The Counterfactual Test
Rees Calder · 2 May 2025 · 6 min read
The question nobody asks
You gave £500 to a homeless shelter last winter. Beds were filled. People stayed warm. You feel good about it, and you should.
But here is the uncomfortable question: would those beds have been filled anyway?
If the shelter was going to hit capacity regardless, funded by the council and three corporate sponsors, your £500 didn't create a warm bed. It freed up £500 of someone else's budget. That is a very different claim from "I housed someone."
This is the counterfactual test. Not "did good things happen after I gave?" but "did my giving cause something that would not have happened otherwise?"
Why it matters for your tally
Most giving dashboards show inputs. You gave X pounds to Y charity. But impact is not inputs. Impact is the difference between the world where you gave and the world where you didn't.
Economists call this additionality. Development researchers call it the counterfactual. Whatever the label, it asks: what changed because of you, specifically?
Three scenarios where counterfactual impact is high:
- Funding gaps: A programme needs £50,000 to launch. It has £48,000. Your £2,000 is the difference between existing and not existing. Your counterfactual impact is the entire programme.
- Matching thresholds: A funder will unlock £100,000 if a charity raises £20,000 from individuals. Your contribution toward that £20,000 has 5x leverage.
- Neglected causes: When few donors fund an area, each marginal pound does more. GiveWell estimates that top charities in neglected health interventions deliver 10-100x the impact per pound of well-funded causes.
Three scenarios where it is low:
- Oversubscribed appeals: The charity already hit its target. Your money goes to reserves or gets redirected.
- Government-matched spending: If a government commits to matching charity spending 1:1, your donation may simply displace tax revenue that would have been spent anyway.
- Popular disasters: After high-profile emergencies, donations often exceed what can be usefully deployed. The 2004 tsunami raised so much that some charities asked people to stop giving.
How to apply the test
You do not need a PhD in econometrics. Three questions get you most of the way:
1. Is this charity funding-constrained?
Check their annual report. If they have more than 12 months of reserves and no stated expansion plans, your marginal pound is likely less impactful. If they report turning away beneficiaries due to budget, your pound matters more.
2. Am I filling a gap or joining a crowd?
Ask the charity directly: "If I give £X, what specifically happens that wouldn't happen without it?" Good charities can answer this. They know their funding gaps. If the answer is vague, the counterfactual is probably vague too.
3. Could this money be redirected?
Sometimes the best counterfactual impact comes from giving to something nobody else will fund. Effective Altruism researchers estimate that the most neglected global health interventions are 100x more cost-effective than the average UK charity spend. Not because average charities are bad, but because the neglected ones have almost no other funders.
Updating your tally
If you track your giving, add a column: counterfactual confidence. Rate each donation High, Medium, or Low.
- High: You know the charity is funding-constrained. You can point to what your money specifically enabled. The cause is neglected.
- Medium: The charity is well-run and probably funding-constrained, but you cannot pinpoint exactly what your donation did versus what other funders would have covered.
- Low: The charity is well-funded, the appeal was oversubscribed, or the government would have stepped in regardless.
This is not about guilt. A Low counterfactual donation to a local food bank still matters socially, still builds community, still signals values. But if you want to maximise the actual difference you make in the world, the test tells you where to point your next pound.
The uncomfortable implication
Most people's giving tallies look impressive on inputs and modest on counterfactual impact. That is not a failure. It is information. The gap between "money I gave" and "change I caused" is where the opportunity lives.
Your next donation could close that gap. The question is whether you want to ask it.