The Global South Resilience Stack

Energy is the foundation. It powers pumps and treatment plants, lights and refrigerators in clinics, cell towers and the phones that connect to them. Without it, the layers above are limited to what can be done by hand, by gravity or by diesel generator. Water is the first lifeline above it, and it has two tiers. Basic water — a protected well or handpump within a 30-minute round trip — can be built without grid power. Safely managed water, available on premises when needed and free from contamination, generally needs pumping, treatment and piped distribution, and so depends on power.

Health services sit on both. A health facility without electricity cannot keep vaccines cold, run an oxygen concentrator or deliver a baby safely at night, and one without water cannot keep infection under control. Health is also the most institution-heavy layer: it needs trained staff, medicines, supply chains and financing, which is part of why it moves slowly. Digital is the coordination layer. It needs power to exist, but once it does it feeds information back into every layer below: smart meters and pay-as-you-go billing for energy, sensors and payments for water, records, referrals, disease surveillance and cash transfers for health, and early warnings for all of them.

The map shows where each layer stands. Switch layers to see that the four maps differ. Electricity gaps are concentrated in sub-Saharan Africa; basic water gaps follow them but extend into fragile states elsewhere; health coverage gaps run across South Asia as well as Africa; and the internet gap is the widest of all, covering most of Africa and large parts of South Asia. The sequence profile layer combines them by asking which layer, taken in order from the foundation up, is the first to fall short.

The clearest dependency in the stack is the simplest. A person cannot use the internet without electricity somewhere in the chain — to charge the phone, run the tower and power the network. The data bears it out almost perfectly. Across 3,072 country-years since 2000 with both measures, internet use exceeded electricity access by more than five percentage points in only 10, a handful of cases in Namibia, Libya, Lesotho and Burkina Faso, which most likely reflect differences in survey timing and measurement between the two series. In every other case, internet use was no more than five points above electricity access, and usually far below it.

For health coverage and safely managed water the ceiling is softer but still visible. Grouping all country-years by electricity access shows a steady envelope. Where fewer than a quarter of people had electricity, the median UHC service coverage index was 30 and the highest recorded was 55; the median share with safely managed water was 10%. Where electricity access was 90–99%, the medians were 69 and 53%, and where it was universal they were 71 and 73%. Basic water is the exception: it reaches a median of 49% even where electricity is below a quarter, because handpumps and protected springs do not need power.

These envelopes are associations. Countries with more electricity are also richer, more urban and further along in time, and all of these raise health coverage and water quality. But the physical mechanisms are well documented — a vaccine fridge needs power, a treatment plant needs power — and the ceiling never runs the other way: there is no country-year with high health coverage and almost no electricity. The practical meaning is that the foundation cannot be skipped. Digital can leapfrog fixed lines, and water can start with handpumps, but no layer above can run far ahead of power.

A second way to read the dependencies is to ask where the other layers stood at the moment a country crossed a threshold on one of them. Among the 41 low- and middle-income countries whose electricity access crossed 90% between 2000 and 2024, the median country had basic water coverage of 90%, a UHC index of 64 and internet use of 21% in the year it crossed. Among the 40 that crossed 50% electricity, the medians were 69% basic water, a UHC index of 44 and internet use of 13%. Water and power rise together at the bottom of the stack; health follows; the internet comes last.

Looking from the top down gives the same order. The 80 countries whose internet use crossed 50% did so at a median electricity access of 99%, and none below 52%. The 30 countries whose UHC index crossed 60 did so at a median electricity access of 91%, with a tenth of them below 62%. The 13 countries whose safely managed water crossed 50% did so at a median electricity access of 94%. Each upper layer has, in practice, arrived on a foundation of power that was already nearly complete.

The order is not a rule that the layers must be built one at a time. Most of these countries were building several layers in parallel; the point is which ones were ready first. It is also partly an artefact of speed. The internet arrived late in history and spreads fast, so it appears at the top of the order everywhere. Health coverage is built slowly over decades, so it lags whichever layers move faster. The sequencing question in section 7 is how to turn this pattern into a plan.

Each layer has its own speed. Over the latest ten years of data, the median low- and middle-income country added 3.3 percentage points of internet use a year, 0.5 points of electricity access, 0.33 points of basic water and 0.26 points on the UHC service coverage index. The averages hide stalls: 34 countries made no measurable progress on health coverage over their latest decade, 14 none on basic water, and only three none on electricity or internet use. Global figures tell the same story. The UHC index rose from 54 in 2000 to 71 in 2023, but the WHO and World Bank project only 74 by 2030.

Extending each country's recent pace gives the year each layer reaches its target: 95% electricity, 95% basic water, a UHC index of 80 and 90% internet use. Seventy countries already have 95% electricity, but on trend 25 countries, with 1.2 billion people in 2050, still do not by 2050. For basic water the figure is 39 countries and 1.8 billion people. For the internet, despite its speed, 26 countries and 1.25 billion people fall short, rising to 32 if internet use cannot run more than five points ahead of electricity. For health coverage it is 91 countries — 3.2 billion people in 2050.

Only 28 of the 129 countries reach all four targets by 2050 at their recent pace. The binding layer in most of the rest is health. That reverses the usual intuition about sequencing. Because digital moves fastest, it can wait for the foundation and still catch up; because health moves slowest, it cannot wait for anything. A country that defers health investment until its power and water are complete will, at typical speeds, still be short of universal coverage in the second half of the century.

If the stack has a single point of failure, it is the primary health facility. The joint 2023 report by the WHO, the World Bank, IRENA and Sustainable Energy for All, Energizing Health, estimated that about a billion people in low- and lower-middle-income countries are served by health facilities with no electricity or an unreliable supply. In South Asia 12% of facilities had no electricity at all, and in sub-Saharan Africa 15%; only about half of hospitals in sub-Saharan Africa had reliable power. Rural facilities are worse off than urban ones in almost every country.

Water is missing in the same places. The WHO/UNICEF Joint Monitoring Programme found that in 2022, 78% of health facilities worldwide had basic water services, which left 1.7 billion people without a basic water service at their health facility, including 874 million whose facility had no water service at all. In the 60 fragile contexts it examined for 2023, only 63% of facilities had basic water and 19% basic sanitation. A clinic without both power and water cannot provide the services that move health coverage up the index.

This is where sequencing becomes a matter of design rather than order. The facility does not need to wait for the grid: solar systems with batteries can power lights, cold chains and equipment, run a pump for the facility's water and charge the tablets and phones that link it to records and payments. UNDP's Solar for Health initiative, largely financed by the Global Fund, reports solar electrification of around 1,000 health centres and storage facilities in about 15 countries, including 405 in Zimbabwe. Building the four layers together at the facility is the fastest way to raise the slowest one.

The explorer below shows the four layers for any of the 129 countries from 2000 to the latest year, with the recent pace extended to 2050 as dashed lines. The tiles give the year each layer reaches its target at that pace and the country's sequence profile — the first layer, from the foundation up, that falls short of its working threshold. Selecting a country on the map in section 1 also selects it here.

The simulator applies a multiplier to the recent pace of each layer in every country at once and recounts how many countries reach each target by 2050. It is a way to ask which layer matters most for the overall date of completion. Doubling the pace of digital, for example, does little if electricity is held back and the energy ceiling is switched on, because internet use cannot outrun power; doubling the pace of health changes the picture most, because health is the binding layer for most countries.

The arithmetic is linear and illustrative. Real acceleration is not free, coverage tends to slow as it approaches 100%, and the most difficult last miles are usually also the most expensive. The tool does not estimate costs or causes; it shows where the binding constraint lies at the speeds countries are actually achieving, and how far each layer would have to move to change the 2050 outcome.

The evidence points to a rule with four parts. Foundation first: because electricity caps every layer above it, closing the energy gap is the precondition for the rest, and in the 44 countries where access is below 80% it is the first priority. Slowest earliest: because health coverage moves at a tenth of the speed of the internet and is the binding layer for 2050 in most countries, health investment cannot wait for the foundation to be complete; it has to start now, at the same time as power. Bundle at the facility: the way to start the slowest layer without waiting for the grid is to build energy, water and connectivity together where health care is delivered.

Digital as accelerator: the internet spreads fast once power exists, and it can speed up the layers below it through metering, pay-as-you-go energy, mobile payments for water, digital health records and early warning. Its role in the sequence is not to come first but to multiply the return on the layers beneath it. Basic water, finally, is the hinge: it can and should be built off-grid wherever it is missing, but moving from basic to safely managed water follows power, as the envelope in section 2 shows.

Applying the working thresholds to the 129 countries gives five groups. Energy-first countries, below 80% electricity access, are 44 countries with 1.2 billion people, nearly all in sub-Saharan Africa. Water-first countries have power but not basic water: six countries, including Afghanistan, Yemen and Cambodia. Health-first countries have both but a UHC index below 65: 16 countries with 0.66 billion people, led by Pakistan and Bangladesh. Digital-first countries have the lower three layers but internet use below 70%: 12 countries with 1.64 billion people, led by India and the Philippines. The remaining 51, with 3.1 billion people, are consolidating.

Electricity access, basic and safely managed drinking water, internet use and population are from the World Bank World Development Indicators, downloaded in October 2026, with projected 2030 and 2050 populations from the World Bank's population projections. The UHC service coverage index (SDG 3.8.1) is the WHO series as published by Our World in Data, to 2023. The Global South is the 131 economies the World Bank classes as low, lower-middle or upper-middle income; 129 have data on all four layers. External figures are from Tracking SDG7 2025, the WHO/UNICEF Joint Monitoring Programme, the UHC Global Monitoring Report 2025, ITU Facts and Figures 2025, Energizing Health (2023), UNDP, the IEA and the World Bank's Mission 300 updates.

Envelope and crossing statistics pool all country-years since 2000 for which both series exist; they reflect time and income as well as dependency, and only the electricity ceiling on internet use is treated as physical. A crossing year is the first year a series reached a threshold after starting below it. Recent pace is the least-squares slope over the latest ten years of data with at least five observations; flat or falling series are given zero pace. Projections extend pace linearly, capped at 100, and arrival is the first year the projection reaches the target. The energy ceiling option caps projected internet use at projected electricity access plus five points.

The sequence profile is the first layer, in the order energy, water, health, digital, below 80% electricity, 85% basic water, a UHC index of 65 or 70% internet use. The UHC index is a composite rather than a share of people, so it is compared with its own target. Projections are illustrative extrapolations, not forecasts. The map uses Natural Earth 1:50m boundaries via world-atlas in an Equal Earth projection; boundaries do not imply endorsement. Related reports: from payments to power, the food–energy–water nexus and the geography of global vulnerability.