Impact investing grows up: from intentionality to additionality (Part 2/5)

Falko and Harald discuss the current state of impact investing: where we are, why we’re here, and where we need to go. They draw on recommendations that commentators made as far back as 2012, and introduce new frameworks that help investors who are looking for “additional impact” to identify the most effective tools. This is part two in a series of five articles. You will find part one here, part three here, part four here and part five here.

In part one Falko and Harald argue that a shift in focus is needed in impact investing: from the strong emphasis on intentionality and identity – which requires asking: “can this impact investment be attributed to me?” – to an emphasis on additionality and systems – which requires asking “how do we ensure adequate funding of needed solutions to global problems?” And: “how do we effect the system change that is often needed for this?” In this article they analyse the concept of additionality and explore how impact measurement can help pinpoint it. For this series they lean heavily on a series of articles published by Stanford Social Innovation Review, that are referenced throughout the text, and four academic papers co-authored by Falko. These sources are also listed at the bottom of this article.

Part 2 – Additionality & measurement

We start here by going back to the basics: why are we doing this? What’s the point? Starr says, “All talk of double- and triple-bottom lines aside, there really is only one bottom line. It’s either impact or profit.”

We agree: the point, the main point, is to have impact through investing. Clearly we also want to generate a financial return, but that applies to any investment – that’s like saying an electric vehicle should have four wheels. Any car has four wheels; what makes an electric vehicle different from another is that it has no internal combustion engine. What sets ‘impact’ apart from ‘mainstream’ investments? It is that the investor wants to (1) cause (2) change (3) that wouldn’t occur anyway. We’ll discuss these three elements, captured in our CC2 concept, one by one.

1. Cause (“C”)

A number of impact commentators argued in their essay Impact investments: a call for (re)orientation: “for investment to be impact-generating, there must be an apparent causal effect on an outcome that can be attributed to the underlying investment.”

We agree (full disclosure: Falko is one of the authors of this paper) and feel causal effect should not simply be ignored, as it often is today. To use another analogy: a child sitting in the backseat of a bicycle is not the one making it move forward – that’s the parent. He’s along for the ride, and will arrive at the destination, but did not cause this to happen: he didn’t make the bicycle ride, make it ride faster, or make it go in the right direction. If we want to know how we reached the destination, we shouldn’t argue the child had anything to do with it. Or simply assume it, just because the bike arrived with the child on it.

If Harald buys shares in the listed bednet manufacturer Bug-Begone from Falko and this company’s bednets help reduce malaria, then Harald’s investment did not cause this change – the money went to Falko. Harald can feel good about being on the backseat while Bug-Begone did its magic, but shouldn’t think his investment caused it.

2. Change (“C”)

Different impact investors have different goals, many go by the Sustainable Development Goals (SDGs) as framework but, regardless of the type of impact, we argue that the point for any impact investor is to see change as a result of the investment. People disagree on what kind of change qualifies – some see nuclear energy as a social good, and so investments in nuclear technologies as impactful; others see it as harmful and therefore nuclear investments to be avoided. We don’t want to arbitrate which kind of change qualifies as impact, just want to state there should be a desired change.

3. That Wouldn’t Occur Otherwise (“TWOO”, or “2”)

We want to know if the change would have also happened without our investment. If it would have, we might still derive warm glow from it – we invested in a project, there was change, we can feel good about being associated with it. But if it would have happened anyway it would be silly to argue our investment had anything to do with it. If we invest in Bug-Begone and malaria rates decline but it turns out this is fully attributable to a vaccine and would have occurred regardless of our bed nets, we’d have reasons to assume the change would also have occurred if we had not made the investment.

You’d think that investors investing for impact want to make sure every one of these CC2 elements is satisfied, but sadly this isn’t the case. Brest said, in 2013: “assessing a particular investment’s additionality in order to determine its investment impact is a novel task that, so far as we know, has not previously been undertaken.” And he says that “a 2010 survey of philanthropists and impact investors suggests that the vast majority are not willing to make any effort to gain information about the actual social or environmental impact of their investments.” 

Starr said, in 2012, “enterprises place surprising little emphasis on assuring real impact. Too often, the case for impact is pretty sketchy… Real impact measurement is a drag on the financial bottom line and investors are usually willing to assume it’s there, so few feel compelled to do it.”

We agree with this but also point out that capturing these elements is very difficult. Which is perhaps why, speaking in the terms of our CC2 framework – Cause, Change, That Wouldn’t Otherwise Occur – the impact measurement efforts we have seen since Brest and Starr’s comments have focused on capturing Change and not the other two, more important elements. 

But what these efforts have obscured is that these elements – Cause and That Wouldn’t Otherwise Occur – have been, not just neglected, but roundly ignored. We would argue that there is no impact measurement methodology that gives an indication of the causal relationship between an investment and a change; and none that gives an indication of the degree of additionality, of whether a change would have occurred even without the investment.

This is also because these dynamics cannot be captured in simple metrics. In the book Radical Uncertainty, Mervyn King and John Kay say: “The public sector is awash with models and metrics, and the mantra that what can be counted counts – and hence that only what is counted counts and what counts can and must be counted – has infected all areas of public and business policy.” This is correct – not everything that counts, that matters, can be counted or measured: it’s unlikely we will design a metric that tells us the degree of causality between our investment in bed nets and malaria decline is 7.4. It’s improbable we’ll invent the “Additionality Rating” – for example, investments with a BB+ Additionality rating have a “reasonable likelihood” of resulting in change that wouldn’t otherwise occur.

Still, as impact investors, we want to get our head around these elements if we want to know whether our investment caused change that wouldn’t have occurred otherwise.

To think how we might do this – if not with quantitative metrics – we turn again to Radical Uncertainty: “Informed judgment will always be required in understanding and interpreting the output of a model and in using it in any large-world situation.” And: “Framing begins by identifying critical factors and assembling relevant data. It involves applying experience of how these factors have interacted in the past, and making an assessment of how they might interact in the future.”

We think this is exactly right. As Alex Edmans at London Business School has said: “sustainability cannot be measured, it should be assessed.” The same applies here: causality and additionality cannot be measured, they should be assessed; argued and debated, using informed judgment, and by understanding the theory of change – how have people dealt with malaria in the past? What do we know about how successful different approaches have been in tackling poverty? What does the last decade of grappling with diversity tell us about which interventions work, which we should invest in? 

Starr says, “In the end the key to figuring out real impact is honest, curious, and constructive skepticism.”

We think Starr is right. What’s more, because we can often establish which interventions were effective in driving change only many years after the fact, sometimes not at all, we should be thinking about these things more ex-ante (before the investment) rather than ex-ante (after the investment). Difficult? Yes. 

But no one said this would be easy.

Before wrapping up this article in the series we talk about two implications of this thinking: one for labels, and one for monetising impact.

Labels 

Our thinking raises the question “can we come up with a label that captures CC2”? The answer is probably “no”. If we assume that impact investors, including consumers who want to invest in impact funds, want to know whether their investment Causes Change That Wouldn’t Otherwise Occur, we need to come up with a label that will capture not only the Change (that we can measure, even if we should get better at it), we need to come up with labels to capture Causality, as well as Additionality. As argued above, we will likely not achieve this, at least not quantitatively. Labels might one day incorporate a qualitative assessment, for example: “This investment is very unlikely / somewhat likely / very likely to contribute to eradicating poverty”. And: “Poverty will almost certainly / probably / probably not / almost certainly not be solved without this investment”. Wouldn’t that be nice?

But we imagine that coming up with them would require panels of poverty experts – development economists, political scientists, sociologists etc. – assessing causality and additionality of different interventions. Possible in theory, but what would it take to do this for all 17 SDGs (and dozens of subgoals), and for all interventions we need to meet them? And would the capital this would mobilize, because of the assurance it gives investors, justify the investment in such a massive undertaking? We think unlikely. But we also think that it is much more important that certain needed investments are made than that we establish who they can be attributed to.

Lee & Singh conducted a study which is illustrative here: “A large fraction of the participants—between one-third and two-thirds – failed to select an outcome-efficient allocation. This pattern held even for participants who were well educated and financially savvy. Why did these individuals fail? It turns out that a critical factor is categorical cognition—a tendency to use simple categories, rather than systematic calculation, to evaluate options. We tested this by giving a random set of participants the same impact investment exercise, but with the options presented without the common categorical labels of for-profit, charity, and social enterprise. By removing these labels, this intervention removed the distraction of categories, allowing people to pay more attention to the actual outcomes. With categorical labels removed, participants failed to choose an outcome-efficient allocation far less frequently; in some versions of our study, the failure rate dropped by nearly half.”

We conclude that, not only is coming up with labels that capture CC2 extremely difficult and costly, it may also be counter-productive – we may not actually need them to drive CC2-type impact.

Monetising impact

The other question is about monetising impact. It has been suggested that putting dollar values on the realised impact might help in further building the impact investing community – the investor could simply assess investments on a promised financial return of, say, 10% IRR and an impact return of $1,718.

We believe this is wishful thinking. First, putting numbers on the Change – that is, the second C in our CC2 framework – says nothing about the degree to which the investment has caused that change or whether the change would have occurred anyway. Second, we think that the types of changes that count as impact – basically any change that contributes to an SDG – are so varied that we will likely never come up with a simple metric that allows them to be compared using dollar amounts. For example, “a case of malaria prevented” equals “6 years of primary education” equals “20 years lived without hunger” equals “400 square feet of coral reefs protected” equals $24,850. Third, while it may be possible to put a dollar amount on some of these impactful outcomes, e.g. carbon pulled out of the air, with most other outcomes that we would like to see this will be impossible. Pricing these things in a way that sees broad consensus assumes there will be markets where these things can be traded, and there is no market for “one year lived without hunger”, or for “gender equality in a medium-sized country” or for “ten years of peace in Tanzania”. 

Brest & Born said, “Although we have no a priori commitment to any particular depth of analysis, we believe that realizing the promise of impact investing depends on all three measures [i.e. assessing financial return, assessing social return, assessing additionality] becoming central to the marketplace.

We would agree – this is in line with our CC2 thinking – but also argue that a significant shift in thinking and effort is needed to get away from the quantitatively oriented methodologies focused on assessing social return, and to get closer to more qualitative assessments that also include the elements of causality and additionality.

In the next article in the series we will discuss the degree to which it is possible to find impact investments that also offer market-rate returns and will argue, using the “SDG Venn diagram” that the number of impact investment opportunities in secondary markets involving listed companies is close to zero.

Our arguments lean heavily on those of a number of commentators who – in a set of articles published by the Stanford Social Innovation Review (SSIR) – have started making similar observations as far back as 2012, but that we feel have been overlooked. We want to bring them back to the current debate and they are therefore referenced throughout our articles:

• Sectors, Not Just Firms; Do No Harm: Subsidies and Impact Investing and Government Matters (2012), By Matt Bannick & Paula Goldman (“Bannick & Goldman”)

• The Trouble With Impact Investing: P1 / P2 / P3 (2012); Kevin Starr (Part 2 with Laura Hattendorf (“Starr”)

• When can impact investing create real impact / Unpacking the Impact in Impact Investing (2013); a long and shorter version of the same article, Paul Brest & Kelly Born (“Brest & Born”)

• How Investors Can (and Can't) Create Social Value (2016); Paul Brest, Ronald Gilson and Mark  Wolfson (“Brest, Gilson and Wolfson”)

• How to Overcome ‘Warm Glow’ and Other Barriers to Effective Impact Investment Decisions (2020); Matthew Lee & Jasjit Singh (“Lee & Singh”)

• Impact Investing Can’t Deliver by Chasing Market Returns (2023); Jim Bildner (“Bildner”)

We picked these from a much larger universe of articles on impact investing because they best encapsulate how to think about its moving parts, but also to show that smart advice on impact investment has been available from the early days, and that we can be more effective by heeding it.

This article is also based on insights from four papers that Falko co-authored together with other academics in recent years:

• Between impact and returns: Private investors and the sustainable development goals (2022); with Timo Busch, Sebastian Utz, Anne Kellers

• Do Investors Care About Impact? (2021); with Florian Heeb, Julian Kölbel, Stefan Zeisberger

• Unlocking the black box of private impact investors (2021); with Sarah Louise Carroux, Timo Busch 

• Wealthy Private Investors and Socially Responsible Investing: The Influence of Reference Groups (2021); with David Risi, Anne Kellers

Finally, in writing this series, we have greatly benefited from conversations with Paul Brest, James Gifford, Robert Boogaard and Jonathan Harris.

illuminem Voices is a democratic space presenting the thoughts and opinions of leading Sustainability & Energy writers, their opinions do not necessarily represent those of illuminem.