The digital divide
The digital divide is a barrier to human progress within the technological factors. The evidence for it, and the methods proposed to address it, follow.
ITU's Facts and Figures 2025 estimates 6 billion people use the internet, but 2.2 billion remain offline. The gaps are sharp.
| Group | Share online |
|---|---|
| High-income countries | 94% |
| Low-income countries | 23% |
| Urban areas | 85% |
| Rural areas | 58% |
| Men | 77% |
| Women | 71% |
Evidence
- Offline population. 96% of those offline live in low- and middle-income countries.
- Quality gaps. Quality gaps are growing even where people are connected. 84% of people in high-income countries have 5G coverage, compared with 4% in low-income countries.
- Beyond access. ITU's Secretary-General describes today's divides as being defined by speed, reliability, affordability, and skills, not just access.
- Learning cut off during school closures. UNICEF used a survey of education ministries and household data from 100 countries. UNICEF estimated that at least 463 million students, or at least 31% of schoolchildren worldwide, could not be reached by digital or broadcast remote learning when COVID-19 closed schools in 2020. UNICEF found this was mainly because their homes lacked the needed devices or the policies did not reach them. UNICEF found that more than 70% of the students who could not be reached lived in rural areas. Over three-quarters came from the poorest 40% of households. UNICEF notes these are best-case estimates and that the real number reached was likely significantly lower (UNICEF, 2020).
- Coverage and poverty in Nigeria. A study linked three waves of a nationally representative household panel to mobile operators' records of 3G and 4G rollout from 2010 to 2016. It found that mobile broadband coverage had large positive effects on household consumption. The study also found that coverage reduced the share of households below the poverty line. This came mainly through higher labor force participation and employment, particularly among women. In the authors' summary, total consumption rose by about 10% and extreme poverty fell by about 8 percentage points after coverage arrived. The earlier working paper version reported a smaller gain in consumption, about 6% after one year of coverage, growing over time. Coverage was not randomly assigned. Several authors worked for the GSMA, the mobile industry association (Bahia et al., 2024).
- Broadband and national growth. The study looked at a panel of OECD countries from 1996 to 2007. In that panel, a 10 percentage point increase in broadband penetration raised annual per capita GDP growth by 0.9 to 1.5 percentage points. To separate the effect of broadband from other causes of growth, the study predicted broadband spread from older telephone and cable TV networks. But it covers only rich countries in the early broadband era (Czernich et al., 2011).
Why it matters: The internet now plays the role that Riebau's bookshop played for Faraday. People who are offline are cut off from most of the world's accessible knowledge, courses, and networks.
What can be done
Actions rated on the strength-rating scale (A strong and replicated, B solid but limited, C weak or debated, D contested or failed), applied to the specific claim made.
- Mobile money services (B, contested). In Kenya, access to the M-PESA mobile money system raised per capita consumption. It lifted an estimated 194,000 households, or 2% of Kenyan households, out of poverty. Effects were larger for female-headed households, partly because women moved out of farming and into business (Suri and Jack, 2016). The estimate rests on differences in households' access to mobile money agents rather than randomization. Critics argue that the poverty claims are "extremely problematic" (Bateman, Duvendack, and Loubere, 2019).
- Extend mobile phone coverage to underserved areas. See "Geography" in the environmental factors.
- Digital skills bootcamps matched to local demand (B for technology jobs, C for earnings). A randomized trial in Peru tested an online digital skills bootcamp for young people. The program raised the probability of working in entry-level technology jobs two years later. But average effects on formal employment and income were null, because only some course tracks paid off. The authors link this to selection into tracks and limited local demand for certain digital skills (Novella, Freund, and Rosas-Shady, 2026).
- Not shown to work: free home computers alone to raise school achievement (B). The largest field experiment to give schoolchildren free home computers found that ownership and use rose substantially. But there were no effects on grades, test scores, credits earned, attendance, or disciplinary actions (Fairlie and Robinson, 2013). In Romania, children who won a voucher to buy a computer had significantly lower school grades but better computer skills. Parental rules on homework and computer use reduced these effects (Malamud and Pop-Eleches, 2011).
Proposed and experimental methods
Methods that are proposed, under trial, approved in some places, or tried and then failed. Each shows a stage label and an evidence rating. A stage label shows how far a method has progressed, not whether it works. The stage labels are explained on the technological factor page.
- Subsidized home broadband for low-income families (Approved but not scaled, B). Internet service providers or governments discount home internet for households below an income threshold. Comcast launched its Internet Essentials program in 2012 for low-income households with school-aged children. The program raised the chance that an eligible low-income adult had a job by nearly 1 percentage point. It also raised incomes among eligible people by over 1%. The estimated benefit for an enrolled household (about $2,202) was more than four times the estimated cost of the service (Zuo, 2021). The study uses a triple-differences design (comparing areas, eligibility, and timing) rather than randomization.
- Technology-neutral rural broadband grants (Approved but not scaled, contested, C). The U.S. Broadband Equity, Access, and Deployment (BEAD) program is a $42.45 billion federal grant program funded by the Infrastructure Investment and Jobs Act (NTIA, n.d.). New guidance on June 6, 2025, dropped the preference for fiber and let satellite and fixed wireless compete. NTIA expects about $21 billion in savings on deployment costs. By June 30, 2026, NTIA had approved 54 of 56 state and territory final proposals. But BEAD-funded infrastructure had been deployed in only two states (Benton Institute, 2026). In Illinois, whose plan was the last approved, 68.1% of locations went to fiber, 23.6% to fixed wireless, and 8.1% to low Earth orbit satellite (Hardesty, 2026). No outcome data exist yet.
- City-owned fiber networks (Approved but not scaled, contested, D). Local governments or public utilities build and run broadband networks. A study commissioned around Chattanooga's municipal fiber network reported $2.69 billion in benefits and 9,516 jobs created or kept in its first ten years (EPB, 2021). But a peer-reviewed difference-in-differences analysis found "no economically- nor statistically-significant effect" on nine labor market outcomes (Ford and Seals, 2021). A 2026 policy bulletin by the same lead author argues that the job estimates misapply a source study that found no effect in urban counties (Ford, 2026).
- Reversed: the U.S. Affordable Connectivity Program (Failed or reversed, B). The program gave monthly internet discounts to low-income households for about two and a half years. The FCC froze new enrollment in early February 2024. It ended the program on June 1, 2024, because Congress did not provide more funding. At that point over 23 million households were enrolled (FCC, 2024).
- Reversed: zero-rated "free basics" internet (Failed or reversed, B). Facebook's Free Basics let people use selected sites without data charges. In February 2016 India's telecom regulator ruled that "no service provider shall offer or charge discriminatory tariffs for data services on the basis of content." The ruling ended the program in India. It came after critics argued that the program violated net neutrality, the principle that all internet content is treated equally (Ghoshal, 2016).
- Failed: stratospheric balloon internet (Failed or reversed, B). Alphabet's Loon used high-altitude balloons to bring internet access to remote areas. Loon ran a pilot service in Kenya. Loon was shut down in January 2021. In its chief executive's words, the reason was that "we haven't found a way to get the costs low enough to build a long-term, sustainable business" (Via Satellite, 2021).
Sources cited on this page
- Bahia, K., Castells, P., Cruz, G., Masaki, T., Pedrós, X., Pfutze, T., Rodríguez-Castelán, C., & Winkler, H. (2024). The welfare effects of mobile broadband internet: Evidence from Nigeria. Journal of Development Economics, 170, 103314. DOI B Moderate
- Czernich, N., Falck, O., Kretschmer, T., & Woessmann, L. (2011). Broadband infrastructure and economic growth. The Economic Journal, 121(552), 505-532. DOI B Moderate
- UNICEF (2020). COVID-19: Are children able to continue learning during school closures? A global analysis of the potential reach of remote learning policies using data from 100 countries B Moderate
- Ford, G. S. (2026). When narratives replace evidence: Employment claims and the EPB municipal broadband network (Phoenix Center Policy Bulletin No. 84). Phoenix Center for Advanced Legal & Economic Public Policy Studies. link B Moderate
- Ford, G. S., & Seals, R. A. (2021). The rewards of municipal broadband: An econometric analysis of the labor market. Telecommunications Policy, 45(8), 102220. link C Limited
- Ghoshal, A. (2016, February 8). Facebook's Free Basics blocked by India over net neutrality rules. The Next Web. link B Moderate
- Hardesty, L. (2026, August 25). NTIA signs off on Illinois BEAD final proposal, the last of 56 proposals. Fierce Network. link B Moderate
- Zuo, G. W. (2021). Wired and hired: Employment effects of subsidized broadband internet for low-income Americans. American Economic Journal: Economic Policy, 13(3), 447-482. link B Moderate
- Bateman, M., Duvendack, M., & Loubere, N. (2019). Is fin-tech the new panacea for poverty alleviation and local development? Contesting Suri and Jack's M-Pesa findings published in Science. Review of African Political Economy, 46(161), 480-495. link C Limited
- Fairlie, R. W., & Robinson, J. (2013). Experimental evidence on the effects of home computers on academic achievement among schoolchildren. American Economic Journal: Applied Economics, 5(3), 211-240. link B Moderate
- Malamud, O., & Pop-Eleches, C. (2011). Home computer use and the development of human capital. Quarterly Journal of Economics, 126(2), 987-1027. link B Moderate
- Novella, R., Freund, R., & Rosas-Shady, D. (2026). Digital skills for better jobs: Experimental evidence from a bootcamp training program in Peru. Journal of Development Economics, 182, 103838. link B Moderate
- Suri, T., & Jack, W. (2016). The long-run poverty and gender impacts of mobile money. Science, 354(6317), 1288-1292. link B Moderate
- Ma, Y., Fairlie, R., Loyalka, P., & Rozelle, S. (2024). Isolating the "tech" from EdTech: Experimental evidence on computer-assisted learning in China. Economic Development and Cultural Change, 72(4), 1923-1962. link B Moderate
Every source for this factor is listed on the technological factor page.