Duncan Green: What are the limits of transparency and technology?

Cross-posted from Oxfam’s From Poverty to Power blog, by Duncan Green.

What are the limits of transparency and technology? From three gurus of the openness movement (Eigen, Rajani, McGee)

After a slightly disappointing ‘wonkwar’ on migration, let’s try a less adversarial format for another big development issue: Transparency and Accountability. I have an instinctive suspicion of anything that sounds like a magic bullet, a cost-free solution, or motherhood and apple pie in general. So the current surge in interest on open data and transparency has me grumbling and sniffing the air. Are politicians just grabbing it as a cheap announcement in austere times? Does it contain some kind of implicit right wing assumptions (an individualist homo economicus maximising market efficiency through open data)? And is there any evidence that transparency actually has much impact on the lives of poor people (after all, the proponents of transparency and results-based agendas are often the same organizations, so I hope they are practicing what they preach….)

I put these fears to three transparency gurus, and here are their fascinating responses, striking in their quality and level of, well, openness. It’s a long read, but I hope you’ll agree, a worthwhile one. Think we’ll just stick with comments on this one – doesn’t feel like a vote would be useful (but let me know if you think otherwise)

What else is needed to make transparency have impact?

Prof. Dr.  Peter Eigen, Transparency international (Founder, Chairman of the Advisory Council)Peter Eigen

Transparency is not a magic bullet but rather part of the arsenal. Transparency, when combined with accountable and responsive institutions as well as the space for civil society to get involved, is a game changer. These other factors are what make transparency meaningful and useful to build trust, fight corruption and achieve development progress. Alone, transparency may be cost-free but it also falls short on realizing all the promises that have been tied up with it.

But what does transparency really encompass or mean? Transparency can generally be defined as the “characteristic of governments, companies, organisations and individuals of being open in the clear disclosure of information, rules, plans, processes and actions”.

Take the case of public spending. Transparency can be used to follow the money to make sure that promised funds are turned into results for the poorest. In Mexico, “social witnesses” are required by law and have overseen more than $50 billion of public spending and tendering procedures. Findings from Transparency International show that in countries where there is greater openness, more checks-and-balances and better rule of law, more pregnant women have healthy births, more children and young people are educated to read, and more families have access to clean water and sanitation.

But in all of these cases, the ‘magic bullet’ of transparency is more part of a broader magic formula. Open government information is combined with governments that are willing to be held to account, official channels that allow for a government’s accountability and the rule of law, and individuals that are able and wanting to hold their government to account.

When it comes to companies, improving transparency can be used to see what businesses are paying into government coffers in terms of taxes, royalties, licenses and other payments. But again, simply putting the information out in the public domain does not equate to changes in practice or market efficiencies unless there are shareholders, governments and individuals that have the means to convert transparency into changes in policies and actions.

Transparency International (TI), the organization that I founded over 20 years ago, has seen how much increased openness can lead to real and lasting change when these other pieces of the puzzle are present. TI can be found in more than 100 countries where it has national chapters.

What TI has learned is that it is not just the quantity of information provided but the quality: is it comparable, timely, accessible, understandable and useful? It is also about matching the supply of information with the demand for certain types of information.

Information may be “cost-free”, but there may be no one interested in it unless it meets these characteristics and speaks to people’s needs.

Sorting this out for development is what will make possible a “data revolution” as has been called for by the UN and others, rather than creating a deluge of information.

We walk the fine line between just opening the information faucets in the name of “transparency” and providing what is needed to allow individuals to participate, provide oversight, prevent corruption and make informed choices.

Why transparency and technology won’t drive accountability

Rakesh Rajani, Head of Twaweza East Africa and the co-chair of the Open Government Partnershiprakesh rajani

Ideas that hold the most promise can also be the most deceptive; for their power and allure can mask the inconvenient hard thinking that come in the way of a good story. The use of technology in development, and in particular its potential to close the gap between citizen voice and state responsiveness, is one such idea.

In the past decade, many a development blogpost, newspaper article and YouTube video has gushed about the transformative power of the internet and mobile telephony – in closing information asymmetries, creating pathways for citizen expression and feedback, monitoring service delivery, visualizing data, and creating new possibilities for collective action. At Twaweza, we have claimed that the spread of communication technologies is a ‘game changer’ in East Africa, that the information space has been ‘democratized’ now that the content and methods that used to be the preserve of a few are open to the many, and because facts and ideas can travel in so many directions, so quickly and at little cost. Who cannot be moved by the memes of citizens mapping their neighborhood, or village women reporting the broken water point, or budget visuals that give you simple, color-coded bubbles to follow your money?

That technology can allow us to do interesting things is amply documented. I’ve just reviewed a useful book of several such case studies that the World Bank will launch on April 9. The trouble is that too many of us too much of the time have oversold its promise. Not so long ago grownups would talk as if all one had to do was to sprinkle mobile phones or internet and the persistent, structural imbalances and power asymmetries that had dogged us for decades would melt away. ‘Apps for Africa’ was the new Live Aid and microcredit revolution rolled into one. Thankfully, for the most part, we are past that stage.

With hindsight, it’s tempting to be snide. But it’s not so easy in practice. At Twaweza we’ve supported a number of thoughtful, caring people to roll out seemingly well-designed technology-driven initiatives (see, for example, here, here, here and here), ideas that many smart people from local NGOs and media to the World Bank and prestigious universities found compelling, but that did not live up to their original promise in uptake or fostering accountability. I suspect that we have no monopoly on a particularly high rate of failure, but are simply more willing and able to talk openly about it. Even those that ‘work’ often do not last beyond their initial phases (for example on the limits of Ushahidi see here, here, and here) or transform the underlying constraints they were set up to address. Snazzier forms do not a function make; when the authorities choose to ignore the content there isn’t much difference between a wooden suggestion box and a citizen feedback website with analytics.

So how do we think sensibly about these issues?

First, just because technologies can allow us to collect, store, analyze and communicate data and ideas in unprecedented ways should not lull us to think they can address old, entrenched problems in unprecedented ways. The primary constraints for human action are non-technological in nature. Most people who do not speak up in public meetings have perfectly functioning voices, and training them on better enunciation will not help matters much. Many technology projects have been hampered by inadequate theorizing, by political economy and social movement analysis, and by the lack of reference to historical evidence. And while clear and imaginative thinking is universally valuable, by necessity this analysis needs to be contextual. In particular we need to be particularly cautious about transferring successful use of technology from one place and time to another.

Second, we need a deep understanding of human motivation; of why people – on both the ‘citizen demand’ and ‘responsive state’ sides – would choose to pay attention, take action, and persist when setbacks happen, and a more granular (than just ‘citizens’ or ‘authorities’) of who would act. Our assumptions on each of these can be ill informed and poorly thought through, leaving us perplexed on why no one showed up for the thing we built so well. A simple set of questions, such as these (see Fig 3 on page 32) developed by a team evaluating citizen action in our Uwezo initiative, can be instructive.

Third, as Twaweza Advisory Board member Lant Pritchett puts it, however smart we are or well we do our homework, we are bound to not get it right the first (or second or third) time. That being so, what matters is how we test and tweak, and retest and retweak; set up agile feedback loops in a structured manner; and establish internal incentives and external relationships that can foster a culture of iteration and learning. The point here is not to experiment all day in boutique labs with little regard to impact, but rather to integrate experimentation and adaptation at the heart of how we implement at scale.

These lessons arise here in reflecting on the limits of technology to drive accountability, but in fact apply more broadly. And that is the point. Technology and transparency don’t drive anything. People, who organize and at times use technology to do so, do.

Motherhood, apple-pie and openness

Rosemary McGee, Institute of Development Studies (UK), Research and Evidence Component Coordinator of Making All Voices Count, and Technical Advisor toRosemary McGee Open Government Partnership

I’ll start by being a typical academic: whether transparency is necessary and important for development purposes depends on one’s definition of a number of terms.  First, development.  That might mean the extension of market economics to the furthest corners of the globe, the achievement of the MDGs, or the ability of people to imagine the world differently and realise that vision by changing the power relations that disadvantage them.

Second, transparency.  To TI, founded 20 years ago, transparency means clear disclosure of information, rules and actions, as in the definition Peter Eigen sets out above.  Since then, definition-creep has set in.  Lots of terms have come to be connected with it.  Is this a definitional issue that only bothers hair-splitters like me, or a real issue that ought to bother all of us?

There’s ‘openness’, which is anything from laissez-faire capitalism and free trade regimes (the economist’s definition), to inter-operability and the access to use and modify data and computer code in a shared environment (the techie’s definition), to a critical attitude to tradition (the open society advocate’s definition).  And several more – these are just the three most relevant to the present debate.

Then there’s ‘open government’, which according to the Open Government Partnership means greater civic participation in public affairs, and governments which are more transparent, responsive, accountable, and effective.

And then there is the usage of transparency which treats it as the harbinger of accountability.  This puts accountability as the goal, and transparency as one contributing factor, linked to it in an ‘uncertain relationship’, as Jonathan Fox warned several years ago.

‘Openness’ and ‘open government’ may have many effects that are good for people in situations of poverty and marginalization.  They may also have effects which are not, and effects which are bad for them, at least relatively speaking.  They may ease the corporate pillaging of natural resources or the monopolistic buying-up of subsistence farm land. They may demonize ‘tradition’ in ways that undermine informal institutions on which poor people or minorities depend for their livelihoods or their dignity.  Or perhaps they provide a smokescreen behind which governments hone their capacities to spy on suspect citizens.  The scope for the bad effects increases exponentially as technologies lower the costs and shorten the timeframes of ‘opening up’, reducing the pressure for careful appraisal.

Whether ‘openness’ and ‘open government’ are good for poor people depends on who is promoting them and using them, and for what.  (A recent learning study on assumptions and realities about the users of tech-based transparency and accountability initiatives suggests that, if the ‘who’ and ‘for what’ questions are asked too rarely about the promoters of tech-for-T&A, they are asked even more rarely about the people the promoters claim to be benefitting – which leads straight to the ‘whether’ question, but that’s another story.)

I believe that accountability in public and corporate governance is a universally good thing that is good for poor and marginalized people.  My belief doesn’t extend universally to all the approaches set out above.  Unlike ‘openness’ or ‘open data’, transparency for accountability means a value-based commitment to a particular kind of change, one rooted in principles of human rights and fairness.  The Open Government Partnership’s definition of its purpose is rooted in these too (although that may come as a surprise given some of the governments that have been let into the OGP).  For governance to get more accountable in any country in the world, African, Asian and Latin American countries, transparency would certainly be necessary, and openness, open data, open government might contribute.

There is not enough evidence yet that transparency does have a positive impact on poor people’s lives, because of the non-linear relationships between transparency and accountability, and between accountability and better lives.  The causal chains are long and complex, and include a lot of false starts and dead ends.  Programmes like Making All Voices Count, backed by funders who are both proponents of transparency and strongly results-based in orientation, are investing considerably in building more evidence on whether it’s so and, when so, how it happens.  But in this field marked by frenetic aid, philanthropic, entrepreneurial and surveillance activity, let’s not leapfrog over the questions of what is being pursued under the heading of ‘transparency’, who is promoting it, and why.

Duncan Green is a strategic adviser for Oxfam GB and author of ‘From Poverty to Power’.

Comments are closed.