Thinking · Obligation

Section 135 and climate-health. How to defend the line item.

Abhijith Magal · 20 September 2026 · 11 min read

Can CSR money pay for climate-health work? Yes, and the question is usually asked in the wrong direction. Schedule VII of the Companies Act, 2013 is broad enough that a Section 135 CSR climate health programme clears eligibility on the first reading: item (i) names promoting health care including preventive health care, item (iv) names ensuring environmental sustainability and maintaining the quality of soil, air and water, and item (xii) names disaster management, including relief, rehabilitation and reconstruction. A programme that keeps a worker or a patient reachable through a heat season or a flood sits inside item (i) on its own terms.

The harder question is the one nobody asks until the second year. Not is this allowed, but will this survive being looked at. A CSR line item gets read four times after it is approved: by the CSR committee, by the Board in its annual report, by the statutory auditor, and, above a threshold, by an independent impact assessor. Each reads for something different. A programme designed only for the first reader fails the other three.

What does Schedule VII actually allow for climate-health?

Three entries carry almost all climate-health programme design.

Item (i) covers eradicating hunger, poverty and malnutrition, promoting health care including preventive health care and sanitation, and making available safe drinking water. Preventive health care is the phrase doing the work. Heat protocols, hydration and shade infrastructure, community health worker capacity, screening camps, water quality and continuity of treatment all read naturally as preventive health care.

Item (iv) covers ensuring environmental sustainability, ecological balance, conservation of natural resources and maintaining quality of soil, air and water. Air quality work and water security work belong here when the outcome is environmental, and under item (i) when the outcome is a health result in a defined population.

Item (xii) covers disaster management, including relief, rehabilitation and reconstruction. Flood and cyclone response sits here, and the entry is the most commonly misused of the three, because relief is easy to spend and hard to evidence once the water goes down.

One project can carry a primary entry and a secondary one. What it cannot do is float between them in the paperwork. The annual action plan recommended by the CSR committee under Rule 5(2) has to name the projects, the entries, the manner of execution, the modalities of fund utilisation, the implementation schedule and the monitoring mechanism. If the entry moves between the plan and the annual report, somebody will ask why.

The defensibility chain, link by link

Every defensible CSR project runs the same sequence. The Rules set it; the only variable is how well each link is evidenced.

Break 1: the employee exclusion

Rule 2(1)(d)(iv) puts activities benefitting employees of the company, as employee is defined in the Code on Wages, 2019, outside CSR altogether. This is the single most common way a climate-health programme dies in review, because the most obvious climate-health population inside a company's reach is its own workforce.

The fix is a design decision taken at the start, not a rewrite at the end. Draw the programme around a geography or a population, not around a payroll. A heat and water programme built for wards around a plant, open to everyone in them, is a community programme that some employees' families happen to live inside. A heat programme built for the plant's workers is a duty-of-care obligation the company owes anyway, which is a good thing to run and the wrong budget to run it from. The duty-of-care route has its own logic, set out in heat is now a workplace duty of care.

The same instinct catches Rule 2(1)(d)(i), the normal course of business exclusion, and Rule 2(1)(d)(v), which rules out activities sponsored for marketing benefit. A pharmaceutical company funding awareness of a condition it sells into has a harder file to defend than the same company funding continuity of treatment for a cohort it does not sell to.

Break 2: an eligible partner in the wrong terrain

Rule 4(1) is a gate, not a ranking. It admits a section 8 company, a registered public trust or a registered society with 12A registration and 80G approval, or an entity set up by government or by statute, and where the entity was not established by the company itself it requires a three-year track record in similar activities. Rule 4(2) adds Form CSR-1 and a unique CSR Registration Number, in force since 1 April 2021.

Pass those and forty organisations are still eligible. Nothing in the statute helps a company choose between them, which is why climate-health projects so often stall in the month the climate arrives. A partner who has never worked through a flood season in that district is not disqualified by anything. They simply stop delivering in June. The comparison problem, and a published method for it, sit in eligible is not the same as suitable, and the document-level checks are in the diligence checklist.

Break 3: an evidence file that stops at the money

Rule 4(5) requires the Board to satisfy itself that funds were used as approved, with the Chief Financial Officer certifying to that effect. That certificate is about money. The impact assessment under Rule 8(3) is about people, and it arrives a year after the project closes, run by an agency the company did not brief at the start.

Most files can answer the money question and not the people question. Utilisation certificates, invoices and photographs establish that spending happened. They do not establish how many people were reached, whether the same people were still covered at the end, or what would have happened anyway. The cheapest correction is also the least popular: fix the denominator before the first rupee moves. Name the cohort, count it, and record the two or three numbers the assessment will ask for later.

A CSR file that can prove the money moved and not that anybody was reached has documented a transaction, not a programme.

For climate-health work there is one more number worth carrying, and almost nobody does: a continuity field. One date per person per entitlement, recording when cover lapses. It costs nothing at enrolment and it is the only way to show, at closure, that a programme delivered continuity rather than a good enrolment week.

Where climate-health CSR should point

If the programme is meant to reduce climate harm to health, the exposure data decides the geography and the sector before any local preference does. The Lancet Countdown's 2025 India data sheet attributes the labour-hour loss from heat in 2024 across sectors, and the split is lopsided enough to plan from.

Two implications follow for programme design. First, a climate-health CSR programme aimed at agricultural or construction households is aimed where the measured harm concentrates, which is an argument that holds in a Board room without any appeal to conscience. Second, these are the two workforces least likely to hold the documents that unlock what they are already entitled to, which is where a CSR rupee does work no company benefit scheme can do.

What changed in 2026

The CSR Rules were amended on 27 May 2026 to allow a fourth route. Under the new Rule 4A, a company may carry out CSR through a zero coupon zero principal instrument issued by a not for profit organisation registered on the Social Stock Exchange segment of a recognised stock exchange. Two conditions are worth knowing before anyone gets excited: spending through such instruments cannot exceed ten per cent of the company's total CSR expenditure for the year, and a project funded that way is exempt from impact assessment.

That exemption is the part to read twice. It removes the one mechanism that independently tests whether a project reached anybody. For a climate-health programme, where the whole question is whether cover survived the season, buying the exemption is a poor trade.

The overhead argument you are probably losing

Two provisions of the Rules answer the objection that design and measurement are expensive overhead, and neither gets used often enough in a pricing conversation.

Rule 7(1) caps administrative overheads at five per cent of total CSR expenditure. Rule 2(1)(b) then defines those overheads as expenses for general management and administration of the CSR function, and expressly excludes expenses directly incurred for the designing, implementation, monitoring and evaluation of a particular project or programme. Design, monitoring and evaluation are project cost. They do not consume the five per cent.

Rule 8(3)(c) goes further for companies above the threshold. Impact assessment expenditure can be booked to CSR for that financial year up to two per cent of total CSR expenditure or fifty lakh rupees, whichever is higher. That is a statutory, recurring line for independent measurement, sitting inside the CSR budget rather than competing with it.

A design that survives all four readers

The programme that holds up is the one built backwards from the impact assessment, not forwards from the budget. Four questions, asked before approval, settle most of it.

• BoundaryWho exactly is in the cohort, and is the boundary geographic rather than payroll-based? Write the number down before the programme starts.
• BaselineWhat is true today for that cohort, on the two or three measures the programme claims to move? Without this, every later number is an assertion.
• ContinuityWhat lapses, and when? One renewal date per person per entitlement turns a coverage claim into a testable one.
• DisruptionWhich months does climate take out of the delivery calendar in this district, and what does the design do in those months rather than around them?

The fourth question is the one that distinguishes a climate-health programme from a health programme running in a warming country. A design that quietly assumes twelve uninterrupted months will under-deliver in the year a heatwave lands on top of a flood, and the annual report will record that as underperformance by the partner.

That gap is what a climate-health CSR programme is buying down. Naming it that way also gives the Board report something better than activity counts to report against, and it is the frame the rest of this site is built on, set out in climate change is a healthcare story and defined at what is climate-access.

Common questions

Can CSR funds be used for a climate-health programme?

Yes. Schedule VII covers promoting health care including preventive health care under item (i), environmental sustainability and the quality of soil, air and water under item (iv), and disaster management including relief, rehabilitation and reconstruction under item (xii). A programme that protects health against climate stress sits inside item (i) on its own terms, and can draw on (iv) and (xii) where the design justifies it.

Which Schedule VII entry should a climate-health programme be booked under?

The entry that matches what the programme delivers to people, not the one that sounds most current. Health outcomes belong under item (i), air, water and soil quality work under item (iv), and cyclone, flood and heat response under item (xii). One project can name a primary and a secondary entry, provided the annual action plan records which activity sits where.

Does a CSR implementing partner need CSR-1 registration?

Yes. Under Rule 4(2), every entity undertaking CSR activity has had to file Form CSR-1 with the Registrar since 1 April 2021 and receive a unique CSR Registration Number. Rule 4(1) sets who may act at all: a section 8 company, registered public trust or registered society with 12A registration and 80G approval, and, where the entity was not set up by the company or a government, a three-year track record in similar activities.

When is a CSR impact assessment mandatory?

Under Rule 8(3), a company with an average CSR obligation of ten crore rupees or more over the three immediately preceding financial years must have an independent agency assess projects with outlays of one crore rupees or more, completed not less than a year before the study. The cost can be booked to CSR up to two per cent of total CSR expenditure for that year or fifty lakh rupees, whichever is higher.

Why do CSR health programmes fail an audit even when the activity is eligible?

Usually for one of three reasons. The programme benefits the company's own employees, which Rule 2(1)(d) excludes. It sits in the normal course of business, which the same rule excludes. Or the file cannot show what the money bought: no need assessment, no baseline, and no beneficiary count that ties back to the disbursal the Chief Financial Officer certified under Rule 4(5).

How much of a CSR budget can be spent on administration?

Rule 7(1) caps administrative overheads at five per cent of total CSR expenditure for the financial year. Rule 2(1)(b) defines those overheads as general management and administration of the CSR function and expressly excludes expenses directly incurred on designing, implementing, monitoring and evaluating a particular project. Design and measurement are project cost, not overhead.

Abhijith Magal

Founder, Syntropy Earth. Nine years across two global pharmaceutical multinationals in patient access and commercial roles, with health-equity work alongside the WHO-Foundation and UNICEF. He works on climate-access: where climate disruption breaks the link between patients and care.

This page describes the law as published. It is not legal or audit advice, and decisions on your own CSR filings should be taken with your advisers.

Last updated: 20 September 2026