## Internet Adoption Trends during COVID-19

## Source details

**Canonical URL:** [Internet Adoption Trends during COVID-19](https://www.imf.org/-/media/files/publications/imf-notes/2023/english/insea2023001.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/imf-notes/2023/english/insea2023001.pdf.md)
- [Structured JSON version](/-/media/files/publications/imf-notes/2023/english/insea2023001.pdf.json)

---

### I. Broad Trends in Digitalization
- Internet users per 100 inhabitants reached 91 in advanced economies in 2021.
- Internet access by region in 2021:
  - Middle East and Central Asia: 61 percent of the population.
  - Sub-Saharan Africa: 31 percent of the population.
- Mobile signal coverage (3G to 5G) showed near convergence across regions, but mobile cellular subscription lagged considerably in Middle East and Central Asia and in sub-Saharan Africa.
- High-speed LTE network coverage displays large cross-regional discrepancies.
- Fixed broadband traffic per capita and international internet bandwidth:
  - Grew fast in advanced economies, emerging and developing Asia, and emerging and developing Europe.
- 5G adoption in 2020:
  - Around 160 million 5G-using mobile devices in 2020, mainly clustered in northeast Asia.
- Definitions and notes:
  - LTE stands for Long-Term Evolution (sometimes referred to as 4G LTE).
  - Internet users are individuals who have used the internet in the past three months via both computer and mobile phone.
  - AE = advanced economies; EDA = emerging and developing Asia; EDE = emerging and developing Europe; LAC = Latin America and the Caribbean; MECA = Middle East and Central Asia; SSA = sub-Saharan Africa.

### II. Developments during COVID-19
- Internet access (extensive margin) did not accelerate substantially during the COVID-19 pandemic.
  - The annual growth of internet users accelerated only in emerging market economies during the pandemic, driven mainly by India.
  - Internet users per 100 inhabitants: example growth notes — the average annual growth of internet users per 100 inhabitants in emerging market economies was 4.6 percentage points between 2019 and 2021, compared with 8.4 percentage points between 2017 and 2019.
  - Advanced economies and low-income developing countries remained at historical trends since the outbreak of COVID-19.
- Internet speed (intensive margin) showed a marked acceleration across all income groups after the start of the pandemic.
  - The average download/upload speed accelerated considerably since the outbreak of COVID-19.
  - Likelihood-ratio (LR) test for a structural break in April or May 2020 (H0: no structural break):
    - Advanced economies: LR = 9.1775 (prob = 0.0102).
    - Emerging market economies: LR = 5.4734 (prob = 0.0648).
    - Low-income developing countries: LR = 4.7477 (prob = 0.0931).
- Data and methodology:
  - This section uses Ookla Speedtest Intelligence® data (self-administered internet speed tests).
  - Self-administered tests capture actual user experiences across locations, devices, and times and may be more accurate than advertised speeds.
  - Ookla’s estimates may have wider confidence intervals in countries with low internet access due to limited observations.
- Time windows used in figures:
  - Data before COVID-19: January 2017 to February 2020.
  - Data after the start of the COVID-19 pandemic: March 2020 to April 2022.
  - The gray line in figures refers to March 2020, when WHO declared COVID-19 a global pandemic.
- Labels used in figures:
  - AE = advanced economies; EME = emerging market economies; IND = India; LIDC = low-income developing countries; mbps = megabits per second.

### III. Price and Affordability of Broadband Services
- Trend in nominal prices:
  - Data-only mobile-broadband basket prices dropped by more than 50 percent since 2015.
  - By 2021, these prices were less than $5 per month in emerging market economies and low-income developing countries.
- Advanced-economy price dynamics:
  - In advanced economies, mobile-broadband prices spiked recently; this increase likely reflects users upgrading data packages even as data-unit prices decline.
- Affordability relative to income:
  - The average mobile-broadband basket price is nine times higher in low-income developing countries than in advanced economies when considered relative to per capita gross income.
- Country and year specifics:
  - A notable kink in broadband price in 2017 was driven by the US, where the data package price increased from $16 in 2016 to $33 in 2017 before dropping back to $22 in 2018.
- Drivers of internet prices (correlates identified):
  - Market structure:
    - More concentrated telecom markets allow broadband providers to exploit market power and charge higher prices.
  - Quality of internet services:
    - Nominal broadband prices are positively correlated with download speed.
    - The average nominal cost of a broadband package remains significantly higher in low-income developing countries even after adjusting for internet speed.
  - Economies of scale:
    - Marginal cost of telecom service tends to decrease with the number of consumers.
    - A negative correlation exists between population size and broadband package prices, consistent with economies of scale.
- Data sources and notes:
  - Prices based on current and historical basket definitions in USD purchasing power parities (PPP) 2020 and as a percent of monthly GNI per capita.
  - AE = advanced economies; EME = emerging market economies; LIDC = low-income developing countries; GB = gigabytes; GNI = gross national income.

### IV. Conclusions and Policy Implications
- Main findings:
  - During the COVID-19 pandemic there was little acceleration in growth in internet access (extensive margin) across all income groups.
  - There were clear improvements in internet speed (intensive margin) for existing users.
  - Affordability of digital services remains a significant concern for low-income developing countries despite declines in nominal internet prices.
  - The average mobile-broadband basket price in low-income developing countries is substantially higher relative to advanced economies when measured against income, exacerbating the challenge of closing the internet access gap.
- Fiscal policy recommendations to promote internet adoption and prevent widening digital divides:
  - Provide and finance digital infrastructure.
  - Subsidize internet adoption directly and indirectly.
  - Support programs for digital literacy.
- Additional contextual notes:
  - Digitalization has important macroeconomic implications; for example, firm-level digitalization during the pandemic shielded productivity and hours worked in digitalized entities.
  - India’s “Digital India” campaign (launched in 2015) is cited as a driver of rapid internet adoption predating the pandemic.

*Source: IMF Note NOTE/2023/001 — “Internet Adoption Trends during COVID-19,” March 2023.*

### Section 1

### Internet Adoption Trends during COVID-19

### I. Broad Trends in Digitalization
- Internet users per 100 inhabitants reached 91 in advanced economies in 2021.
- Internet access by region in 2021:
  - Middle East and Central Asia: 61 percent of the population.
  - Sub-Saharan Africa: 31 percent of the population.
- Mobile signal coverage (3G to 5G) showed near convergence across regions, but mobile cellular subscription lagged considerably in Middle East and Central Asia and in sub-Saharan Africa.
- High-speed LTE network coverage displays large cross-regional discrepancies.
- Fixed broadband traffic per capita and international internet bandwidth:
  - Grew fast in advanced economies, emerging and developing Asia, and emerging and developing Europe.
- 5G adoption in 2020:
  - Around 160 million 5G-using mobile devices in 2020, mainly clustered in northeast Asia.
- Definitions and notes:
  - LTE stands for Long-Term Evolution (sometimes referred to as 4G LTE).
  - Internet users are individuals who have used the internet in the past three months via both computer and mobile phone.
  - AE = advanced economies; EDA = emerging and developing Asia; EDE = emerging and developing Europe; LAC = Latin America and the Caribbean; MECA = Middle East and Central Asia; SSA = sub-Saharan Africa.

### II. Developments during COVID-19
- Internet access (extensive margin) did not accelerate substantially during the COVID-19 pandemic.
  - The annual growth of internet users accelerated only in emerging market economies during the pandemic, driven mainly by India.
  - Internet users per 100 inhabitants: example growth notes — the average annual growth of internet users per 100 inhabitants in emerging market economies was 4.6 percentage points between 2019 and 2021, compared with 8.4 percentage points between 2017 and 2019.
  - Advanced economies and low-income developing countries remained at historical trends since the outbreak of COVID-19.
- Internet speed (intensive margin) showed a marked acceleration across all income groups after the start of the pandemic.
  - The average download/upload speed accelerated considerably since the outbreak of COVID-19.
  - Likelihood-ratio (LR) test for a structural break in April or May 2020 (H0: no structural break):
    - Advanced economies: LR = 9.1775 (prob = 0.0102).
    - Emerging market economies: LR = 5.4734 (prob = 0.0648).
    - Low-income developing countries: LR = 4.7477 (prob = 0.0931).
- Data and methodology:
  - This section uses Ookla Speedtest Intelligence® data (self-administered internet speed tests).
  - Self-administered tests capture actual user experiences across locations, devices, and times and may be more accurate than advertised speeds.
  - Ookla’s estimates may have wider confidence intervals in countries with low internet access due to limited observations.
- Time windows used in figures:
  - Data before COVID-19: January 2017 to February 2020.
  - Data after the start of the COVID-19 pandemic: March 2020 to April 2022.
  - The gray line in figures refers to March 2020, when WHO declared COVID-19 a global pandemic.
- Labels used in figures:
  - AE = advanced economies; EME = emerging market economies; IND = India; LIDC = low-income developing countries; mbps = megabits per second.

### III. Price and Affordability of Broadband Services
- Trend in nominal prices:
  - Data-only mobile-broadband basket prices dropped by more than 50 percent since 2015.
  - By 2021, these prices were less than $5 per month in emerging market economies and low-income developing countries.
- Advanced-economy price dynamics:
  - In advanced economies, mobile-broadband prices spiked recently; this increase likely reflects users upgrading data packages even as data-unit prices decline.
- Affordability relative to income:
  - The average mobile-broadband basket price is nine times higher in low-income developing countries than in advanced economies when considered relative to per capita gross income.
- Country and year specifics:
  - A notable kink in broadband price in 2017 was driven by the US, where the data package price increased from $16 in 2016 to $33 in 2017 before dropping back to $22 in 2018.
- Drivers of internet prices (correlates identified):
  - Market structure:
    - More concentrated telecom markets allow broadband providers to exploit market power and charge higher prices.
  - Quality of internet services:
    - Nominal broadband prices are positively correlated with download speed.
    - The average nominal cost of a broadband package remains significantly higher in low-income developing countries even after adjusting for internet speed.
  - Economies of scale:
    - Marginal cost of telecom service tends to decrease with the number of consumers.
    - A negative correlation exists between population size and broadband package prices, consistent with economies of scale.
- Data sources and notes:
  - Prices based on current and historical basket definitions in USD purchasing power parities (PPP) 2020 and as a percent of monthly GNI per capita.
  - AE = advanced economies; EME = emerging market economies; LIDC = low-income developing countries; GB = gigabytes; GNI = gross national income.

### IV. Conclusions and Policy Implications
- Main findings:
  - During the COVID-19 pandemic there was little acceleration in growth in internet access (extensive margin) across all income groups.
  - There were clear improvements in internet speed (intensive margin) for existing users.
  - Affordability of digital services remains a significant concern for low-income developing countries despite declines in nominal internet prices.
  - The average mobile-broadband basket price in low-income developing countries is substantially higher relative to advanced economies when measured against income, exacerbating the challenge of closing the internet access gap.
- Fiscal policy recommendations to promote internet adoption and prevent widening digital divides:
  - Provide and finance digital infrastructure.
  - Subsidize internet adoption directly and indirectly.
  - Support programs for digital literacy.
- Additional contextual notes:
  - Digitalization has important macroeconomic implications; for example, firm-level digitalization during the pandemic shielded productivity and hours worked in digitalized entities.
  - India’s “Digital India” campaign (launched in 2015) is cited as a driver of rapid internet adoption predating the pandemic.

*Source: IMF Note NOTE/2023/001 — “Internet Adoption Trends during COVID-19,” March 2023.*

### Section 2

### insea2023001 - Section 2

### Cited works
- Jaumotte, Florence, Longji Li, Andrea Medici, Myrto Oikonomou, Carlo Pizzinelli, Ippei Shibata, Jiaming Soh, and Marina M. Tavares. 2023. “Digitalization during the COVID-19 Crisis: Implications for Labor Markets and Productivity in Advanced Economies.” IMF Staff Discussion Note 23/03, International Monetary Fund, Washington, DC.
- Kumar, Utkarsh, David Amaglobeli, and Mariano Moszoro. 2023. “Determinants and Social Dividends of Digital Adoption.” IMF Working Paper 23/065, International Monetary Fund, Washington, DC.
- University of Chicago. 2022. “Internet Equity Initiative.” Data Science Institute. Accessed February 28, 2023.

### Document identifiers and headings present in this section
- NOTESNOTES
- IMFIMF
- Internet Adoption Trends during COVID-19
- NOTE/2023/001

*insea2023001 - Section 2*

---


_Source: https://www.imf.org/-/media/files/publications/imf-notes/2023/english/insea2023001.pdf_