How can we understand the impact that mobiles are having on the livelihoods of the poor?
We all know that mobile phone use has grown exponentially in developing countries. And that phones are having an increasing impact on the livelihoods of the poor by providing market prices, by supplying health information, by enabling financial transfers, etc.
But we know a lot less about how to conceptualise all this. Can we just pull some development studies ideas off-the-shelf? Or do we need to do more than this?
A new working paper in the Development Informatics series – “Understanding Mobile Phone Impact on Livelihoods in Developing Countries: A New Research Framework” – argues the livelihoods approach is a good starting point. But that it needs modification.
The livelihoods approach suggests four potential impacts of mobiles on the assets that underpin all livelihoods:
− Asset substitution: saving time and costs for journeys, but adding costs for mobile expenditure.
− Asset enhancement: greater efficiency in use of other assets e.g. for agricultural production or relationship-building.
− Asset disembodiment: the conversion of assets to digital form e.g. the codification of social contacts, or digitisation of money.
− Asset exchange/combination: e.g. the exchange of airtime or m-cash.
Important intermediaries – mobile operators, their agents, community-based organisations and NGOs, family and friends – help shape the extent and distribution of these impacts. These are also shaped by the three livelihood strategies to which the poor apply mobiles:
− Maintaining existing livelihoods and mitigating vulnerability: e.g. use of mobiles to maintain social networks that can assist in an emergency.
− Expanding and enhancing existing activities: e.g. using mobiles to obtain greater earnings from existing produce, to save more effectively, or to obtain greater remittances from existing social contacts.
− Diversifying into new activities: e.g. employment in the mobile sector, or use of mobiles to complete micro-work tasks.
These components of the livelihoods approach – assets, intermediating organisations and institutions, strategies – are therefore very useful in understanding the role of mobiles in development. But the approach also has four shortcomings.
i. Reconceiving assets. The assets pentagon was developed within the context of traditional agriculture, and it underplays recent understandings of the importance of networks, agency and capabilities in development. It would be better replaced by a three-way categorisation of assets:
− resource-based assets (RBA) that are tangible (physical, financial, natural capital);
− network-based assets (NBA) that derive from connections (social, political, cultural capital);
− cognitive-based assets (CBA) comprising human and psychological capital including competencies (knowledge, skills, attitudes).
ii. Incorporating information. Mobiles expose a truth that information is the lifeblood of development, and yet it is essentially ignored within the livelihoods framework. Information is essential to individuals’ awareness of, and ability to utilise, all assets; and the use of information requires other assets to turn it into decisions and livelihood strategies. Those processes need to be recognised within any understanding of livelihoods.
iii. Recognising bottom-up processes. The livelihoods framework tends to see intermediating processes and structures in macro-terms (government, laws, policies, culture). But diffusion and use of mobile has equally been shaped by more bottom-up processes including the functioning of specific market transactions, and user appropriations and adaptations within poor communities. The latter need to be recognised.
iv. Categorising impacts. If the core interest is impact of mobiles, the homogenising of that impact into a single “livelihood outcomes” box is not particularly helpful. Better to borrow from the ICT4D value chain and differentiate a broadening scale: from direct changes in behaviour, through process-level outcomes, to broader impacts on development goals.
Adapting the livelihoods framework on the basis of these four points, we arrive at the revised framework shown below, for use in conceiving and researching the impact of mobiles on livelihoods in developing countries:
The framework immediately helps to identify possible research questions:
− What is the effect of contextual factors – processes of globalisation, processes of technological innovation, population migration, etc – on the livelihoods impact of mobiles?
− How are markets and market processes shaping the impact of mobiles, including the tension between seeking to make markets more inclusive, and markets’ tendency towards exclusion and inequality?
− What exactly is the impact of mobiles on the substitution, enhancement/diminution, disembodiment, exchange and combination of livelihood assets at the household level?
− Are mobiles forging new forms of connection to the intermediating structures and processes that govern the enactment of livelihood strategies?
− What new livelihood strategies are mobiles enabling; how do they come into being and come to sustain; and what impact are they having?
− What factors mediate the conversion of mobile behavioural outputs into broader outcomes and development impacts?
No doubt there are many other questions that the framework can be used to identify and conceptualise.Follow @CDIManchester
Actor-network theory (ANT) has been around since the 1980s, and significantly utilised in some disciplines, such as information systems. But – oddly – it has hardly been applied at all in development studies, including within ICT4D research. That is recently starting to change but to give some further impetus, we organised an international workshop in June 2011: “Understanding Development Through Actor-Network Theory”. You can find online both a summary of the workshop and abstracts and presentations from the nine papers (the papers should appear in a journal special issue in 2012).
Actor-network theory began as a means to explain how science works, such as the operation of scientific laboratories and projects. However, it has subsequently grown to be seen as a full-blown social theory. In particular, ANT says three things.
First, it says, “Hey, sociologists, you’ve been so obsessed with humans that you’ve been ignoring all the objects in the world. But those objects – documents, mobile phones, plants, websites, etc – play an important role; just like humans they shape the people and other objects around them. So ANT is going to treat them the same as people, and call them both ‘actors’.”
Second, it says, “Hey, sociologists, because you’ve been so obsessed with humans, you think that society and social contexts or social factors are what explains everything in life. But you’re wrong. In fact you’re so wrong you’ve got your basic equation of life the wrong way around. You think that society explains what goes on in the world. Nope. What goes on in the world is what explains society. So ANT is going to focus on the mechanics of life: the ways in which people and objects interact with each other.”
Third, it says, “Hey, more recent French-type sociologists, you’ve been so obsessed with breaking things apart to understand the bits of grammar and bits of history that made them that your idea of researching a clock would be to smash it to pieces with a hammer. That is not how to research a clock. To research a clock you need to understand how all the pieces got put together, following the network of people and objects that interacted in order to make that clock. So ANT is going to focus on how networks are assembled.”
Much ANT writing is horribly obscure, so full of hideously complex sentences and words that the writers must surely have done this deliberately in the hope of avoiding Oscar Wilde’s dictum, “to be intelligible is to be found out”. But, done well, ANT can tell a good story and even occasionally give you the sense that you are suddenly seeing the world in a whole new light. A whole new light that – because it’s about dynamics and innovations and technology and networks – seems especially relevant to ICT4D.
A couple of good entry points – good because they each provide a fairly clear and portable conceptual framework that you can re-use in your own research – are:
- Callon, M. (1986) Some elements of a sociology of translation: domestication of scallops and the fishermen of St Brieuc Bay, in: Power, Action and Belief, J. Law (ed.), Routledge & Kegan Paul, London, 196-233
- Law, J. & Callon, M. (1992) The life and death of an aircraft: a network analysis of technical change, in: W.E. Bijker & J. Law (eds), Shaping Technology/Building Society, MIT Press, Cambridge, MA, 21-52
Also not too unreadable is Latour’s Reassembling the Social, though had Latour been shot half-way through the dialogue with a PhD student that is reported in the book, I can’t help feeling a verdict of justifiable homicide would have been returned.
Although, as noted, use of ANT in ICT4D research has been limited there have been enough examples, at least from developing country cases within the information systems field, that we get a sense of the questions ANT is good at answering:
- How do you explain the trajectory of an ICT4D project?
- What role does technology play in an ICT4D project?
- How does power manifest itself in an ICT4D project? How were apparently powerless actors able to influence the direction of an ICT4D project? How was it that apparently powerful actors didn’t get their way on an ICT4D project?
- How does a particular ICT4D innovation (be it a new technology or business model or idea) diffuse or scale-up or sink without trace?
- How did a particular ICT4D impact or ICT4D policy come about?
If you’ve identified other ICT4D questions that are especially suitable for an ANT lens, then do contribute them.
If you want an example of applying ANT in ICT4D that also includes a reflection on the pros and cons of the theory, and some thoughts on applying it in your research, I can recommend:
- Stanforth, C. (2007) Using actor-network theory to analyze e-government implementation in developing countries, Information Technology and International Development, 3(3), 35-60
There is also a discussion of the relation between ICT4D and ANT in:
- Rubinoff, D.D. (2008) Towards an ICT4D geometry of empowerment: using actor-network theory to understand and improve ICT4D, in: Developing Successful ICT Strategies, M.H. Rahman (ed.), Information Science Reference, Hershey, PA, 133-154
And feel free to comment on other ICT4D literature that makes use of ANT.
If you would like to participate in discussions about ANT, you can join our online forum on LinkedIn at: http://www.linkedin.com/groups/ActorNetwork-Theory-in-Development-Studies-3995328
We will also be populating a group on Mendeley with reference details, and welcome contributions: http://www.mendeley.com/groups/1255941/actor-network-theory-in-development-studies/Follow @CDIManchester
Sometimes with ICT4D projects, you build it and they don’t come. Why is that? Why do potential users resist, object, reject?
One explanation comes from the concept of “institutional dualism”.
First developed to explain how Japan reacted to the import of Western ideas and technologies in the 19th century, this can also be used to understand any innovation – including ICT4D initiatives – in which there is some separation between designers and intended adopters.
Each of those two groups sits within its own institutional network: a complex of institutional elements (e.g. norms, rules, beliefs, values) and organisations and actions. Left to its own devices, any institutional network will tend to “self-reproduce”. For example, its cultural values will encourage particular actions, and those actions will in turn reinforce the network’s cultural values.
But innovations like an ICT4D application will bring two different institutional networks – those of the application designers, and those of the application adopters – into contact. We can call this institutional dualism because of the two institutional networks that come into play (see figure).
A strong example of institutional dualism would occur if a team from a European university designed an ICT4D application and then introduced it into a rural location in Africa. The European designers’ behaviour is enabled by a set of Western organisations and shaped by a set of Western institutional forces, some of which will be inscribed into the ICT4D application. During implementation this network is drawn into contact with the very different network of rural Africa, with different organisational structures, behaviours, and institutional forces.
Many ICT4D projects will be a bit less starkly drawn than this, but will still involve institutional dualism because designers and adopters almost always come from different places and different spaces.
What then happens? There are four possible outcomes from a situation of institutional dualism:
- Domination: one of the institutional networks prevails over the other in the ICT4D project. If the designers dominate, the project could fail due to its mismatch to the broader local context. If the adopters prevail, that requires a complete re-design of the project to have occurred.
- Contest: neither of the institutional networks prevails, but there is ongoing competition between them. The ICT4D project may stagger on, but always in difficulty as it is pulled in two different directions.
- Parallel-Running: a separation is arranged with some aspects of the project guided by the designers’ institutional network, some by the adopters’. This is only possible where the ICT4D project has a broad flexibility and scope.
- Hybridisation: the ICT4D project becomes the site within which the two institutional networks blend, forming a mixture of institutional values and hence a set of hybrid actions within the organisational structure of the project itself.
Of these four, only hybridisation and some types of domination are likely to lead to a sustainable ICT4D project. We have seen this in practice with a large-scale ICT4D case study of institutional dualism from the Brazilian public sector. Although giving some outward signs of hybridisation, beneath the surface this remained a story of ongoing contest and parallel-running even some years after its first implementation. It was still contingent, and it demonstrates the great difficulty ICT4D projects have in institutionalising themselves when operating in environments of strong institutional dualism.
ICT4D projects and policies can best be understood through a value chain model. As shown in Figure 1 below, this builds on a standard input—process—output model to create a sequence of linked ICT-for-development resources and processes. The model can be used for projects and policies in various ways: to trace their history; to analyse their content; to assess and evaluate.
The ICT4D value chain offers four main domains that can be the focus for historical or content analysis or evaluation:
- Readiness: the systemic prerequisites for any ICT4D initiative; both the foundational precursors that we might conceptualise mainly at the national level such as ICT infrastructure, skills and policy; and the more specific inputs (both ‘hard’ and ‘soft’) that feed into any individual initiative. Assessment could focus on the presence/absence of these resources and capabilities, or the strategy that converts precursors into inputs.
- Availability: implementation of an ICT4D initiative turns the inputs into a set of tangible ICT deliverables; typical among which might be a telecentre or mobile phones. Again, assessment can focus on either the delivered resources and/or the delivery process.
- Uptake: the processes by which access to the technology is turned into actual usage; also noting that key concerns around this process and its ability to contribute to development have related to the sustainability of this use over time, and – for various innovations that are prototyped – the potential or actuality of scaling-up. In practice, usage indicators are more often assessed than the various uptake processes.
- Impact: which can be divided into three sub-elements:
- Outputs: the micro-level behavioural changes associated with technology use.
- Outcomes: the wider costs and benefits associated with ICT.
- Development Impacts: the contribution of the ICT to broader development goals.
Figure 1: The ICT4D Value Chain
How has interest in these four domains changed over time?
One way to trace this is through key staging posts for the ICT4D community:
- The Digital Opportunity Taskforce (DOTForce) arose from the 2000 G8 summit in Okinawa. In 2001, it produced its “Digital Opportunities for All” report which encompassed four focal areas. Three – readiness, connectivity and human capacity – were related only to the Readiness domain; and one – participation in e-networks – looked mainly at Readiness and Availability issues.
- In 2003, the first World Summit on the Information Society was held in Geneva. Its main report was, tellingly, entitled “Building the Information Society” and not surprisingly the main focus was on building ICT connection and access; again looking mostly at the Readiness and Availability domains.
- The second World Summit on the Information Society was held in Tunis in 2005. Unlike its predecessor, its agenda did start to talk about impact. It still had a strong focus on precursors like financing and governance, but it included additional discussion about the application of ICTs, thus starting to encompass the Uptake and Impact domains.
- The largest subsequent meeting was the GK3 event in Kuala Lumpur at the end of 2007. It was shaped by twelve main sub-themes. Analysing these shows a fairly even spread across the four domains, though with Impact by now the largest single focus, followed by Availability.
There has been no subsequent comparable single event in the area drawing together many thousands of participants as these staging posts did; rather, a growing number of smaller events drawing several hundreds. However, a useful bellwether is the Information and Communications for Development Report produced by the World Bank. In its 2009 edition, the ratio of mentions of ‘readiness’ to ‘impact’ was 1:35.
Such evidence is best seen as straws in the wind rather than definitive, but it does suggest a similar pattern to that seen in other areas of ICT application, and summarised in Figure 2.
Figure 2: Changing Focus of ICT4D Priorities Over Time
Whatever the exact shape of the graph, it reflects the relative lack of attention that has been paid to ICTs’ contribution to development until quite recently. That is problematic because, as you move from left to right along the value chain, assessment becomes more difficult, more costly but also more valuable. Of course there has been literature assessing the connection to development including the summary Compendium on Impact Assessment of ICT4D Projects, and the 2010 Journal of International Development policy arena: “Do Information and Communication Technologies (ICTs) Contribute to Development?“.
However, donor agencies, governments, academic departments and others must still do more to shift the focus of attention along the ICT4D value chain; and to demonstrate ICTs’ development impact.