One common argument against using welfare metrics in impact investing is attribution.
If multiple actors influence outcomes, how can investors claim that their investment produced the welfare improvement?
This concern comes from the traditional evaluation model of attribution analysis, which attempts to establish a direct causal link between an intervention and an outcome, usually by constructing a counterfactual.
Methods like randomized controlled trials (RCTs) or quasi-experimental designs are often used for this purpose.
But many impact investments operate in complex environments like health systems, energy access markets, financial inclusion ecosystems, where outcomes are shaped by many actors simultaneously.
In these contexts, isolating one actor’s causal share can be extremely difficult.
What to do instead? Contribution analysis.
Where attribution analysis asks: Did this intervention cause the outcome?
Contribution analysis asks: How did this intervention help produce the outcome alongside other actors?
Instead of isolating a single cause, contribution analysis builds a credible evidence narrative about how an intervention contributed to observed outcomes within a broader system.
Contribution analysis is also relatively cheaper to do. RCTs/experimental studies are resource intensive, in terms of time, money, and personnel.
These are resources a lot of social enterprises cannot afford. Because of this, they use outputs as proxies for impact evidence.
