Digital Inequality and the Signature of Digital Technologies and the Digital Ecosystem: Analysis of Deviations in the Rank-Size Rule of Internet Access Data

Abstract

The world is today a more digitally integrated place; however, digital inequality still prevails and its repercussions (e.g., poor access to information, e-commerce, remote education, remote work, and remote healthcare) have aggravated with the Covid-19 pandemic. In this debate, Lafuente, Acs, and Szerb adopt a power-law approach to scrutinize digital inequalities in 107 countries during 2000–2019. The authors take the digital inequality discussion to a more qualitative level by connecting their findings to the quality of countries’ digital ecosystem. Building on the nuance that digital integration encompasses digital technologies and a healthy digital ecosystem, the empirical exercise shows significant progress in digital integration during 2000–2019; however, digital integration is slowing down since 2015. The inspection of countries’ digital ecosystem suggests that digital policies targeting governance (e.g., regulation and data privacy) and platforms’ activities (e.g., social media and online payments) are critical to enhance the digital system and, consequently, reduce digital inequality and its negative manifestations.

Appendix

1.1 Appendix 1: Power law exponent for the number of Internet users (selected years: 2000, 2005, 2010, and 2015)

This material presents the results of the power law estimations for the distribution of the number of internet users for the years 2000, 2005, 2010, and 2015. Figure 8.3 shows the log-log plots for the analyzed data, while full OLS regression results are presented in Table 8.2. Finally, the data used in this analysis, available at the World Bank databases (URL: https://data.worldbank.org/indicator), is presented in Table 8.3.

Fig. 8.3

Log-log plots of the number of internet users for the years 2000, 2005, 2010, and 2015. We followed Gabaix and Ibragimov (2011) to compute the standard error (SE) of the power law exponent $\left( SE=\sqrt{2/N}\times \beta \right)$. For all power exponents the p-value < 0.001 (two-tailed)

Table 8.2 OLS models: results for the power law analysis of the number of Internet users

Table 8.3 Data for the number of internet users (millions of people) for selected years (2000, 2005, 2010, and 2015)

1.2 Appendix 2: Data for the empirical analysis

This material describes the variables employed in the empirical exercise of the study (Table 8.4). We also include the data used in our analyses (Tables 8.5 and 8.6) and the descriptive statistics for the selected variables (Table 8.7).

Table 8.4 Description of the study variables

Table 8.5 Data

Table 8.6 Digital platform economy (DPE): Data for the DPE and its sub-indicators

Table 8.7 Descriptive statistics for the study variables (N = 107 countries)

1.3 Appendix 3: Kernel density plots for the digital platform economy (DPE) index and its sub-indicators

Table 8.8 in this material presents the results of the statistic tests (t-test, Mann-Whitney U test, and Kolmogorov-Smirnov test of equality of distributions) evaluating differences in the DPE indicator and its four sub-indicators, namely: digital multisided platforms, digital technology entrepreneurs, digital governance, and digital citizenship. In the table, the comparison of the values for the digital ecosystem proxy variables distinguish between outperforming countries that improved their rank between 2015 and 2019 (DiffRank > 0) and underperforming countries whose rank value worsened between 2015 and 2019 (DiffRank < 0).

Table 8.8 Mean comparisons

1.4 Appendix 4: OLS regression results

This material presents the results of the OLS models relating the digital platform economy (DPE) index to the digital integration variable (normalized difference in ranks: normDiffRank) and the variation (%) in the number of Internet users.

Table 8.9 Regression results (OLS)