ICT access: Testing for convergence across countries

ABSTRACT

Our objectives for this article are twofold: first, to examine the convergence of nations with regard to ICT access; and second, to see whether countries in the same continental region (South and East Asia, West and Middle East Asia, Africa, Eastern Europe, Western Europe, North America, South America, and Oceania) or having same initial ICT access have coalesced into groups. We constructed an ICT access index using three indicators (mobile phone, Internet, and broadband use per 100 population) for a sample of 198 nations for the 16-year period from 2000 to 2015. Results show that digital divide has declined in relative terms, but not in absolute terms. The countries, starting at lower ICT levels, are not adopting ICT faster than the leaders. Continental region-wise analysis also shows that the catch-up is maximum for already developed countries, and minimum for countries in African and Oceania continental regions.

KEYWORDS:

• Catch-up
• convergence
• developing countries
• digital divide
• ICT access

JEL Classification Code:

• L86
• L96

Notes

1. Several studies have documented the application of ICT on these spheres as well as their positive impact for countries like Brazil and Tunisia (e.g., World Economic Forum 2015, 2010).

2. Adjusting for quality improvements, computer equipment in 2008 cost only about 10 per cent of what the same equipment used to cost in 1998 (Perry, 2009). Similarly, fixed broadband prices as a share of gross national income (GNI) per capita have dropped by nearly 65% in last five years (ITU, 2014a).

3. There is sufficient empirical evidence for Korea (Kim 2003), Taiwan and India (Kumar 2003), which shows that these countries in their initial phase of industrialization adopted a weak IP regime and relied on imitation products. This facilitated faster technology and knowledge diffusion.

4. “Convergence” (also known as the catch-up effect) is the hypothesis that poorer countries' PCI will tend to grow at faster rates than richer countries because of three reasons: first, they need not `‘reinvent the wheel’’ and can replicate the production methods and institutions already developed in advanced countries. Second, capital (and other factors) has a diminishing return (i.e., its marginal contribution to output declines when other inputs are kept constant) and these diminishing returns are stronger for developed or capital-rich countries. Third, the growth is higher if one is starting with a lower base (James 2012). As a result, all economies should eventually converge in terms of PCI. There are many examples of poorer countries that have converged with developed countries, thereby validating this catch-up theory. In the 1970s and 1980s, the East Asian Tigers, namely Singapore, Hong Kong, South Korea and Taiwan, rapidly converged with developed economies—all of them are today considered developed countries.

5. In the case of PCs, in 2001 the world average was 9% ownership, North America 61.1%, Europe and Central Asia 18.1%, Sub-Saharan Africa 1%, and South Asia 0.5% (Chinn and Fairlie 2007). In 2005, when Africa had only 3 computers per 100 persons, the corresponding figure for Europe was 26. In 2006, when Rwanda had only 0.3 computers per 100 persons, the corresponding figure for Canada was 89 (ITU 2009).

6. A probable reason for this is the release of Europe 2020 Strategy in 2010, which envisages a smart, sustainable and inclusive growth for European economy, which can be accomplished by (among other things) developing a digital economy based on knowledge and innovation (European Commission, 2010).

7. For our ICT access index, we wanted to include number of PCs also but could not do so because for most countries the data was not available beyond 2012. Moreover, in some countries where mainframe computers are prevalent, the number of PCs may understate the total use of computers (ITU, 2010a).

8. Ideal values are the “goalposts” that could be reached for different ICT variables. ITU (2015b) has estimated that 120 is the ideal value that a country could attain, irrespective of whether it has pre-paid or post-paid type of mobile subscription. For details about how these ideal values have been computed for different ICT variables, please see ITU (2015b). These “goalposts,” however are not fixed because the content and targets of the digital divide gap changes over time according to the different needs of users and the different technologies used (Kyriakidou et al. 2011).

9. All estimations were done in STATA 12.

10. The results though not reported, South America (9%) and Oceania (7.7%) continental regions have the lowest convergence rate.

11. ICT price basket as computed by ITU (2009) consists three sub-baskets: fixed telephone, mobile cellular, and fixed broadband Internet tariffs and is computed as the sum of these three tariffs, as a percentage of monthly per capita Gross National Income.

12. The countries chosen are Ethiopia from Africa, Mongolia from East Asia, Nepal from South Asia, Paraguay from South America, Tajikistan from Eastern Europe and Tunisia from Africa.

13. There is some evidence for this happening in rural areas too (see Aker and Mbiti 2010 for Sub-Saharan Africa and Jensen 2007 for India).

Additional information

Funding

National Research Foundation of Korea from Government of Korea (NRF-2014S1A2A2027622).