H Heuristics Digital Reports

H Heuristics · Digital Report № 2026-04 · September 2026

Breaking the Pollution–Poverty–Climate Nexus in Emerging Economies

Integrated strategies for cleaner growth, public health, resilience and economic development

Three legs, three clocks. The bundle is the same either way — what changes is which leg makes the case.

AuthorHunter Hughes
InstitutionH Heuristics
Published10 September 2026
Report №2026-04
Reading time21 min

Abstract

Air pollution is the second-leading risk factor for death worldwide, killing an estimated 8.1 million people in 2021 — about one death in eight — of which some 3.7 million are linked to household air pollution from burning solid fuels indoors. That last figure is better read as a poverty statistic than a pollution one: those deaths occur because roughly 2.1 billion people cannot afford or cannot obtain a cleaner way to cook. This report treats pollution, poverty and climate not as three problems that coincide but as a nexus of mutual causation, and asks why integrated strategies to address them, which have been analytically well understood for years, so consistently fail to be implemented.

The report's analytical core is that the three legs differ less in magnitude than in the speed at which an intervention pays back and in who receives the payment. Cleaner air pays back in months, locally, to the identifiable person breathing it; escape from poverty pays back over years to a generation, to the household; avoided emissions pay back over decades, globally, to no identifiable party. Institutions are correctly organised around those time constants — health ministries on annual mortality, social ministries on poverty headcounts, environment ministries on 2050 targets — with the consequence that an intervention paying out across all three timetables is scored by whichever ministry sponsors it, on that ministry's clock alone, using only the share of the return that lands in its own column. The full return is never assembled in one place.

Both directions of the pollution–poverty link are established. Poverty produces the pollution: a household chooses the fuel whose up-front cost it can meet this week, which makes energy poverty the proximate cause of household air pollution, with Sub-Saharan Africa carrying roughly half the clean-cooking deficit and about 85 per cent of the electricity-access deficit. Pollution produces the poverty: in 2022 an estimated 2.1 billion people faced financial hardship from out-of-pocket health spending, including 1.6 billion pushed into poverty or driven deeper into it, and in low-income countries out-of-pocket payments are the single largest source of health financing at slightly more than 40 per cent. The largest financial flow in the structure runs the wrong way and is widely mis-described: of an estimated US$7.4 trillion in fossil-fuel subsidies in 2024, only US$725 billion is explicit budgetary support, while US$6.7 trillion is implicit — unpriced air pollution and climate damage, which no treasury appropriates and which is settled instead in the lungs and household budgets of the least responsible and most exposed.

The implication is a sequencing rule rather than a new instrument: appraise on the fastest clock, and bank the slowest. Justify clean cooking, household electrification, clean urban transport and waste control on health and household-income returns that land inside a political cycle, and count the climate benefit as a dividend requiring no separate appropriation. The rule needs a portfolio filter, because the overlap between the legs is large but not total — remote utility-scale renewables deliver emissions cuts with little local air-quality gain, end-of-pipe controls cut exposure while entrenching combustion, and consumption subsidies relieve household budgets while worsening the exposure that is making the household ill. Subsidy reform in particular must run in the right order, with cash compensation and the clean alternative arriving simultaneously, as in Morocco's 2013–2015 reform, rather than as a price shock imposed ahead of any substitute.


Executive Summary

Pollution, poverty and climate are managed as three problems by three ministries on three timetables. At the level of the household that suffers all of them, they are one condition — and the reason it persists is that the same intervention pays out on three different clocks.

FINDING 01

A nexus, not a hierarchy

The three legs cause one another in both directions. Pollution drives poverty through illness and the health bills that follow it; poverty drives pollution because a household burns what it can afford; climate amplifies both and is fed by both. This is mutual causation, and it cannot be fixed one leg at a time.

FINDING 02

Three clocks

Cleaner air pays back in months, to the person breathing it. Escaping poverty pays back over years to a generation, to the household. Climate pays back over decades, to everyone and no one. Budgets are organised by clock — which is why an intervention that pays on all three gets appraised on whichever one its sponsoring ministry runs.

FINDING 03

Appraise on the fastest clock

Filed as climate policy, clean cooking is judged on a decadal, globally diffuse return and loses. Filed as health policy, it is judged on a mortality return that lands inside a single budget cycle — and the climate benefit rides along free. The bundle does not change. The case for it does.

Air pollution is the second-leading risk factor for death worldwide. The Health Effects Institute's State of Global Air 2024 attributes 8.1 million deaths in 2021 to it — roughly one death in eight — of which some 3.7 million are linked to household air pollution, overwhelmingly from burning solid fuels indoors. That last figure is the hinge of this report, because it is not really a pollution statistic. It is a poverty statistic wearing a pollution uniform: those deaths occur because roughly 2.1 billion people cannot afford or cannot obtain a cleaner way to cook.

8.1M
Deaths a year from air pollution — the second-leading risk factor for death worldwide
Health Effects Institute, State of Global Air 2024
1.6bn
People pushed into poverty, or driven deeper into it, by out-of-pocket health spending in a single year
WHO, Tracking Universal Health Coverage (2023)
2.1bn
People without clean cooking — the proximate cause of 3.7 million deaths a year
Tracking SDG 7 (2025)
$6.7tn
Annual implicit fossil-fuel subsidy — unpriced air pollution and climate damage, paid by households rather than treasuries
IMF, subsidy data update (2025)

An earlier report in this series treated fossil-fuel combustion as a shared root feeding four separate risk registers — a tree, with one driver and four branches. That structure does not describe what this report is about. Poverty is not a branch of combustion; it is both a cause and a consequence of the pollution it sits beside. The right picture is a triangle with feedback in every direction, and the policy implications are correspondingly different: not one intervention at a root, but a sequencing rule for a bundle.

That rule is the report's practical contribution. Integrated strategies fail in practice not because the integration is analytically wrong but because integrated things still have to be bought by a single budget holder answering to a single timetable. The way through is to justify the bundle on its fastest-returning leg — which, almost everywhere in the emerging world, is health — and to treat the slower-returning legs as dividends that require no separate appropriation.

Filed as climate policy, clean cooking is judged on a return that arrives in the 2050s. Filed as health policy, it is judged on one that arrives before the next election. The sequencing rule

1. Three Problems, One Household

The nexus is easiest to see from below. At the level of the institutions that manage it, the three legs are separate portfolios; at the level of the household that lives it, they are a single circumstance.

Consider a household on the edge of a fast-growing city in South Asia or Sub-Saharan Africa. It cooks on charcoal or wood because bottled gas costs more per month than it can reliably spare and the grid connection, if it exists, is intermittent. The smoke fills a room where children sleep. The nearest paved road carries diesel traffic past the door. When someone falls ill, the consultation and the medicine are paid in cash, from the same budget that buys the fuel. In the wet season the settlement floods; in the dry season the heat makes outdoor work shorter and less productive.

Every element of that description belongs to a different ministry, a different statistical series, and a different international negotiation. None of them, taken alone, describes the household's situation. What the household experiences is a single circuit in which each element funds the next.

Figure 1 — The nexus, and the clock each leg runs on

The pollution–poverty–climate nexus Three nodes — pollution, poverty and climate — connected by double-headed arrows showing mutual causation. Pollution causes poverty through illness and health costs, and poverty causes pollution because households burn the cheapest available fuel. Climate shares a combustion source with pollution and pushes households back into poverty. Each node carries a different time constant: pollution pays back in months, poverty over a generation, climate over decades. POLLUTION 8.1m deaths a year CLOCK: MONTHS POVERTY 2.1bn without clean cooking CLOCK: A GENERATION CLIMATE shared combustion source CLOCK: DECADES illness, lost work, health bills you burn what you can afford one source: combustion heat and ozone worsen exposure shocks push households back · poverty forecloses the clean option THE NEXUS every arrow runs both ways

Schematic. The distinguishing feature is that no leg is upstream of the others: each is both a cause and a consequence. Death figures: State of Global Air 2024; access figures: Tracking SDG 7 (2025).

The institutional consequence is severe. A circuit with no upstream point has no natural owner. A ministry of health owns the illness but not the stove; a ministry of energy owns the stove but is measured on connections rather than on mortality; a ministry of environment owns the emissions but has neither budget nor mandate for either. Each can act correctly within its remit and leave the circuit turning.


2. Three Clocks

The analytical core. The three legs differ not mainly in size but in the speed at which an intervention pays back, and in who receives the payment — and policy is organised around exactly those two variables.

Take the three legs in turn and ask a single question of each: if the intervention happens today, when does the benefit arrive, and to whom?

Pollution runs on months. Exposure and harm are close together in time, and the relationship is reversible: reduce the concentration of fine particulates a household breathes and the respiratory and cardiovascular consequences begin to fall within the same year. The benefit is local and it accrues to the identifiable person who was breathing the air.

Poverty runs on a generation. A household's escape from a poverty trap is measured in years — recovered earnings, avoided catastrophic spending, a business that can hold stock. The deepest part of the return, a child who completes schooling because the household did not have to liquidate to pay for treatment, is realised over a working life.

Climate runs on decades. An emission avoided today changes a temperature path over decades and a damage path over a century. The benefit is global, non-excludable, and accrues to no identifiable party.

Figure 2 — Time to first measurable benefit, by leg

This figure renders the report's own analytical ordering, not measured lags: the values are judgements about when a benefit first becomes measurable, drawn from the mechanisms discussed in Sections 3 and 4, and are intended to show the spread between the legs rather than to date any particular programme. The logarithmic axis is deliberate — the argument is that the three clocks differ by orders of magnitude.

The spread is the finding. For every intervention in the figure the health return arrives first, the poverty return second, and the climate return last — typically an order of magnitude later. And crucially, the ordering is the same for all of them: there is no clean-energy intervention whose climate benefit lands before its health benefit, because the health benefit is produced by the same act of not burning something, at the point where people are.

2.1 Why the clocks fragment the budget

Institutions are built around time constants. A health ministry is accountable on annual mortality and morbidity statistics. A social-protection ministry works to poverty headcounts revised every few years. An environment ministry answers to targets set for 2035 and 2050 and reports into a global process. These are not failures of coordination; they are the correct design for each mandate taken separately.

The trouble is that they make an integrated intervention unbuyable. Suppose a clean-cooking programme returns benefits on all three legs. Whichever ministry sponsors it must justify it against its own hurdle, on its own clock, using only the share of the return that lands in its own column. A health ministry counts the mortality benefit and ignores the emissions. An environment ministry counts the emissions, discounts them across four decades, and finds a poor result per dollar against a utility-scale alternative. The full return is never assembled in one place, and the programme is either undersized or unfunded.

The sequencing rule

If the bundle is the same either way, the question is not which leg matters most but which leg makes the case. Appraise on the fastest clock; bank the slowest. Justify clean cooking, clean transport and household electrification on the health and household-income returns that arrive inside a political cycle, and count the climate benefit as a dividend that required no separate appropriation. The physical intervention is unchanged. What changes is whether it gets bought.


3. How the Trap Closes

Both directions of the pollution–poverty link, and the mechanism that converts a health event into a permanent loss of income.

3.1 Poverty produces the pollution

The direction usually left out of environmental analysis is the one that runs from poverty to pollution. A household does not choose solid fuel over gas or electricity because it prefers smoke. It chooses the fuel whose up-front cost it can meet this week, and solid fuels are collected or bought in tiny quantities while a cylinder or a connection requires a lump sum. Energy poverty is therefore not merely correlated with household air pollution; it is its proximate cause.

Figure 3 — The access deficit that produces household air pollution

Roughly 2.1 billion people lack clean cooking and about 655 million lack any electricity access. Sub-Saharan Africa carries around half of the clean-cooking deficit and roughly 85 per cent of the electricity-access deficit; regional splits here apply those reported shares. Sources: Tracking SDG 7 (2025) and the 2025 joint release; regional shares as reported by the IEA.

This is why the 3.7 million annual deaths from household air pollution are best read as a poverty outcome. The WHO's assessment of household air pollution makes the distributional point plainly: the burden falls on women and young children, who spend the most time near the hearth, in the households least able to change the arrangement.

3.2 Pollution produces the poverty

The return direction is usually discussed in mortality terms, which understates it. Death is the extreme of a distribution whose bulk is illness — and in most of the emerging world, illness is a direct financial shock, because it is paid for at the point of use.

Figure 4 — Financial hardship from out-of-pocket health spending, 2022

An estimated 2.1 billion people faced financial hardship from out-of-pocket health spending, including 1.6 billion pushed into poverty or driven deeper into it. Source: WHO, Tracking Universal Health Coverage: 2023 Global Monitoring Report; see also the WHO summary.

That figure is the mechanism by which a pollution problem becomes a poverty trap. It is not the value of a statistical life; it is cash leaving poor households to pay for conditions that clean air would largely have prevented. And the reason the shock lands on the household rather than on a public system is structural.

Figure 5 — Who pays for health in low-income countries

In low-income countries, out-of-pocket payments accounted for slightly more than 40 per cent of current health expenditure as of 2022, external aid for roughly 31 per cent, and domestic government resources for only about 22 per cent; the remainder is other sources. Source: WHO Global Monitoring Report.

When households are the largest single financier of health care, every avoidable illness is a transfer out of household savings and into treatment. Prevention, in that setting, is not a public-health nicety. It is one of the more effective income-protection instruments available.

3.3 The exposure inversion

The two directions compound into an inversion that is the moral centre of the nexus. The households that contribute least to ambient pollution and least to emissions are the ones most exposed to both, and the least able to buy their way out of exposure: no filtration, no air conditioning, no option to live away from the road or the kiln, and work that is disproportionately outdoors. Exposure, in other words, is inversely related to responsibility and to means at every step.

A household does not choose solid fuel over gas because it prefers smoke. It chooses the fuel whose up-front cost it can meet this week.

4. Climate: Amplifier and Free Rider

The third leg does two distinct things in this structure, and conflating them is a common source of bad policy.

Climate enters the nexus in two quite separate roles, and it is worth being precise about which is which.

As an amplifier, warming makes the other two legs worse through mechanisms that are already operating. Heat is directly lethal — the WHO estimates roughly 489,000 heat-related deaths a year — and it interacts with pollution chemistry, since higher temperatures accelerate ground-level ozone formation and wildfire smoke adds to particulate loads. The 2025 Lancet Countdown assessment frames climate inaction explicitly as an ongoing mortality event rather than a future one. On the poverty leg, climate shocks push households back down the ladder they were climbing, and a household that has just lost a harvest returns to the cheapest available fuel.

As a free rider, climate is the leg that gets its benefit without needing to make its own case. Because ambient and household pollution and greenhouse gases very largely share a physical source — combustion — an intervention that removes the combustion for health reasons removes the emissions as a by-product. The climate return requires no separate justification, no separate budget, and no separate constituency. It arrives attached to something else that was worth doing anyway.

4.1 The overlap is large but not total

It would be convenient to claim that everything good for one leg is good for all three. That is not true, and the exceptions matter for portfolio design.

The design implication is that the sequencing rule needs a portfolio filter in front of it. Appraising on the fastest clock is the right move only for interventions that genuinely pay on more than one leg. Section 5 sets out which those are.


5. The Integrated Portfolio

Which interventions actually pay on all three legs, sorted by the clock they can be sold on.

Table 1 — Interventions by leg and by clock

Intervention Health Poverty Climate Sell it as
Clean cooking and electrified heat Very strong, months Strong, years Moderate, decades Health and household budgets
Household and community electrification Strong, years Very strong, years Moderate, decades Income and services
Clean urban transport Very strong, months Moderate, years Strong, decades Urban health and time saved
Industrial emission controls Strong, months Weak Weak to moderate Health, but not a nexus fix
Grid decarbonisation Moderate, years Moderate, years Very strong, decades Cost of power and security
Waste management and open-burning control Strong, months Moderate, years Moderate, decades Municipal health
Fossil-fuel consumption subsidies Negative Positive, immediate Negative Replace with cash transfers

Strength ratings are analytical judgements informed by the evidence in Sections 3 and 4, not measured effect sizes; the timings are those of Figure 2. The final row is included because it is the intervention emerging economies most often already have in place, and the only one in the table that runs against two of the three legs.

Read by row, the table identifies a small set of interventions that are strong on the two fast legs and moderate on the slow one. These are the nexus-breaking investments, and clean cooking is the archetype: the health return is the largest and fastest available in the entire adaptation and mitigation repertoire, the household-budget return follows within a year or two, and the emissions reduction arrives on its own schedule without anyone having to pay for it separately.

5.1 Cost is no longer the obstacle for power

One objection to integrated strategies has expired. For electricity generation, the clean option is no longer the expensive option.

Figure 6 — Unsubsidised levelised cost of new generation

Ranges reflect resource quality and project conditions. Source: Lazard, Levelized Cost of Energy+ (2024). Note that these are generation costs, not delivered costs: the cost of capital, grid access and last-mile distribution determine what a household actually pays.

That last qualification matters and should not be glossed. Cheap generation does not automatically become cheap electricity in a household that has no connection, no meter and no capacity to pay a lump sum. The binding constraint at the household end is finance and distribution, not the cost of a megawatt-hour — which is precisely why the poverty leg cannot be dropped from the analysis.


6. The Subsidy Inversion

The largest single financial flow in the nexus runs the wrong way, and it is not paid by the treasuries that are usually blamed for it.

The IMF's 2025 update to its global fossil-fuel subsidy estimates puts explicit subsidies — direct budgetary support that undercharges for supply cost — at US$725 billion in 2024. Implicit subsidies, chiefly the uncosted externalities of local air pollution and climate damage, come to a further US$6.7 trillion, for a combined total approaching US$7.4 trillion, or close to seven per cent of world GDP.

Figure 7 — The fossil-fuel subsidy, 2024

The explicit component is roughly a tenth of the total. Source: IMF (2025); see also the IMF's energy subsidies portal.

The word "subsidy" invites a fiscal reading — money a government pays out — and for the smaller explicit component that reading is correct. For the implicit component it is actively misleading, and the correction is the point of this section. Nobody appropriates US$6.7 trillion. That figure is the value of harm that is done and not paid for: the respiratory disease, the shortened lives, the lost working days and the climate damage. It is not a transfer from a treasury to a fuel. It is a transfer from the people who absorb the harm to the people who avoid its cost.

Sections 3.2 and 3.3 identified who those people are. The implicit subsidy is settled in the out-of-pocket health spending that pushed 1.6 billion people into or deeper into poverty, and it is settled disproportionately by households with the least exposure choice and the smallest contribution to the harm. Read this way, the largest financial flow in the nexus is a regressive transfer roughly ten times the size of the fiscal one that attracts the political attention.

The reform trap

This does not make subsidy reform simple, and the honest caution is important. Removing an explicit consumption subsidy raises fuel prices immediately, and the burden falls first on poor households — visibly, in the same week. The implicit subsidy they were also paying is invisible and diffuse. A reform that removes the visible transfer without compensating for it therefore worsens the poverty leg in the short run in order to improve the pollution and climate legs in the long run, which is the sequencing rule of Section 2 run backwards. Reform works when the removed subsidy is replaced by targeted cash transfers and by the clean alternative arriving at the same time — not before, and not after.


7. Evidence from the Record

What the experience of scaling access and reforming prices actually shows.

7.1 Where the fastest clock has been used

The clearest demonstrations of the sequencing rule are clean-cooking programmes, which have been driven overwhelmingly by health and household-welfare arguments rather than climate ones. Progress is real: the 2025 joint Tracking SDG 7 assessment records continued gains in both electricity and clean-cooking access, driven largely by South and East Asia, even as the deficit remains concentrated in Sub-Saharan Africa.

The IEA's assessment of universal clean-cooking access in Africa sets out the same logic in its cost structure: the sums required are small by the standards of energy investment, the health returns are immediate and large, and the emissions benefit is genuine but is not what makes the case. On the electricity leg, Mission 300 — the World Bank and African Development Bank effort to connect 300 million Africans by 2030 — is likewise argued on services, income and opportunity rather than on carbon.

7.2 Morocco: taking the price signal seriously

Morocco reformed fossil-fuel subsidies decisively over 2013 to 2015, removing price support for petrol, diesel and fuel oil — a politically difficult step most peer economies have postponed. The reform removed the artificial cost advantage of dirty fuel and freed fiscal space, and it was accompanied by a large renewable build-out rather than preceding it. That ordering is the substantive lesson: the subsidy came out as the alternative came in.

7.3 The counter-pattern

The failures follow a recognisable shape. A government removes a fuel subsidy under fiscal pressure, without a compensating transfer and without the clean alternative in place. Prices rise, transport and food inflation follow within weeks, the political cost is immediate and concentrated, and the reform is partially or wholly reversed. The pollution and climate benefits that would have accrued over years and decades never arrive, because the programme did not survive its first month. Nothing in that sequence is a failure of the underlying economics; it is a failure to lead with the leg that pays fastest for the households bearing the cost.


8. Appraise on the Fastest Clock

An agenda organised around the sequencing rule, because the analytical case for integration has been made repeatedly and has repeatedly failed to be bought.

8.1 Change the appraisal, not the intervention

8.2 Break the poverty→pollution direction

8.3 Break the pollution→poverty direction

8.4 Correct the price, in the right order

8.5 Conclusion

The analytical case for treating pollution, poverty and climate together is not new and is not seriously contested. What has been missing is an account of why integration keeps failing in practice despite being obviously correct in principle. This report's answer is that integrated problems still have to be bought by non-integrated institutions, each accountable on a different timetable, and that an intervention paying out across three timetables gets scored on whichever one its sponsor happens to run.

That diagnosis has an unusually cheap remedy, because it requires no new technology, no new money and no new institution — only a change in which leg carries the argument. Clean cooking, household electrification, clean urban transport and waste control are all strong on health within months and on household budgets within a year or two. Their climate benefit is real, is large in aggregate, and does not need to be sold, because it arrives attached to a decision that a health or finance ministry can already justify.

The prize is the circuit itself. A household that stops burning solid fuel indoors stops generating the illness that empties its budget, keeps the income that would have paid for treatment, retains the schooling that would have been sacrificed, and stops emitting — in that order, on three different clocks, from one decision. Breaking the nexus does not require acting on all three legs at once. It requires acting on one leg in a way that happens to run through all three, and making the case on the leg that pays first.


References

Every quantitative claim above is attributed inline. The principal sources are collected here.


Metadata

Keywords
air pollutionhousehold air pollutionclean cookingenergy povertypollution poverty nexusco-benefitsout-of-pocket health spendingcatastrophic health expenditurefossil fuel subsidiesimplicit subsidiesemerging economiesintegrated policytime preferencesubsidy reform
Topics
Public Health Emerging Markets Energy Transition Institutions & Governance
JEL classification
Q53, I14, I32, Q56, O13 — air pollution and environmental externalities; health and inequality; measurement and analysis of poverty; environment and development; agriculture and natural resources in development
Data and method
This report synthesises institutional research on air pollution, health financing, energy access and fossil-fuel pricing, including the Health Effects Institute's State of Global Air 2024; the WHO's Tracking Universal Health Coverage 2023 Global Monitoring Report and its fact sheets on household and ambient air pollution and on heat; the 2025 Lancet Countdown; Tracking SDG 7 (2025) and the accompanying ESMAP release; the IEA's assessment of universal clean-cooking access in Africa; the World Bank and African Development Bank's Mission 300; the IMF's 2025 fossil-fuel subsidy data update and its account of Morocco's 2013-2015 subsidy reform; and Lazard's Levelized Cost of Energy+ (2024). Every quantitative claim is attributed inline. Figure 1 is a conceptual schematic. Figure 2 renders the report's own analytical ordering of when a benefit first becomes measurable, not observed lags, and its caption says so; Table 1's strength ratings are likewise analytical judgements rather than measured effect sizes. Regional splits in Figure 3 apply reported regional shares to reported global totals. The report is analytical rather than predictive.
Report
H Heuristics Digital Report № 2026-04 · Published 10 September 2026
Licence
CC BY-NC-ND 4.0
Cite as
Hunter Hughes (2026). Breaking the Pollution–Poverty–Climate Nexus in Emerging Economies: Integrated strategies for cleaner growth, public health, resilience and economic development. H Heuristics Digital Report 2026-04. https://digitalreports.hheuristics.com/reports/pollution-poverty-climate-nexus/
↑ Back to top