British International Investment

Do investors care about impact?

This blog is authored by our Director of Research and Policy, Paddy Carter. These are personal views and do not necessarily represent the views of British International Investment.


The problem at the heart of impact investing is that impact is not observable. One of the most important tasks of the teams who manage impact within development finance institutions and impact investors is to find ways of making impact visible. We need to observe impact so that investment professionals can recognise which investments deliver it, and the organization can set about maximising it.

I am exaggerating a little – it would be more accurate to say that impact is not easily observable. Sometimes, with a dedicated evaluation effort, it is possible to estimate how much difference an investment has made to people’s lives. Usually, however, all investors have to hand are various impact “metrics”. These are usually outputs, such as electricity generated, or numbers of workers employed. What we need is some way to make sense of these metrics.

That’s why impact scoring tools, such as ours, are so important. Even if they do not really measure impact in way that would allow a comparison across different investors (who use different scoring tools), within investors these tools identify which investments the institution regards as more or less impactful. It is impossible to even think about constructing a portfolio to maximise impact without having some way of arriving at relative judgements about the magnitude of impact.

With that in mind, I was alarmed to read about a recent experiment that explored how much impact-oriented investors are willing to pay for impact.[1] The experiment exploited one of the rare cases in which the magnitude of impact is quantifiable – carbon dioxide emission reductions. The experiment did not involve genuine professional asset allocation decisions, but there was money at stake (a $1000 investment corresponding to the chosen investment was made in a random subset of cases) and the subjects were experienced professional investors. Each investor was offered a choice between two options, presented as equity funds: a sustainable investment offering some level of carbon dioxide abatement, and a conventional investment offering none. The level of carbon dioxide abatement was varied across subjects. Willingness to pay was elicited by varying a front-end fee that the investors must pay for the sustainable investment.

The researchers found that investors are willing to pay a substantial amount for a sustainable investment with some impact, but that willingness to pay does not differ significantly between an investment that saves 0.5 tons of carbon dioxide emissions and one that saves five tons. The authors investigate a number of possible explanations and conclude that investors are motivated by the “warm glow” of doing something impactful, without regard for magnitude.

That is not encouraging from the point of view of optimal portfolio construction, and it also imperils one theory about why impact investing has an impact: by reducing the cost of capital more for more impactful enterprises.

I thought of two reasons why impact investors might improve on this disappointing performance. The first is simply that if impact investing decisions are made more transparent, the people giving money to asset mangers to make impactful investments might start shouting at them, if they can see they are ignoring the magnitude of impact. The second is that investors might perform better if impact is expressed in a unit that is even easier to understand and compare to financial returns: dollars.

That latter idea bolsters the case for impact monetisation, but it does not make it any easier. Many impact investors think the arrival of corporate accounts that report impact measured in dollars is only a matter of “when”, and not “if”. George Serafeim of Harvard Business School is one of the leading researchers in that field – he sets out a framework of impact weighted accounts in this paper with Katie Trinh.

I fear – rather like self-driving cars, which were once assumed to be around the corner – that is going to prove to be easier said than done. I think some of the more important dimensions of impact are too hard to measure, and the counterfactual which defines what difference an enterprise is making to people (which is the core concept of impact) is too hard to define. One illustrative example of these difficulties is how to think about market power. Economists are increasingly concerned that fewer and more powerful firms are holding down wages, holding up prices and reducing business dynamism, to most people’s detriment. How would a firm monetise that sort of impact?[2] That might sound esoteric, but how credible are reported dollars of positive impact going to be if there are good reasons to believe a business could be having a negative impact?

It would be easier to focus solely on monetising the impact felt by consumers of goods and services. That could be done by eliciting “willingness to pay” and comparing that to how much people actually are paying. The problem here is adjusting ability to pay. We would not want to say that a clean cookstove that low-income people in Africa would pay $12 for has less impact than a solar swimming pool heater that Californians would pay $1200 for. And even without that problem, willingness to pay is not obviously what impact investors should care about. That figure of $12 for a cookstove comes from a recent publication in which researchers found that an improved charcoal cookstove saved people $237 over its two-year lifespan and resulted in a reduction in carbon emissions that they valued at $295, and yet on average households were only willing to pay $12 for the stove.[3] Which of those numbers best represents impact?

So I have more hope for “impact investors will get better if they get shouted at” idea, than the “impact investors will get better once impact is monetised” idea.

One final thought. How much does inefficient allocation by impact investors matter? I guess the cost has to do with how the sum of money to be invested compares to the quantity of investment opportunities. If money is scarce relative to investment opportunities, an indiscriminate approach to impact would result in low impact investments being chosen at the expense of high impact investments. It matters less if money is plentiful relative to opportunities. In that case, all the high impact investments on offer would probably be made anyway. I know that BII feels the need to screen investments for impact because we have many more investment opportunities than money, but I don’t know about the impact investing industry more generally.

[1] Florian Heeb, Julian F Kölbel, Falko Paetzold, Stefan Zeisberger, Do Investors Care about Impact?, The Review of Financial Studies, 2022;

[2] See the book “The Great Reversal” by Thomas Phillipon for example

[3] Berkouwer and Dean (2022). “Credit, Attention, and Externalities in the Adoption of Energy Efficient Technologies by Low-Income Households.” American Economic Review.

Subscribe to receive our latest news and updates