is used include data analysis to uncover new insights, presentation and visualization to make the information more comprehensible, aggregation and commingling of multiple datasets to