Digital Labour

Development and crowdsourcing

A recent publication at the IFIP 9.4 conference in Indonesia “Understanding the Development Implications of Online Outsourcing” focused on analysing interviews with Upwork freelancers in the Pakistan Himalayas.  The framework we used was based on the DFID livelihoods and this was useful in making sense of the extent to which the Upwork platform was contributing to development.  The Youth Employment Programme (http://www.youthemp.com) of Pakistan’s Khyber Pakhtunkhwa (KPK) provincial government which aims to train 40,000 young people to use Upwork and similar platforms.  This is an integral part of Pakistan government policy in the region to bring marginalised people into gainful employment.  The results of our study were mixed, some of the freelancers were unable to engage at all even though they had been trained.  The frictionless marketplace was too intense and demanding for many of them. Others were seen to engage in a very limited way but a few were able to make a living and there was evidence of collectives developing into small businesses.  The success of the winners is encouraging but a lot are falling by the wayside.  There are many policy initiatives:

  • The NaijaCloud initiative in Nigeria, sponsored by the World Bank and supported by the national government, which provides awareness workshops on online outsourcing.
  • Malaysia’s eRezeki initiative (https://erezeki.my/en/home) which has trained hundreds of freelancers and now uses a “walled garden” approach with US crowdsourcing platform Massolutions: a managed portal for OO work such that the work process is controlled by government.

Protecting the individual worker from the intensity of competition inherent in multinational freelancing platforms is a facet of ERezeki .  The workers never actually engage with the platform, instead a government controlled portal mediates the work and protects the worker from bidding and managing clients.  This is not the only mechanism for reintermediating controls over the frictionless work platform marketplace.  A good example is cooperative platforms  where the workers also manage the platform thus offering a more equitable relationship for the workers.  Clearly there is more work to be done on how best to mediate plaform based “gig economy” for development exploring the limits and downsides of Upwork and other purely commercial logic driven platforms; in country alternatives such as Iran’s Ponisha to PPP arrangements and other business models including cooperatives.

Digital Labour

Platforms, crowdsourcing and the “poverty – risk” thesis

At the recent DIODE meeting in Indonesia we participants split off into groups to focus on particular topics one of which was the development implications of crowdsourcing platforms.  There is a significant critique of crowdsourcing platforms such as Amazon MTurk focussing on the same criticisms levelled at the “gig economy”: lack of any holidays, sickness, pension and health and safety.  Whether the quintessential “gig platform” Uber has employees or is just a technology company is a topic for Union activism and courtroom debate.  Amazon MTurk, Crowdflower, Upwork, Fiverr etc. have yet to come under the same level of scrutiny as Uber who remain in the spotlight of Union and other stakeholder activism.

This is perhaps because exemplary platform-based organisations such as Samasource take paid work to marginalised individuals where the work options are limited and who otherwise would either be unemployed or working in much worse conditions.  However, this does not mean the work is risk free and little is known about the nature of the work done in developing countries and the effects.

The late sociologist Ulrich Beck in his book “Risk Society”[1] argues that there is a “systematic attraction between extreme poverty and extreme risk” citing the examples of people in developing countries poisoned by engaging in work such as manual processing of asbestos brake linings, fertiliser application and afflicted by the lax safety standards that caused the Bhopal chemical disaster.

How could crowdsourcing carry such a relationship?  During the course of the discussion we focussed on some of problematic issues around the issue of content moderation.  It would appear there is some evidence of Beck’s “poverty-risk” hypothesis applying here.

The task of removal of extreme pornographic and other images considered offensive such as ISIS beheadings reported as offensive from platforms such as Facebook is undertaken by workers in the Philippines[2].  The mental health implications of this are as yet unknown but the interviews the Wired article below indicate that significant trauma is experienced by the workers who are responsible for identifying and removing many hundreds of such images per day.

There is some literature derived from other occupations such as the police who are dealing with paedophilia images as part of investigations. It is worth considering the conclusions of the study :

“The results revealed that Internet Child Exploitation (ICE) investigators experience salient emotional, cognitive, social and behavioural consequences due to viewing ICE material and their reactions can be short and long term. The degree of negative impact appears to vary markedly across individuals, types and content of material and viewing context, with variation based on individual, case-related and contextual factors both in and outside the workplace.”[3]

Viewing these images clearly has negative effect and measuring and monitoring the effect is under researched. The Police force recognise there is a problem with repeated exposure to such images and video content and have put in place policy for management of the effects such as counselling etc.

As yet there is no evidence that crowdsourcing platforms have considered the need for employee health and safety (partly because of the lack of distinction on responsibilities of platform based organisations for health and safety, ethics regarding freelance gig workers vs. employees).

There is clearly a research agenda for us to pursue regarding the ethics, training needs and mental health implications of platform based crowdsourcing.

[1] Risk Society Towards a New Modernity Ulrich Beck

[2] https://www.wired.com/2014/10/content-moderation/

[3] Powell, M., Cassematis, P., Benson, M. et al. J Police Officers’ Perceptions of their Reactions to Viewing Internet Child Exploitation Material  Police Crim Psych (2015) 30: 103. doi:10.1007/s11896-014-9148-z