Impact of Over-the-top Services, Pricing, and Equity on the Digital Divides in Sub-Saharan Africa

With the convergence within the digital ecosystem today, access to digital technology is now a multilevel phenomenon closely tied to the access to one or more of the following: a compatible device, the Internet, and a facilitating service/application. This makes it difficult to disentangle
mobile services and Internet services in studies on the digital divide. As countries in Sub-Saharan Africa (SSA), which have the lowest adoption rates globally, look to leverage digital technologies as a tool to drive economic and social development, there is a need for continued and novel approaches to understanding the digital divide. This thesis proposes a new approach to conceptualizing the digital divide and characterizes the three levels of the digital divide:
inequalities in access, use, and benefits from use in SSA, using Nigeria as a case study. This work also critically examines the effect of recent pricing policies in Nigeria on the digital divide, as well as the effect of other sociodemographic, socioeconomic, and behavioral factors at the individual level. In Chapter 2, I run a choice-based conjoint experiment to understand the impact of access to
over-the-top (OTT) services on individual preferences for different mobile services – cellular calls, text, or the different services on the Internet – or not using any mobile service (the first level digital divide). I find that when OTTs are introduced into the market, mobile users are less
likely to go without mobile services or to use a traditional service. I also find that this effect is significant in a market with a pay-as-you-go business model. The results also reveal that mobile users are price sensitive, therefore pricing policies may aid in bridging the first and second levels
of the digital divide. The findings indicate that customers’ preferences in the mobile market are changing and OTT access could be a tool in closing the first-level digital divide. Therefore, I recommend that policies to drive Internet access, especially OTT, should be explored. In Chapter 3, I use a panel data approach to estimate the effect of reduction in the prices of mobile Internet plans on the volume of use of the Internet, cellular calls, and text messaging services (the second-level digital divide). I find that the reduction in the prices is associated with an increase in the volume of data used and a decrease in the volume of texts sent by an individual. However, reducing the prices of mobile Internet plans does not “close” the second level digital divide across socioeconomic groups. I did not see a convergence in the volume of use of any of the mobile services across any demographic subgroups. These findings suggest that more robust policies that are targeted at specific subgroups are needed to reduce the existing
second-level mobile technology digital divide that exists in developing countries. In Chapter 4, I draw on the Uses and Gratifications Theory, the Unified Theory of Acceptance
and Use of Technology, and factor analysis to examine the differences in the frequency and type of Internet use (second-level digital divide) and the differences in outcomes from Internet use (third-level digital divide). I find that females, the older population, and individuals with a lower
level of education are the digitally disadvantaged subset of the population. I also find that high technical skills are associated with high frequency of use of the Internet for personal development, social, and business activities. I also find that encouragement from family and friends as well as intention to increase Internet use in the future are associated with increased frequency of use of the Internet for consuming news content and social interaction respectively. This supports arguments in an earlier work that improving access to over-the-top applications such as WhatsApp could increase Internet use. In examining the determinants of the third-level digital divide, I find that using the Internet for social activities such as using social networks and communicating with family and friends have the greatest impact on offline outcomes. I also find
that individuals with a high level of education are more likely to get positive health outcomes and less likely to get personal development outcomes, such as getting a job or completing an online training, from the Internet.
In Chapter 5, I discuss this work’s contribution to the literature and some of the policy implications of the findings in Chapters 2 through 4. Findings in Chapters 2 through 4 suggest that supporting the use of OTT and social networks in developing countries would have benefits. These Internet activities, although they may not directly contribute to personal development or economic gains, typically require little technical skills. Therefore, by engaging with these
Internet activities, the digitally disadvantaged subset of the population would be able to develop the required skills to achieve benefits from Internet use. In Chapter 3, I learn that addressing the affordability barrier in a developing country is not enough to bridge the second-level digital divide. More robust policies are needed to bridge the second-level digital divide in developing countries. In Chapter 4, I learn that the digital disadvantage may simply be a reflection of
societal inequalities in the online space. Therefore, in order to bridge the digital divide, target policies that address these preexisting inequalities are recommended.