Are information services ‘disintermediated’ and ‘disembedded’ from place and social relations?

*This post was originally posted May 30th, 2013*

In this blogpost, I want to reflect on our recent workshops with BPO/ITES managers in Kenya and Rwanda and answer this question (tentatively): No, the global market for information and knowledge based services is still very much embedded in place and social relations.

Please examine Exhibit A:bpo value chainsbpo value chains 2

This diagram comes from the Canadian government’s guide for small and medium sized enterprises interested in ‘linking in to global value chains” (

Such diagrams suggest an easy dispersal of business activities around the globe; R&D wanders into California while Sales drifts to Europe and Assembly drifts to India. Our project is concerned with Business Process Outsourcing (basically white-collar service employment) in East Africa: how does the internet enable the flow of such services across geographical, national and social boundaries? Do Indian, Kenyan and Rwandan companies compete on a global scale or are their value chains embedded in their geographic and social positions?

A couple weeks ago, I was in Nairobi and Kigali to share our preliminary ideas with our research participants and to get their own ideas about such research questions. In our discussions about barriers to getting work, two clear consensuses emerged:

  1. Accessing global (US and European) services requires some kind of ‘presence’ in client markets.
  1. Accessing local or regional (East African) markets requires either local social/domain knowledge or internationally standardized knowledge.

First of all, among our BPO participants, there was a strong feeling of disconnection, a dependence on international intermediaries and an inability to provide secure employment in the current climate. This was starkly demonstrated in the following participant drawings:

bpo 1

The participant above has clearly distinguished between ‘international associate’ (presumably a consultant or intermediary) and an ‘international client’. Without direct contact with clients, the manager feels unable to provide full-time stable employment to his workers. This sentiment was reflected in other participant value chains. For example in the drawing below, we can again see his desire for ‘direct contact’ and dissatisfaction with the subcontracting relationships that currently predominate.

bpo 2

A number of participants expressed a desire to have international employees or offices to bring in more work. From the perspective of the BPO managers interviewed in our project, it would appear that place matters a great deal and that they currently feel far away from their clients.

This feeling was also expressed in our workshop discussions about policy. One participant explicitly stressed the importance of ‘presence’ with others stressing ‘access to the end client’ as a key variable in determining the success of a business model. These participants wanted the government to facilitate presence in foreign markets. The two following tables again emphasize the importance of having ‘presence’:

bpo 3bpo 4

This was also reflected in the ‘Samasource’ chain and it is interesting that a number of participants in both the workshop and from our broader sample of interviewees rely on Samasource to bring in a constant and steady stream of work. While the margins are lower when using an intermediary like Samasource, this reliability was extremely important to managers. Here is Samasource’s value chain:

bpo 5

Another thing that might be stressed is the unreliability of platforms like elance in the eyes of Kenyan managers. This was starkly shown in the following drawings of two participants:

bpo 6 bpo 7

Their dissatisfaction with elance/ and their desire to connect with international consultants comes across very strongly. In the second drawing, the participant writes that ‘international mediation’ and ‘a lack of a national portal’ are two of the biggest barriers from having the kind of chain he/she wants.

From the perspective of these two managers, it would appear that the transaction costs of conducting small-scale piecemeal projects available on platforms like elance and make this kind of distintermediated work unfeasible for serious companies. Instead, there was a general consensus that managers want more ‘introductions’ through international workshops, foreign offices, social networks, international consultants, donors and a ‘national’ portal so that they can come into contact with bigger, more stable clients.

When I again asked Rwandan participants (primarily in the software development field) to reflect on the key barriers stopping them from getting higher value, more international work, they gave the following suggestions:

  • “Lack of transparency in offering jobs”
  • “Lack of references”
  • “Lack of information”
  • “Lack of enough information and finances to get in the track”
  • “Lack of access to capital”
  • “People don’t understand/awareness (of local clients)”
  • “Fake Materials (in terms of software and hardware)”
  • “Lack of standardized skills”
  • Skills gap

Despite the proliferation of fibre optic internet and mobile internet connections, information does not appear to have completely ‘moved online’. By far the most common barrier to accessing work opportunities was a lack of transparency and a lack of information about work opportunities and clients. This reflects the social embeddedness of information about work opportunities in the local market; in order to find out about an opportunity, one must be connected to the right kinds of social networks. These may include overlapping government/political networks, local private sector networks and transnational networks. Within the ICT sector, we may presume that all players are computer literate and speak English or French, so it is not a lack of access to internet connectivity or a lack of awareness about how to use the internet; but rather that the information itself is not available online.

Interestingly, this contrasted with our discussion with players from the Rwandan tea sector. When I asked Rwandan tea farmers what kind of information they needed to improve their positions in value chains, they wanted auction price information, production cost information and information about the prevailing conversion ratios (the ratio of green tea leaf needed to produce processed black tea). Most of this information is already available online either from the East Africa Tea Trader’s Association website, Rwanda’s National Agricultural Export Development Board (NAEB) or through the reports of factory boards (different factories may publish such material differently but tea farming unions are usually minor shareholders). The tea farmers were surprised to learn that this information was already available, emphasizing that such information was incredibly important to them in bargaining with factories for better prices. It might be the case that more powerful players currently feel comfortable publishing information online because they do not expect weaker players to be able to access it. In other words, the internet might allow for the free flow of information between players in the chain who have incentives to share information (stronger players with shared interests), while the ‘digital divide’ that currently prevails prevents weaker players from tapping into that same flow of information in order to collectively bargain and change the underlying power relations within the sector. It will be fascinating to examine how changing information structure may or may not change these bargaining relations over time. Will stronger players try to ‘hide’ or restrict access to information once weaker players tap into online information? Will they move information out of public forums into private emails and paid-subscription/member-only websites? Or will online information allow farmers to negotiate better prices and shareholding arrangements with factories? It may be too early to answer these questions so further research should be conducted in future. Our project can provide that incredibly important ‘baseline’ to look at change over time.

Overall, we might say that there are two kinds of ‘information divides’: cases where information is not online because players do not publish information (work opportunities within the ICT sector) and cases where information is available online but there is a lack of awareness, a lack of computer literacy or a lack of English literacy (tea information). These two problems have different policy implications.

In the case of the tea sector, government bodies can extend infrastructure, carry out English and technical training, conduct awareness campaigns and make internet and hardware/software more affordable. But even once they have done all this, the second kind of problem can kick in- the fact that some players may not want to share information online. It is much harder to proscribe policy recommendations here.

How do you encourage players who have a reluctance or even a disincentive to share information more widely? They may benefit from information asymmetries. They may not trust other players to carry out contracts successfully. They may have wider commercial, social or political networks of dependence/redistribution that might prevent them from engaging with new players. This issue really strikes at the heart of many e-governance and other ICT interventions- it is not just about technology, but about trying to change the fundamental governance relations on the ground.

When it comes to work opportunities within the ICT sector of East Africa, there is this balance between:

  • choosing foreign players whose systems, skills and experience allow one to easy ‘check boxes’ on tendering documents,
  • choosing local players who skills and experiences one knows intimately or finally,
  • choosing unknown local players who do not have neither the right documentation or social knowledge to prove themselves worthy and rewarding of opportunity.

Numerous interviewees stressed that the software sector is notoriously hard to standardize. A good developer is often judged by his last project, so in order ‘to know’ if someone is good, one must have social knowledge about the relevant players. This social/domain nature of knowledge and expertise means that the software sector is incredibly difficult to ‘certify’ and standardize. Foreign players get around this lack of ‘social or domain knowledge’ by being able to draw on huge amounts of systemized and regularized experience and documentation from other domains, but it is much harder for new entrants in the local domain who do not have this wider network of knowledge.

Places like kLab and iHub are good spaces for ‘becoming known’ by key information brokers (those who may have privileged access to information about opportunities) but there is a danger that such spaces can become biased towards certain groups (in-groups over out-groups) over time. Again this is an area that needs more research- how these inherently social spaces became ‘gateways’ for fresh talent to enter into circles of expertise and opportunity. These spaces always begin as ‘open places’ but over time, they become popular and the institutions themselves begin to realise how they too must filter and act as gatekeepers. This process is already happening in the iHub in Nairobi; initially the space was completely open. Now they are developing a revised membership policy and a board to judge who can and cannot have access. Spaces like kLab also need to think about how they will manage this transition.

In other words, this problem of information access really conveys an important message- information is about power and for that reason, one cannot assume that having the right information infrastructure is necessarily going to lead to more openness and equality. Information brokers and ‘infomediaries’ have strong incentives to keep information to themselves and only share it with others when they feel they share interests or can really trust those individuals.

Attention needs to be paid to how and why information gets stuck and when it does not filter down to other groups. Is it a matter of trust and social knowledge/standardization or is it a matter of power relations? Both these problems can be approached by policy-makers, but their policy interventions will ultimately be about brokering power, and not just about making hardware, connectivity or ‘open spaces’ more available. How do you really open up a space to new participants when existing players (both local and international) have incentives to close that space either to protect their own value and profits or to protect the wider environment or reputation of the place as a whole?

This research ( is being carried out by Laura Mann and Mark Graham from the University of Oxford, Timothy Waema and Charles Katua from the University of Nairobi and Felix Akorli, Grace Magamno and Claude Bizimana from the National University of Rwanda. If you would like a fuller summary of these meetings or more information about the project, please do get in touch!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s