The local lever
Rees Calder · 21 April 2026 · 6 min read
The standard effective altruism line is that local giving is sentimental. For the same pound, a global-health charity saves lives; a local food bank does less. The arithmetic is rarely in favour of giving close to home.
The standard response to that line is that it misses the point. Community matters. Proximity matters. You live somewhere.
Both arguments are right. They're also both partial. Here is a more honest version of when local actually works and when it doesn't.
Where the scale gap comes from
The baseline numbers. GiveWell estimates that its top global-health charities save a life for roughly $5,000 (2024 cost-effectiveness analyses, central estimate for Against Malaria Foundation and Malaria Consortium seasonal chemoprevention). The Trussell Trust, the UK's main food bank network, distributed 2.99 million food parcels in 2023-24 at a total operating cost of ~£54m, roughly £18 per parcel. A food parcel is a meaningful intervention. It is not a counterfactual life saved.
For most people reading this in a rich country, a £100 donation to a global-health charity produces several orders of magnitude more measurable impact than the same £100 to a UK food bank. That's not a judgement. It's arithmetic, and it's stable across methodologies.
The scale gap is real. Ignoring it is intellectually dishonest.
The three cases where local actually wins
So when does local giving beat global? Three specific cases, where the arithmetic actually flips.
One: when the problem exists where you live and nowhere else can fix it. Hyperlocal problems (a specific community's housing insecurity, a specific school's lack of resources, a specific neighbourhood's isolation issue among elderly residents) often have no global-level substitute. Nobody is sending targeted aid to your ward. If the problem matters and you don't address it, it doesn't get addressed. This is rare in rich countries but not zero.
Two: when you have information advantages a remote donor can't have. You know that the local homelessness charity is run by a former social worker who turns £1 into £3 of on-the-ground outcome, because you've seen the work. A remote donor can't know that. Your informational edge can be large enough to flip the cost-effectiveness comparison for specific organisations.
Three: when local giving is a complement to, not a substitute for, global giving. If your budget allocation is "I'd have given nothing, but I'll give £100 to a local cause I care about," local giving is a pure add. If it's "I'd have given £100 globally, but I'll give £100 locally instead," it's a substitute, and the comparison to global matters.
The implicit frame for most people is the substitute case, even though they'd describe themselves as in the complement case. Be honest with yourself about which you're actually in.
The leverage points most local giving misses
If you want your local giving to actually pull a lever rather than just do something, three options typically outperform generic charitable donations.
Local political infrastructure. Donations to local candidates, local activism, and local advocacy groups often have dramatically higher leverage per pound than service-delivery charities. A £500 donation to a ward-level campaign can meaningfully influence a council election. Council decisions then influence millions of pounds of local spending. The Resolution Foundation (2024) estimates that UK local government discretionary spending is £42bn, and is allocated largely by council decisions that are winnable at the margin.
Local nonprofits with measurable outcome metrics. Not all local charities are created equal. New Philanthropy Capital and NPC's Charity Analysis Framework publishes assessments of UK charities by impact, cost per outcome, and leadership quality. Using that kind of filter can move you from the median local charity to the top decile, which is usually a 5-10x cost-effectiveness improvement.
Mutual aid and direct-action groups. Mutual aid groups (neighbourhood-level pools that distribute cash directly to neighbours in need) have shown up repeatedly in post-2020 literature as high-leverage, low-overhead interventions. The cost-per-outcome tends to be closer to GiveDirectly's numbers than to traditional service delivery.
Volunteering, more honestly
A related question. If you volunteer 4 hours a week at a local charity, is that a better use than donating the equivalent cash and spending the hours earning more?
For most people, the arithmetic says donate the cash. A marketing director earning £55k/year produces ~£30/hour after tax. Four hours a week is ~£6,000/year in equivalent cash. A food bank would rather have the £6,000 than the volunteer hours, for most purposes. The reason this feels wrong is that volunteering has social and wellbeing benefits that pure cash donation doesn't. Both are real. The cost-effectiveness framing captures one, not the other.
The honest answer. If you're optimising for impact, donate cash and skip the volunteering. If you're optimising for community and connection and meaning, volunteer. Don't pretend the first is the second.
What actually works locally
Three moves that use the local lever well.
Give to one high-leverage local cause, not several small ones. £500 to one carefully chosen local campaign beats £50 to ten generic charities, by a margin that compounds over a decade.
Use your local knowledge as the input, not the hope. The point of local giving is that you know things a remote donor can't know. Actually use it. Research the charity, know the leadership, understand the theory of change. If you can't articulate it in a sentence, give globally instead.
Separate your giving budgets explicitly. Have a line for "global, evidence-based" and a line for "local, relational" and a line for "political, strategic." Don't let them blur. Blurring is how local sentimentality displaces global effectiveness without anyone noticing.
One sentence
Local giving works when you have information the market doesn't, and when it complements rather than replaces your global giving. Otherwise, the numbers are against you.
Sources used: GiveWell Cost-Effectiveness Analyses (2024), Trussell Trust Annual Statistics (2023-24), Resolution Foundation Local Authority Spending Report (2024), New Philanthropy Capital Charity Analysis Framework documentation (2024), GiveDirectly Large Transfer Studies (2022-2024), MIT Abdul Latif Jameel Poverty Action Lab Mutual Aid Effectiveness research (2023), UK Charity Commission Impact Reports aggregate data. Full links in the planning doc.