Why I would like to see more critical debates on the ‘open aid data’ discourse
I enjoyed reading the summary of David Eaves’ presentation on aidinfo. Especially the concluding remarks seem to be a good summary of why the discourse on open data seems to gain some momentum, they sound right and they are also right in many ways:
With a good set of examples, patterns and common practices, the use of data will become embedded in development practice. Better information will make aid more effective: data portals improve data literacy, an open culture promotes learning, and using information to add value in the aid sector will ensure data remains open and will promote the release of more [my emphasis].
But then I read ‘Story project aims to put people back at the heart of development’ and it confirmed some of the discomfort I have with the rising popularity of aid data:
Concern that an obsession with numbers is leading to development donors distancing themselves from those they seek to help has led to the creation of a new initiative which seeks to bring stories and story-telling back into the heart of development communication.
(...)
Andrea Cornwall points out that it is not that development does not need numbers. 'Counting the number of people living with AIDS, the number of women dead or maimed because they were denied safe abortions or the number of people unable to access safe water can make policy makers pay attention to issues they might otherwise not make a priority.'
But, she said: 'It is time to speak back to the folly and the arrogance of those who try to reduce the complexities of life to proxies and metrics, and create with their numbers the emperor's new clothes.
I already posted a short piece on my scepticism with ‘better numbers’ and I want to structure my short critique around four issues/themes that I think require a bit more attention and discussion:
1. I’m sceptical because organisations like the World Bank and DFID like it2. I’m sceptical because it’s essentially another attempt to rationalise processes that are inherently political and therefore not simply determined by the best data or the better argument3. I’m sceptical because it prefers data and numbers over qualitative indicators, stories and examples and therefore does not challenge the substantialist development discourse as such4. I’m sceptical because it focuses on the past and ex-post data therefore the focus on current political processes, alternatives in evaluation etc may be diverted
1. This is not about conspiracy theories or a general bashing of large donor organisations. The history of development shows that if ‘everybody’ likes a term/concept/discourse it is probably too broad (see for example ‘participation’ or ‘good governance’), no longer an innovative concept or not challenging to the fundamentals of how development is ‘done’. I have a feeling that maybe all three elements may be true to some extent when everybody is starting to like aid transparency. I am not saying ‘do not publish data’, but I am still missing a critical debate on the ‘small print’, e.g. questions on what counts as ‘data’ and whether engaging with numbers is sufficient. Is there a danger of ‘data’ as the new ‘tyranny’, as two well-known books on the participation discourse have inquired?
2. Policy would be so much better and more ‘effective’ if we had all the/more information. Since the early days of public policy analysis (e.g. Aaron Wildavsky’s work) this seems to be one of the holy grails. But more aid data will not substitute the reality that many political decisions are not based on ‘facts’. Maybe a look into DFID’s budget will reveal some interesting, misguided spending, but I am pretty sure that 2 senior bureaucrats, 3 academics and 1 external advisor were against this project in the first place. It happened anyway, regardless of more, better and critical data. There was a political will to support this country, that government or this initiative and all the data will tell you is probably that political will and effective aid very often do not make good partners.
3. Related to introductory point on the power of stories, more aid data are not likely helping to challenge the ways big organisations are thinking about aid. Maybe it can ‘improve’ the managerial, substantialist approach to aid, but similar to, say, numerous UN reforms in the past it may only scratch at the surface rather than changing the membership of the Security Council. Many of the ways ‘development’ thinks, reports or learns need to be addressed and I think aid data can only have a limited effect on more fundamental flaws of the aid system.
4. Most of the aid data is collected/published ex-post. Interesting for researchers, but for many others they will be old data. Nothing can be changed, nobody is likely to be taken to account and as political processes go, they are difficult to forecast. Elections, changes in the bureaucracy, a new ‘white paper’ all can have an impact-and they are likely to produce more ‘better’ data or create new mistakes without learning from the old ones. How can data be used to change funding before it is spent? How can complex bureaucratic processes become more responsive if an outsider discovers a problem? And how much ‘real time’ influence is really desirable? Maybe people or experts should vote on how actual money is spent in development, but that sounds like a complex process that needs more experimentation.
To sum up, I am still all for publishing as much data as possible, but I also would like to see more critical voices that question some of the assumptions – even if it is just an exercise for open aid enthusiasts to make a better, stronger, more effective case for this important and valuable project.
"Better information will make aid more effective" is a hypotheses that needs testing, not a belief that should simply be adopted.
ReplyDeletePlease consider, amongst other things, the experience of the Guardian funded Katine project in Uganda, implemented by AMREF. It was subject to intensive journalistic coverage over three years, all of which went up on the web, and which is still available (540+stories, plus progres reports and evaluations). My impression (as a marginal participant in this project) was that the degree of public exposure actually helped push the project strategy away from long term development activities (like training) and towards more immediately useful investments, such as buildings and supplies of different kinds. This project's experience needs to be studied. As well as the Guardian website, you can find more on my Evaluating Katine blog
A good call for more critical debate around this Tobias.
ReplyDeleteIn short I think that:
1. Aid transparency is starting to mean many things to many people; this isn't necessarily bad but it does stress the need to be explicit when we're talking about "open aid data".
2. At least the the "who's doing what, where?" piece of open aid data is very well defined as part of the International Aid Transparency Initiative (IATI).
3. Yes: development decisions can be political; access to aid data is one thing which may change the politics.
4. Open data is an enabling step, not an end in itself. Transparency more generally is about making the systematic changes that allow ongoing, incremental improvements to aid as a whole.
I wrote a more complete response at:
http://www.aidinfo.org/open-data-in-development-the-missing-debate.html
I can sympathise with the IDS blog on donors increasingly collecting quantative data for their use and for telling their story. However this is not my understanding of aid transparency. Aid transparency as it is discussed e.g. in IATI is about donors publishing data (quantitative or qualitative) primarily for other stakeholders to increase accountability. So it is precisely because decision making in development is political, that open data can be valuabe to permit those hitherto excluded from political bargaining to voice their concerns and interests.
ReplyDeleteOf course open data is no silver bullet. Other aspects like capacity and resources of civil society stakeholders and conducive environment for political debate need to be considered.
It would be worthwhile to take a broader view and follow the open data debate in the European context, e.g. in the Pan European e-participation network, and learn from it in development cooperation.
And one important caveat for all those following the aid transparency debate in the UK: Aid transparency is very high on the agenda in the UK and a bit less on the agenda in Sweden, the Netherlands and the USA. The rest of the world doesn't really care!