Community welfare paths: micro-granting or direct aid?

In a single year, the Wandsworth Climate Action Microgrant programme routed £100–£500 awards through 60 community and school projects. In a separate randomised experiment in rural Kenya, researchers delivered one-time cash transfers of roughly US$1,000 to more than 10,500 poor households across 653 villages and recorded an estimated local fiscal multiplier of 2.7. Both interventions sit at the small-money end of the social welfare budget, yet they pursue fundamentally different theories of change: one finances locally designed activity, the other transfers purchasing power directly to people in need. Choosing between them is not a moral question. It is a question of which bottleneck the funder is trying to relieve, on what timeline, and under what regulatory ceiling.

Defining the scope: from individual cash to community-led projects

Before any comparison can be drawn, the two terms need to be pinned down. Neither is a fixed budget category, and treating them as interchangeable leads to misallocation.

"Microgrant" carries no standard dollar value. One local-government programme in Wandsworth defines the band as £100–£500, awarded to charities, schools, and resident groups for neighbourhood projects; a 2024 rural health-promotion evaluation reports grants of roughly $4,890–$5,000 disbursed to community-based organisations. The size of a microgrant is therefore not defined by a threshold; it is defined by the programme's design, the recipient's capacity, and the activity being funded. A £300 grant to a school parents' association for rainwater harvesting is structurally the same instrument as a $5,000 grant to a district non-governmental organisation running a sanitation campaign. Both transfer restricted, project-tied resources to a local actor.

"Direct aid" is even more elastic. In one framing it means unrestricted cash to a household, in another it means conditional cash tied to school attendance or clinic visits, in another it means vouchers redeemable at approved merchants, and in another it means the in-kind distribution of food, materials, or services. The World Bank and Innovations for Poverty Action evidence review published on 1 March 2024 grouped cash, in-kind, and vouchers as distinct transfer modalities and concluded that cash generally appears more effective than the alternatives for reducing monetary poverty, improving health outcomes, and strengthening food security, while cautioning that no modality dominates across every context.

Micro-granting buys organised activity; direct aid buys household choice. The two instruments build different kinds of social capital and should not be treated as substitutes.

The shared feature across these definitions is the question of decision rights: who decides what the money does. In a microgrant, the decision is made by a local group through a project proposal. In direct aid, the decision is made by the household receiving the cash, voucher, or goods.

The mechanics of micro-granting: capacity, friction, and practitioner support

Micro-granting is a funding instrument, not a development outcome. The mechanics vary, but three features recur in the better-documented programmes.

First, the recipient is usually an organised local entity rather than an individual. Schools, parent-teacher associations, women's self-help groups, neighbourhood welfare committees, and small registered charities are typical awardees. The Wandsworth case study records 60 such recipients in a single year, with grant sizes between £100 and £500 and the council acting as administrator. The local entity is the unit of accountability: it signs the grant agreement, holds the funds, and reports on outputs.

Second, the grant is usually tied to a defined activity. Funds are released against a project plan, occasionally in tranches, and accompanied by a reporting requirement. This converts the disbursement into a performance contract: the recipient commits to deliver a specific output, the funder retains visibility, and the local group acquires a small amount of project-management experience in the process. Over time, this is how the instrument is supposed to build durable capacity.

Third, and this is the implementation finding that practitioners should weight heavily, microgrants appear to perform best when paired with hands-on support. The Wandsworth local-government case study documents that pairing grant funding with practitioner guidance from application through delivery reduced barriers to participation, improved project quality, and strengthened the recipient organisation's evaluation capacity. Standalone grants, without that scaffolding, frequently produced weaker results. This is implementation learning from a single programme rather than a universal causal rule, but it is consistent with what programme auditors see on the ground: money without management support tends to under-deliver.

The administrative cost of micro-granting is non-trivial. Every award needs an application, a review, a disbursement, and a close-out report. At grant sizes of £100–£500, transaction costs can exceed the grant itself unless the programme is consolidated under a single administrator handling multiple awards. This is one reason many successful programmes cluster around schools, district NGOs, or municipal offices — entities with the institutional overhead already in place to absorb the paperwork.

Direct aid dynamics: cash, multipliers, and what the evidence actually shows

Direct aid is the cleaner instrument to evaluate because the disbursement unit is the household and the outcome metric is household welfare. The evidence base here is meaningfully larger than for microgrants.

The headline finding from the World Bank / IPA review is that cash transfers outperform in-kind transfers and vouchers on the welfare dimensions most programmes care about — monetary poverty, health utilisation, nutrition, and food security — across the studies reviewed. The same review is explicit that this is a general pattern, not a universal rule: study contexts, objectives, and designs vary too widely to support a blanket claim that cash dominates in every setting.

The strongest experimental evidence on cash at scale comes from the rural Kenya general-equilibrium study published on 19 December 2019. Researchers delivered one-time transfers of about US$1,000 to more than 10,500 poor households across 653 villages. The study reported positive spillovers to households and firms that did not receive the transfer, minimal local price inflation, and a local fiscal multiplier of 2.7. That figure — every dollar transferred generating an estimated 2.7 dollars in local economic activity — is specific to this one-time transfer experiment in rural Kenya. It should not be projected onto every welfare programme. It is, however, the most defensible quantitative anchor available when comparing the economic reach of cash against the slower, more localised reach of community grants.

More recent experimental work continues to test the design parameters. A 2026 randomised study in Burkina Faso, covering 225 villages, compared three arms: cash alone, cash plus information, and cash plus information plus home visits. The factorial design isolates the marginal value of layering behavioural and home-visit support on top of a cash transfer — directly relevant for any organisation considering whether direct aid alone is sufficient or whether it should be bundled with case management.

The administrative profile of direct aid is the inverse of microgranting. Disbursement is fast — often within days of beneficiary identification. Reporting is light at the household level, but the front-end burden of identifying recipients, running identity checks, and meeting safeguarding standards can be substantial. Cash programmes also carry reputational and political risk that grant programmes largely avoid: a household receiving cash is harder to defend in public communication than a school receiving materials for a repaired roof.

Navigating the regulatory landscape for individual and group support

The choice between micro-granting and direct aid is shaped, sometimes decisively, by compliance obligations. For organisations operating in the United States under tax-exempt status, two Internal Revenue Service frames apply.

For public charities, the operative test for direct assistance to individuals is the charitable-class doctrine: the beneficiary population must be large enough that aid recipients cannot be individually identified, or sufficiently indefinite that the community as a whole benefits rather than a pre-selected list. A disaster-relief fund for "households affected by the 2024 cyclone" qualifies; a fund for "the named families on this spreadsheet" does not.

For private foundations, the rules are tighter. Grants to individuals for travel, study, or similar purposes are taxable expenditures unless the foundation operates an approved grant programme with objective selection criteria and has secured advance approval from the IRS for the specific purpose. This advance-approval rule applies to private foundation grant programmes for individuals and is not a general requirement attaching to every direct payment by every nonprofit. It is, however, the rule that most often surprises organisations that have shifted from public-charity to private-foundation status without re-mapping their disbursement practices.

Outside the United States, the regulatory profile changes. In India, for example, the Foreign Contribution (Regulation) Act governs cross-border welfare funding and imposes registration, reporting, and end-use conditions that apply to both grant and cash programmes. In the European Union, data-protection rules affect how beneficiary data is collected for any direct-aid programme. The unifying point: the disbursement mechanism determines the compliance work, and a strategy that ignores compliance is not a strategy, it is a liability.

Side-by-side comparison

The differences can be made operational by lining the two instruments against the parameters that matter to a funder deciding where to place the next rupee, dollar, or pound.

The table is descriptive, not normative. Each row names a parameter on which the two instruments diverge; the right-hand column does not imply that direct aid is the correct column to choose.

Strategic alignment: matching funding models to community needs

The two instruments are not mutually exclusive. The strongest welfare portfolios use both, sequenced against the problem they are trying to solve.

Direct aid is the appropriate first response where the bottleneck is household consumption: a flood has wiped out grain stocks, a lockdown has eliminated daily-wage income, a healthcare cost has pushed a family below the survival line. The Kenya evidence suggests cash handles this kind of acute shock with measurable reach and measurable spillovers, particularly when the local market can absorb the spending without price distortion.

Micro-granting is the appropriate response where the bottleneck is organised local activity: a school roof needs fixing, a village water committee needs seed funding for a piped system, a women's group needs working capital for a sanitation-products enterprise. In each case, the work needs an entity to lead it, a project plan to bound it, and a reporting line to verify it. Direct cash to households would not produce any of these outputs; the grant instrument does.

The strongest welfare portfolios do not pick a side. They sequence the two instruments against the specific bottleneck each was built to relieve.

Several decision rules follow for practitioners — whether the funder is a volunteer biker network rebuilding a rural school, a district NGO running a health campaign, or a diaspora collective remitting into a community trust.

1. Use direct aid when the target outcome is measurable at the household level, the time horizon is short, and the local market can absorb the spending without distortion.

2. Use micro-granting when the target outcome is a community-level asset or service, the recipient can be organised into an accountable entity, and the funder can absorb the higher administrative cost per dollar deployed.

3. Use both when the problem has a household component and a community component — for instance, post-flood recovery that needs immediate cash for households and a community grant to rebuild the school that anchors the village.

4. Avoid direct aid as a substitute for organisational funding. Cash to households does not produce a repaired roof; it produces consumption, which is its own good. A welfare strategy that funds household needs but starves local institutions will see those institutions collapse under the next shock.

5. Avoid micro-granting as a substitute for emergency relief. A £500 grant to a school committee does not feed a family tonight. Programmes that blur the line between grant and relief spend their grant overhead on ineligible activities and underperform on both axes.

Closing position

For any grassroots welfare organisation, the choice between micro-granting and direct aid is not ideological. It is an allocation problem with measurable trade-offs. The evidence supports direct cash for household-level welfare outcomes; the evidence on microgrants is thinner but consistently points to a different use case: building the organised local capacity that produces community assets and runs future interventions. The highest-yield portfolios use both, segmented by the bottleneck they are addressing, sequenced against a timeline the funder can actually govern, and audited against outcomes the recipient can be held to.

Funding decisions made on this basis will not always be popular. They will, however, survive the next audit, the next funder review, and the next community asking why the money went where it did.

The strategic question is therefore not which instrument is morally superior. It is which bottleneck is binding, on what timeline, and under what compliance ceiling — and whether the portfolio can hold both instruments in reserve, deployed in the right order, against the right problem.