More than just clean towels
The term ‘greenwashing’ was thought to be first coined in the 1980s, when an undergraduate surfer snuck into a beach resort to steal some clean towels. He was struck by the resort’s plea to guests that reusing the towels would help them to protect the ocean and coral reefs, even as the resort was constructing ever more bungalows on said reefs.
Over time, as interest and activism has grown in not only environmental issues, but also social and governance ones, these types of claims have also broadened out and grown across sectors.
In finance, the past 3 years have seen a dramatic increase in environmental, social and governance (ESG) and sustainable investments (see Figure 1). This has also led to increasing concerns about firms confusing or even misleading consumers about the nature of some of these investments.
Figure 1 European Sustainable Fund Flows (USD Billion)
To examine some of these concerns, we have carried out exploratory analysis to understand how different information about the degree to which a fund is ESG/sustainable may affect consumers’ investment decisions. In particular, we have focused on answering the following questions, based on some hypothetical scenarios:
- What ESG/sustainability information about a fund sways how consumers say they will invest?
- What is the effect of greenwashing on how consumers say they will invest? More specifically, what is the effect of implying a fund is ESG/sustainable when a grading presented as more objective says that the fund is not?
Comparing features with conjoint analysis
To answer these questions, we designed an online conjoint analysis. This is a technique that asks participants to view products that have different features and make a hypothetical choice between them, allowing us to infer how influential each feature is for their choice.
We ask 1530 participants to choose between 8 pairs of hypothetical fund factsheets. These factsheets are designed to appear as simplified versions of real fund factsheets consumers might encounter. We include the most essential pieces of information (such as risk or expected return), and items that we want to test the effect of explicitly (such as the fund’s image, the fund’s description, and fund’s strategy – for an example see Figure 2 below). For each pair, they can say they would choose to invest in one fund, the other, or not to invest in either.
For around half the group (744 randomly assigned participants), the hypothetical, simplified factsheets consist of 5 different attributes, so we can hone in on the effects of these. These are fund image, fund description, fund strategy, rate of return, and risk level.
Each attribute can take one of a number of options. Participants see a random combination of these options in each of their factsheets. For fund image, fund description, and fund strategy, half of all the random options that participants see presents the fund as ESG/sustainable (for example, described as an 'ESG' fund or a 'sustainable' fund), and the other half as ESG neutral (for example, described as a 'growth' or 'global' fund in the fund description). Any combinations are possible, meaning there may not be consistency between ESG and neutral (non-ESG) attributes. So, a fund description could present the fund as ESG, but the fund image may not do.
The remaining (786) participants are shown similar pairs of factsheets, but their version includes a sixth attribute – a randomly assigned ESG grading presented as a ‘medal’, that is designed to appear as a reliable, impartial, objective and salient grading of a fund’s sustainability orientation. These ‘medals’ range from a black medal ('No positive impact') to a gold medal ('Impact leader'). See Figure 3 below for details on the medals. While ESG gradings, ratings or awards do exist in reality, they are sometimes used inconsistently, and are not always reliable or objective. We add these medals to help give an indication of what effect such salient and objective gradings might have on consumers’ decisions, as well as also giving us a way to assess greenwashing (more detail below). An example of these factsheets is provided in Figure 2.
 We implemented the conjoint analysis using the oTree platform.
Figure 2. An example of a pair of fund factsheets that participants were asked to choose between with an additional medal attribute
Figure 3. Medals gradings and accompanying text
Objective gradings such as medals sway consumers’ decisions
We then set out trying to answer our first question on what ESG/sustainability information would sway consumers’ decisions.
Using our conjoint analysis design, we first compare the hypothetical investment choices that consumers make when presented with ESG versus neutral attributes (ie ESG versus neutral fund images, fund descriptions and fund strategies), for the group of participants that do not see medals.
We find that ESG fund images, fund descriptions and fund strategies have no statistically significant effect on how participants invested in our analysis setup. Participants appear no more likely to choose funds based on the factsheet having an ESG attribute compared to a neutral one.
We do find that medals have a significant effect on participants’ investment choices in our analysis setup.
We compare funds with each of the positive ESG impact medals (bronze, silver, and gold) with those that have the 'No positive impact' medal. The bronze medal increases the likelihood of participants saying they would invest in the fund by 9%, and the silver and gold medals increase that likelihood by 13% each. This suggests a salient, and objective grading of a fund’s ESG and sustainable credentials would have a significant effect on which funds consumers decide to invest in (although we note we did not test detailed comprehension of what the medals stood for, so consumers may have taken medals more as a sign of general quality, for example).
Figure 4. Effect of positive ESG medals on consumers saying they would invest in our hypothetical funds
Participants did not react to ‘greenwashing’ in our setup
To test our second question on greenwashing, we look at the effect of factsheets that present ESG fund images, fund descriptions, and fund strategies, but where the medal says 'No positive impact'. This best simulates greenwashing in our setup. The left image in Figure 5 below provides an example of this.
We find that participants’ investment choices are not swayed by ESG information that conflicts with the medal’s grading. We find no statistically significant differences in the effect of ESG fund images, fund descriptions and fund strategies on participants’ choices when we compare cases where the medal conflicted with this information and cases where it agreed. This is perhaps unsurprising given our first finding above, where fund images, fund descriptions and fund strategies have no significant effect on the participants who did not see medals on their factsheets.
This could suggest consumers may not always spot greenwashing, but it could also be that they just do not respond to it. Of course, our setup also does not test what happens if the greenwashing is presented as objective information. For example, because the funds are all hypothetical, we cannot test the effect of adding a gold medal to a fund that, in fact, is non-ESG. This is a limitation of our research. Our previous finding on medals suggests that consumers may respond fairly strongly if something is presented as objective information.
Figure 5. An example of funds shown where one fund has been greenwashed and the other has an objectively presented grading of positive ESG impact
A note on our analysis
We use conjoint analysis in this context because we deem it the best tool for generating the types of exploratory (rather than causal) insights set out above, but it is worth noting that its strengths bring some weaknesses. Most importantly, our conjoint design relies on simplified and hypothetical funds and fund information, as well as imagined rather than real investment decisions. While this setup allows us to focus in on the effect of particular pieces of information, it is also not as realistic as it could be (so its ‘external validity’ may be weaker). We have also chosen to recruit a broad panel of participants (who are approximately nationally representative on age, gender and region) because we think ESG and sustainable investments could have a broad-based appeal in the future, but they are unlikely to reflect fully the characteristics and behaviour of current investors who invest their own money.
Within our conjoint analysis, we did spot an issue with profile-order effects. We find that whether hypothetical funds appear on the left or right hand side of the page has some influence on how likely it is to be chosen.
Overall, we still consider our analysis provides helpful insight, but should be interpreted with a lens that considers the issues set out above.
Coming out in the wash
Decades on from when the term ‘greenwashing’ was first coined, policymakers continue to grapple with how best to inform consumers about whether firms actually have the ESG/sustainability effect that they claim.
Our analysis helps to provide some insight on what information on ESG/sustainability might influence how consumers choose to invest. In particular, we find a more objective grading of ESG/sustainability, does have a significant effect on which funds participants say they will invest in.
Our analysis adds to the case for better ESG/sustainable related disclosures and the wider need for an environment in which market participants can manage the risks from moving to a more sustainable economy and capture opportunities to benefit consumers.
 'The troubling evolution of corporate greenwashing'. The Guardian, 20 Aug 2016.
 See Hainmueller, Hopkins, and Yamamoto (2013) for details of the method we followed. We implemented the conjoint analysis using the oTree platform.
 See Hainmueller, Hopkins, and Yamamoto (2013) for details.
 Specifically, we found the image on the right was slightly (and statistically significantly) more likely to be chosen than the image on the left. However, this was not a pre-specified part of our analysis, so care should be taken drawing strong conclusions from this finding.
 In case of interest, see the FCA’s sustainable finance strategy.