Last November, in the context of the Open Government Partnership (OGP) Americas meeting in Buenos Aires, a group of activists, civil servants and experts gathered over lunch to discuss the status of open data in the region, particularly in relation to accountability efforts. Or, in other words, has the open data agenda delivered in terms of promoting accountability and impact? As you would expect, there was a mix of opinions with some participants arguing that open data was delivering some results, while others were far more skeptical about the role that open data play has played in this conversation. In this post, I offer my thoughts on this matter taking advantage of the valuable feedback our colleagues provided.
First, we are not dealing with a new discussion. Transparency and accountability are intertwined and, as this early paper from Jonathan Fox notes in 2007, transparency (understood as the free flow of information) could lead to “soft” or “hard” accountability. It could also lead to governance arrangements remaining the same. It often happens that right to information law requests uncover stories that are quite powerful and would demand an explanation from national or local authorities, and yet there is no visible effect. When you extrapolate this (very old) discussion to the use of data the same problem emerges: if data is obtained and released and nothing changes, what was the point of transparency in the first place?
I would argue that when incorporating the data (and more generally ICT into the mix) some things change. The region is full of examples of research performed by young news enterprises and NGOs such as Ojo Público in Perú, established newspapers such as La Nación in Argentina, or new ventures such as Chequeado; all of them using data to perform accountability functions in their democracies. In other words, the “data world” is aiding a new generation of NGOs, businesses and leaders to develop practices that can be shared across regions and can help to compare and connect problems, and develop collaborative solutions. The later scandals around the Panama papers (based on a giant database coming from various jurisdictions) is another good example. In this way the discussions and the problems, are not new but the methods and the potential to scale and collaborate are indeed new as the cost of sharing and connecting is now very low. Further with a few datasets available, and others obtained through traditional means, certain organisations can act as a powerful engine to start new initiatives. New methods, new actors, new ways (this is all good news) but if the fundamentals of accountability are not addressed the problem remains the same.
Open Data initiatives supporting fiscal transparency, open contracts, service delivery and anti-corruption efforts were (and still are) very much in vogue. All the participants found value in these initiatives but with a general caveat: use of data to achieve accountability and thereby impact is what matters, not just the releasing data for its own sake. Since the early days of the open data movement, back in 2010, a small group of people (myself included) have been repeating that what matters is the people around the data, as well as the challenges you were trying to solve, and not the data per se. Even with all the data available about one issue, a group of people need to make sense of it and use it. The more complex the topic is, the more challenging this task becomes demanding several uses of data by different actors in a given system. Some will act raising the alarm and will perform in-depth research activities. Others will assemble evidence to potentially carry a remedy to a potential wrongdoing and taking the issue to a resolution mechanism. Once the problem is exposed other actors will try to correct it. But all these behaviours happen in a given system (or country) in a way that is not linear, and according to the institutional framework available and the capacities available to the key stakeholders in that country context.
If use is important, what is the role of releasing and structuring open data? In the early days of the open data movement a few of us noted that data should not be considered oil, or gold (not precisely the most graceful analogy considering the history and present of these industries), but as infrastructure that countries could build. It is a less sexy analogy but probably more useful. Currently, the Open Data Institute (ODI) in London also shares this view. Take for instance the case of Paraguay. In 2004 Transparency Paraguay (a chapter of Transparency International) celebrated when they managed to access and publish the list of suppliers for the Paraguayan government. In itself this was a huge achievement as gathering this information was excruciating and the process did not please the Paraguayan government. Today this data is a click away, in a standardised format. As you are probably guessing Paraguay still faces serious issues in terms of transparency and accountability, and publication of important information did not act as a magic fix. Other efforts, such as the current experiment that the Open Data Charter is advancing in Mexico, could be good examples of making relevant data available. All these efforts have some value, often helping the administration to get the house in order, but do not necessarily lead to greater accountability or impact.
Will adopting a given open data standard be enough to deliver greater accountability? The same question could be posed 10 years ago where RTI laws were supposed to deliver a massive change in government across the globe. RTI laws are legal standards on how government releases public information and occasionally are the institutional bedrock for open data policies. We are certainly better off with the existence of an RTI law, but I think all commentators and academics would agree that the effects of these laws were not a radical change in the way government operates in Latin America. This line of argument also applies to technical standards, with the caveat that standards are more malleable, easy to engage and often encounter a different community of practitioners. But as all participants agreed, standards are not the magic wand for accountability. In Europe or Latin America is often equally daunting to find and compare meaningfully public contracts, let alone to advance accountability.
Should we abandon open data standards and in general publishing information and focusing on the use gap as some institutions are now suggesting? This is a false dilemma. We probably need to focus on better data and information and better use. It takes two to tango. Nevertheless, we need to consider what is going to be effective in the field, and the level an initiative or project is operating. In other words, who is using the data, for what purposes and how is this relevant in terms of improving a given polity. There is an interplay between the global, regional and national sphere. In the current stage, I advocate for a more localised approach that helps us to understand better the causal chain(s) connecting data, accountability, and impact.
There is now some good evidence of what works and does not work in the developing world (Asia and Africa and Latin America) in terms of transparency and accountability. The latest summary of MAVC project is an interesting read. In my view further research is needed to a) determine the roles enablers should play when promoting the use of data for accountability b) to determine which are the most effective strategies actors can deliver c) to determine what “success” looks like in this field. For instance, are we looking to establish better governance as an end in itself or as a contribution to addressing development challenges? By all means, this is an interdisciplinary effort and as such a complex one. It is likely to yield results when owned and delivered by local or national stakeholders taking into account a broader governance setting. This is hard work and just releasing a few datasets won’t fill the gap, just as passing RTI laws in the 90s didn’t. How to tackle these problems, will depend on a mix of international, regional and local initiatives. The next OAS Summit of the Americas, as well as the next International Open Data Conference in Argentina in September 2018, are good venues to advance this issue and explore how to meaningfully tackle it. Not an easy gap to fill.
Fabrizio Scrollini is the executive director of the Open Data Latin American Initiative (ILDA), and member of the Open Data Network for Development (OD4D). @Fscrollini
The author is thankful to the comments made by Alan Hudson, Silvana Fumega and Michael Jarvis as well as the notes from Alison Miranda, and the insights provided by the participants in the OGP meeting in November 2017. The views in this post remain the sole responsibility of the author.