I’ve made a series of posts on the potential uses of welfare-adjusted metrics in impact investing.
Now, a bit of devil’s advocate. Are there cons/drawbacks to doing so? Of course there are:
1. Measurement uncertainty can mislead decisions
Welfare metrics often rely on modelling assumptions, generalized evidence, and imperfect data. When translated into #investment decisions, this can create a false sense of precision. For example, a model might suggest one investment is “more impactful,” when in reality the difference is driven by assumptions. This can be risky in capital allocation.
2. Monetising welfare is ethically and methodologically contested.
Converting DALYs into money requires assigning a value to human life or health. This often involves metrics like the Value of a Statistical Life (VSL)/Value of a Life Year. Critics argue this can oversimplify complex social outcomes, embed ethical biases, and create uncomfortable comparisons (e.g., whose life is “worth more”)
3. Risk of distorting investment behaviour.
If investors optimise for welfare-adjusted #returns, they may prioritise easily measurable interventions, favour short-term, quantifiable outcomes, and avoid complex or systemic investments.
4. High implementation costs
Robust welfare measurement requires data collection, modelling expertise, and ongoing validation. Unlike outputs, these are not easily standardised. And because most funds are not set up to run quasi–cost-effectiveness analyses across portfolios, this becomes a barrier to adoption.
5. Attribution still matters
Even with contribution analysis approaches, welfare estimates still depend on assumptions about causal pathways and uncertain #counterfactuals. Critics argue that without strong #attribution, welfare metrics risk becoming well-informed guesses rather than decision-grade metrics.
