How impact is measured when impact investors invest in mission driven enterprises

(1)

Geurtzen, Saskia (2017) 11120843 University of Amsterdam Master of Science, Business Administration Entrepreneurship & Innovation

How impact is measured when impact investors invest

in mission driven enterprises

23rd of June 2017

First reader: Ieva Rozentale Second reader: tbd

(2)

Page 1 of 72

Statement of Originality

This document is written by Student Saskia Geurtzen

who declares to take full responsibility for the

contents of this document.

I declare that the text and the work presented in this

document is original and that no sources other than

those mentioned in the text and its references have

been used in creating it.

The Faculty of Economics and Business is responsible

solely for the supervision of completion of the work,

not for the contents.

(3)

Page 2 of 72

Abstract. This research aims to enrich the existing scientific literature and the field of impact

investing with insights on how impact investors measure the impact created by their social enterprise investee. From the research, it was found that there are in general two ways of measuring impact. The first category measures social results simultaneously with financial results and therefore obtains an incorporated measurement relationship. This way of measuring focuses on increasing the direct impact: the impact the investors have through their enterprises. The other way of measuring has a separate measurement relationship: social results are monitored and measured separately from financial results. This second category shows a focus on increasing both direct and indirect impact (i.e. the impact on beneficial groups or systems). Other important topics found in the field of impact measurement among investors are the quantification of impact results and a financial reward system on both financial and impact result achievements.

Keywords: impact investing, social enterprises, impact measurement, direct impact,

indirect impact, financial results and social results.

“Measurement is the first step that leads to control and eventually to improvement. If you can’t measure something, you can’t understand it. If you can’t understand it, you can’t control it. If you can’t control it, you can’t improve it.” ― H. James Harrington (GoodReads, 2017)

(4)

Page 3 of 72

1. Introduction

‘Tackling poverty with social enterprises.’

This headline was used in the article written by Miller & Teacher (2013) for UK’s biggest newspaper The Guardian. The authors wrote a blog on the effects of social enterprises.

‘In the last 20 years, the Fairtrade1_{movement has done a remarkable job of connecting} developing world farmers into international markets. [..] In sum, the old trope “give a man a fish” barely scratches the surface of the web of interventions necessary to end food insecurity. [..] Organizations that leverage each other’s work are the route to success. It is only through backing a thriving ecosystem of organizations at different levels of the supply chain that [food] poverty will become a problem of the past.’ Miller & Teacher (2013)

Miller & Teacher (2013) are not the first to point out that setting up social enterprises is a sustainable way to solve poverty issues in developing countries. Over the last decades, several organizations have showed their interest in the way social enterprises achieve their societal mission by creating financially sustainable businesses. The most notable development in this field is the rapid expansion of investors seeing a huge (business) potential in supporting the growth of social enterprises.

This field, nowadays called impact investing, has developed itself as a new source of funding for these mission driven initiatives. This type of funding now covers six percent of

(7)

Page 6 of 72

the total venture capital market (Miller et al., 2010) and is becoming a well-established notion in the financial field (Scholtens, 2007).

1.1. Societal relevance

Impact investors make investments with the aim to achieve social and environmental impact as well as financial returns with the enterprises that they have invested in (The GIIN, 2017). The Global Impact Investing Network (The GIIN) adds one other characteristic to this core definition, which is – according to The GIIN – the hallmark of the field of impact investing: having the commitment to measure the societal performance (The GIIN, 2017).

This core definition shows a combination of achieving and measuring on two seemingly contradictory goals: social impact as well as financial returns. In light of the more traditional method, the creation of social impact is attained through charities and donations, without any financial returns. Creating profit by helping to solve poverty issues seems, and is proven, to be consisting of conflicting goals and values. The field of impact investment, and social entrepreneurship in general are therefore filled with a large amount of debates on whether it is possible or not to achieve both financial returns and real social impact (Smith et al., Bannick et al., 2017).

Where the growing amount of successful social enterprises has proven to tackle the difficulties in combining a social mission with financial returns, the social-financial conflict issues have unfortunately not been tackled yet in the measurement process of impact investments. The collection of financial numbers by impact investors seems to be transparent enough since it is based on the widespread experiences from the commercial investment field. But the addition made to the measurement process of measuring societal

(8)

Page 7 of 72

impact as well makes it more difficult (Smith, et al., 2013). This impact measurement process, which is viewed as the hallmark of impact investing, still consists of a lot of questions and different views on what to do to measure societal impact (So & Capanyola, 2016). Several authors have tried to grasp the notion of impact measurement within social enterprises, but not one scholar has studied the dimensions of impact measurement within the impact investment field. These questions, doubts and different views make that in practice diverse ways of measuring impact exist among impact investors (So & Capanyola, 2016). Besides, there is no clarity on what this impact measurement process looks like among different impact investors.

1.2. Scientific relevance

The existing scientific field is still too underdeveloped to answer the raised practical questions on impact measurement. Not much is written about impact investing in general, let alone on how impact measurement is carried out in this field. The few scholars that focus on creating insights in elements of the impact investing field, write about how investors make their investment choice (Miller & Wesley, 2010), the performance of the investments itself (Miller & Wesley, 2010; Scholtens, 2007) and what the concept of impact investing means (Höchstadter & Scheck, 2015). None of these scholars have focused on conducting research on how impact is measured among this group of impact investors, despite the importance of this so-called hallmark in the field of impact investing. This research therefore focuses on shedding a light on how the process of impact measurement is executed and valued by impact investors.

(9)

Page 8 of 72

This makes the main research question of this thesis:

How do impact investors measure the impact created in their collaboration with social enterprises?

To answer this main question, several sub questions have been established.

1. What are the most important themes in the process of measuring impact according to existing scientific literature?

2. What are the most important themes in the process of measuring impact according to the participating impact investors in the research?

3. What patterns can be found in the impact measurement process among different impact investors?

4. What do these patterns tell us about how impact investors measure the impact? 5. What can be concluded from how impact measurement is done by impact investors?

The first question is answered by conducting a literature review, whereas the rest of the questions are answered by conducting and analyzing empirical data of the participating impact investors.

This thesis starts with a discussion of the literature review on relevant scientific articles about impact measurement within in social enterprises. The next chapter explains and accounts for the methodology used to come to certain results. The results are then described with a theme-by-theme approach in the ‘Results’ section. The ‘Discussion’ chapter highlights the patterns found across the themes and cases, to come to a general conclusion

(10)

Page 9 of 72

on the findings in the second-last chapter ‘Conclusion’. This thesis ends with an overview of the limitations and implications of the research.

(11)

Page 10 of 72

2. Literature review

As mentioned earlier, the notion of impact investing is starting to become well-established in the financial field (Scholtens, 2007). The growing interest of financing social initiatives is encouraged by the rise of social enterprises as well. Social enterprises, as businesses that tackle societal issues by using commercial tools (Smith et al., 2013), created the opportunity to achieve financial returns whilst making social impact. Among other things, these enterprises made it attractive for investors to invest in social solutions.

Social enterprises are therefore one of the main stakeholders involved in the field of impact investing. A social enterprise is defined by several authors as an organization that reinvests the profit returns into their business or cause, instead of distributing the profit (Hochstädter, 2015).

Next to the social enterprise, called the demand-side actors by Jackson (2012), other actors as the asset managers (impact investment funds), asset owners (impact capitalists) and service providers (e.g. nongovernmental organizations (NGO)) are involved in impact investments (Jackson, 2012). In many cases, the asset manager is the stakeholder who is managing the investment and the maintenance of the relationship with the demand-side actor (social enterprise). The asset capitalists are the wealthy families, individuals or foundations that bring in capital and receive a financial return on the investment. The fourth actor is for example an NGO that gives input in the impact investment by sharing their network and knowledge. The latter two take a more passive role in the impact investment process: they mainly support by sharing financial or social (network, knowledge) capital (Jackson, 2012). This research therefore focuses on the two main actors in the impact

(12)

Page 11 of 72

investing process: the relationship between the asset manager (i.e. impact investor) and the demand-side actor (i.e. social enterprise).

Another important stakeholder in the impact investment field, and in the impact measurement process, is the beneficiary (i.e. user, customer, employee) of the social enterprise. This stakeholder is in most cases the one with certain needs that the social enterprise is trying to address, and therefore the one experiencing the outcomes and impact of the enterprise. It is in this sense highly relevant to add beneficiaries to this overview, and therefore to this research on impact measurement.

As illustrated, this research focuses on three main stakeholders: the impact investor (asset manager), the social enterprise (demand-side actor) and the beneficiary – the group or system that benefits from the intervention of the enterprise. The next section elaborates on the overall characteristics of the impact investment field.

2.1. The process of impact investing

Impact investing, as a party investing in a social initiative to achieve societal as well as financial impact (The GIIN, 2017), is not the same as Corporate Social Responsibility (CSR) policies or Socially Responsible Investments (SRI) (Tan, 2014). The CSR and SRI initiatives focus mostly on how to decrease the negative impact that corporate enterprises or other businesses are making. In contrast, impact investing is about ‘doing good’, so creating actual positive impact, instead of ‘doing no harm’ (Tan, 2014).

The following four criteria describe the basic characteristics of impact investments (Tan, 2014):

(13)

Page 12 of 72

2. Intentionality. These businesses are primarily focused on tackling societal issues. 3. Locality. It regularly targets rural areas.

4. Accountability. The business reports on social aspects, next to financial reporting. Two of the four elements that describe impact investments are comparable to the private, ‘regular’ type of investing in businesses: the venture capitalists. Venture capitalists pursue one goal: to make a profit by developing a business that increases the capitalizations and therefore its share value (Nikkonen, 2013). Risk is mitigated by this investor by taking a board position to keep control on decision making and the financial results (Nikkonen, 2013). These financial results need to be monitored and in most cases reported to internal and external stakeholders. The importance of both ‘profitability’ and ‘accountability’, as emphasized here, affects both private and impact investments. Both venture capitalists and impact investors have the aim to create a sustainable business that creates profit results, which they monitor to be able to report to internal and external stakeholders. In the case of impact investors, the results often need to be reported back to their own investors, the

impact capitalists (Jackson, 2012).

What makes impact investing different and more difficult from regular investments, is the addition of a second goal. Where venture capital has one major goal, according to Nikkonen (2013), which is making profits, impact investors face the challenge of achieving two goals. In addition to being profitable, impact investments have the aim to achieve societal goals as well. This is gathered in the concept of intentionality, which describes the characteristic of having the intention to create positive societal impact as part an enterprise’s core business (Tan, 2014). This is often called by practitioners as having an ‘impact first’ approach. The element of locality describes that the solution of the enterprise in most cases

(14)

Page 13 of 72

affects rural areas; areas where often people with poor circumstances live (Tan, 2014). These two elements add an extra achievement to the field of impact investments.

When combining the four factors, it shows that impact investments have two goals to achieve: next to profit goals, impact investors aim to achieve societal, local goals. The concept of accountability becomes then twofold as well: being accountable for both social and financial goals.

2.2. Difficulties in measuring conflicting factors

As argued by several authors, the combination of intentionality, profitability and accountability creates more complexity within impact investments due to the associated tensions in social-business goals. These tensions, identified by Smith et al. (2013), are caused by the contradiction between financial objectives (i.e. efficiency and commercial tools and social objectives (social mission and locality) (Smith et al., 2013). This duality between both goals emphasizes the relevance to be accountable, and thus measure, the achieved results of both goals (Smith et al., 2013; Yunus et al., 2010; Onyx, 2012; Bagnoli & Megali, 2009). However, the situation of seemingly conflicting goals can also cause the measurement process itself to consist of conflicts between measuring financial and societal factors. This duality in measurement is explained by Smith et al. (2013) as performing tensions, and is a central and important topic in this research.

Duality in measurement is seen in the characteristics of financial factors and societal factors. The characteristics of financial measuring show that it addresses a small group of shareholders, that the dimensions are more objective, quantitative and much easier to standardize (Smith et al., 2013). This is in contrast to the characteristics of societal goals,

(15)

Page 14 of 72

which address a broad eco-system of involved stakeholders, where metrics to assess performance are more qualitative, subjective and hard to standardize (Smith et al., 2010). The contradiction and possible conflicts between both factors can affect the way they are measured. Measuring financial and social results in the impact investing field is called impact

measurement. The scientific field of impact measurement in now explored.

2.3. Impact measurement literature

The existing scientific literature shows a lack of research on how impact investments measure both financial and social factors (i.e. their measurement process). Much is written about CSR-policies and SRI, or impact measurement within social enterprises, but the knowledge on how impact investors measure their impact is still underdeveloped. Since the above explained duality is comparable to the duality within social enterprises, the literature on measurement processes within social enterprises is explored in this research. Yet, before the literature on impact measurement within social enterprises is reviewed, an explanation is given on the process (and definition) of impact measurement.

2.3.1. Impact: what does it entail?

Mostly used by practitioners, but also recognized among scholars (Bagnoli & Megali, 2009; Arena et al., 2015) is the usage of the impact value chain to define the levels on which impact results can be measured. This impact pathway is based on the theory of change, developed by Connell et al. (1995) and popularized by co-editor Carol Weiss, as an evaluation tool to create insights in performances of societal community initiatives. This theory of change

(16)

Page 15 of 72

illustrates a causal process of the levels on which societal programs can be evaluated and measured on. The different levels in this impact value chain are described below.

Figure 1. Impact value chain

Source: Rosenzweig (2004)

Impact is defined here as the changes to societal systems (e.g. peoples’ lives, environmental damage reduction) minus the external factors that cannot be connected to the intervention of the social enterprise – changes that would have happened anyway.

The impact value chain defined by Rosenzweig (2004) has its roots in the social sciences. Impact accounts here for counterfactuals that happened by comparing the target group with a control group – as is also been done in social science experiments. The control group consists of people who were unable to receive any effects or outcomes of the intervention that is being measured (Rosenzweig, 2004). This definition of impact is also used by impact measurement companies as Sinzer, Steward RedQueen and by nongovernmental organizations as Oxfam Novib. In addition, this chain is also used by several scholars (i.e. Bagnoli & Megali, 2009 and Arena et al., 2015) to build impact measurement frameworks. Inputs What is put into the enterprise Activities Enterprises’ primary activities Outputs Results that can be measured Outcomes Changes to societal systems Impact Societal changes solely caused by enterprise Outcomes MINUS counterfactuals

(17)

Page 16 of 72

The impact value chain clarifies the definition of impact and on what levels measurement can be conducted. Next section elaborates what stages the impact measurement process entails.

2.3.2. Stages of impact measurement in impact investments

In addition to the different levels, impact measurement also occurs in different phases. So & Capanyola (2016) established a cycle of impact measurement, based on their research with leading impact investors and practitioners in the field. This cycle consists of four phases:

1. Estimating impact. Due diligence research to assess potential return (social and financial).

2. Planning impact. Confirm the metrics and methods used to monitor the effects. 3. Monitoring impact. Measuring and analyzing impact throughout the investment. 4. Evaluating impact. Measuring the social consequences of the investment after the

program has finished.

The whole process is important to take into account, since all stages are relevant in order to see how the actual measurement is taken place; and how the different stakeholders collaborate in that process.

As mentioned, the impact measurement process among impact investors has not been researched yet by scientific scholars. Since the explained duality in measuring two goals (i.e. financial and social) is seen in social enterprises as well, the literature on measurement processes in social enterprises is now explored in order to structure this research.

(18)

Page 17 of 72 2.3.3. Impact measurement processes within social enterprises

Amongst practitioners, the process of measuring impact results is called impact measurement. Scientific authors describe this process as performance measurement (Arena et al., 2015; Bagnoli & Megali, 2009; Alvord et al., 2004) or impact assessment (Vanclay, 2003; Onyx, 2012; Grieco et al., 2015)). All these concepts describe the same thing: the results and outcomes obtained by the deliverables and mission of the social enterprise. In this research, the term ‘impact measurement’ is used to describe the above stated concepts. In the existing literature on impact measurement within social enterprises, a focus exists on what models and indicators (i.e. methodologies) should be used to measure impact. Many scholars have conducted research on which complete models are needed to properly assess impact. The most well-known model of impact measurement is the Balanced Scorecard (BSC), which translates the vision, mission and strategy of the enterprise into measurable indicators (Hugues, 2014). The purpose of this scorecard is to balance three elements in the measurement process: i) financial versus non-financial, ii) long-term versus short-term and iii) external versus internal (Hugues, 2014). Besides the model of BSC, a couple of other models have been established to quantify the diffuse notion of impact. The two most famous models are: i) the Social Return on Investment (SROI) model, which uses monetary term to translate the social value that is created, and ii) the model of Global Reporting Initiative (GRI); a reporting tool for social enterprises (Hugues, 2014).

Other scholars tried to do the opposite of creating standardized models. These authors used an independent indicator approach to measure impact. For example, Bagnoli & Megali (2009) established a list of autonomous indicators that are based on broader dimensions (i.e. economic/financial). For each of the dimensions, corresponding indicators

(19)

Page 18 of 72

were established. For example, the social dimension is divided between indicators of key factors as human resources (% of disadvantaged workers), work conditions (safety and health indicators) and characteristics of the products/services (quality and quantity) (Bagnoli & Megali, 2009). The purpose of establishing autonomous indicators based on broader dimensions was to enable social enterprises to tailor the indicators to the unique needs of their social enterprise.

The division between models and indicators to measure impact illustrates the discussion that is going on about what method (model, indicator) should be used to properly measure impact. Bagnoli & Megali (2009) argue that impact indicators should not be formalized into a pre-established system (e.g. model), since it would harm the flexibility of social enterprises to frame the indicators to their own specific context (Bagnoli & Megali, 2011). This notion of specificity is currently supported by Grieco et al. (2015). These authors argue that no single model fits each organization that measures impact, due to a large variety among social enterprises of ways in which they measure impact. The current diversity forced the development of existing models to be tailored to the single needs of different types of organizations (Grieco et al., 2015). Due to this development, it is hard to study the field of impact measurement, since the mentioned diversity eliminates the opportunity to compare the impact measurement processes.

This literature review therefore focuses on the categorizations, i.e. themes, on impact measurement found in the literature. These categories are overarching themes where existing models and indicator dimensions are based on. Using categories creates the possibility to compare different cases amongst each other.

(20)

Page 19 of 72 2.4. Categories in impact measurement

The categories found in existing literature on impact measurement are used to build the conceptual framework for this research. The categorizations that are identified by Hugues (2014), Smith et al. (2013), Bagnoli & Megali (2009), Rosenzweig (2004), Arena et al. (2005) and So & Capanyola (2016) form the overall basis of this framework. The categories used in this overview are:

1. Relationship social and financial factors 2. Collaboration investor-enterprise 3. Measure all level impact

4. Stakeholder involvement 5. Goals of impact measurement

2.4.1. Relationship Social-Financial factors

As noted before, the conflicts between financial and social goals can have a reflection on how impact is measured (Smith et al., 2013). This expected duality in measuring makes it relevant to see how both factors are managed in the process of measuring impact.

The importance of measuring both social and financial factors is emphasized by Bagnoli & Megali (2009) as well. They argue that the measurement of efficiency and profitability should be accompanied with the measurement of social effectiveness (Bagnoli & Megali, 2009). Social effectiveness, in the context of social enterprises, is a fundamental element to meet the social needs the enterprise has been designed to address, and therefore pursuing the mission of the social enterprise (Bagnoli & Megali, 2009). The financial part of this theme is carried out by means of financial accounting systems and statements to get the

(21)

Page 20 of 72

information that is needed to measure efficiency and profitability (Bagnoli & Megali, 2009). Hugues (2014) emphasizes the relevant distinction between monetary and non-monetary factors in the impact measurement process as well. According to Hugues (2014), this element immediately shows the focus (or absence of focus) on monetization when measuring results.

The relationship between how social and financial factors are measured can be made visible by creating insights in the following: how financial results are reported on, what they do to measure social results and what the connection is between the two factors within the measurement process.

First category Relationship social-financial factors

2.4.2. Collaboration investor-enterprise

The expected conflicts between financial and social factors within impact measurement make it highly relevant to see how the impact investor and social enterprise collaborate. The division of roles in the measurement process and where the impact investors support the social enterprise on, should give valuable insights in how social and financial elements are valued and managed. In addition, it is interesting to see how agreements on impact are made, as part of the planning phase, identified by So & Capanyola (2016): what does the collaboration on planning the impact (and financial) goals look like? And how does the impact investor support on achieving and measuring both goals later in the process?

(22)

Page 21 of 72 2.4.3. Measure all levels of impact

The levels of the impact value chain, as defined by Rosenzweig (2004), show on what levels impact should be measured. As supported by Bagnoli & Megali (2009), social enterprises should have the ability to achieve goals and implement strategies that correspond to the needs of their stakeholders (Bagnoli & Megali, 2009), doing this on each level of the impact value chain. To be effective, the social enterprise should not only consider results on one level of the impact value chain, but it should look at the sustainability of all levels of the impact value chain (Bagnoli & Megali, 2009). This theme therefore aims to create insights on how impact investors measure the results on all levels of the value chain.

Third category Measure all levels of impact

2.4.4. Stakeholder involvement

The next category of stakeholders’ involvement is one of the main categories in the performance measurement model of Arena et al. (2015) and found by Bagnoli & Megali (2009) as well. Involving stakeholders is crucial when dealing with dual information requirements and several – possibly conflicting – stakeholder perspectives (Arena et al., 2015). Taking stakeholder’s perspective into account involves knowing who the stakeholders are and what their needs, wishes and satisfaction are. This shows that involving the stakeholder’s interests is about knowing what their needs are and measuring whether the outcomes of the enterprise contribute to those needs (Arena et al., 2015).

(23)

Page 22 of 72 2.4.5. Goal of impact measurement

The fifth category in this research, identified by Hugues (2014) as highly relevant, is the purpose of measuring. The purpose or goal of impact measurement can be labeled as ‘internal’, when it is mainly about: i) making managerial decisions for the organization itself, ii) improving the performance or iii) when measuring the progress of accomplishing the mission (Hugues, 2014). This internal management purpose is also underlined as important by Straub et al. (2010) and Onyx (2012). According to them, the insights gained from measurement enables successful management of the social enterprise.

On the contrary, external purposes have an outwards direction. The activities corresponding with an external purpose are for example: i) the activity of reporting to stakeholders, ii) communicating to outside parties, iii) increasing the access to resources and iv) activities to build legitimacy for the enterprise (Hugues, 2014).

These internal and external characteristics bring along certain values and effects of conducting the impact measurement. Hence, it is important to see how impact investors use the obtained insights.

Fifth category Goal of impact measurement

2.5. Conceptual framework

The categorizations that came out of the literature review help to structure the empirical part of this research, to be able to discover how impact investors execute and value in process of measuring impact. Based on previous section, the conceptual framework of this research is illustrated in figure 2.

(24)

Page 23 of 72

Figure 2. Conceptual framework

2 Collaboration

investor

-enterprise

3 Measure all level

impact

4 Stakeholder

involvement

5 Goal of

measurement

Insights in how these two actors collaborate can give important

information on what they both value in the process of measuring impact and whether there are conflicts visible on both sides.

Taking stakeholders’ needs in account throughout the whole value chain: - Measure the alignment of achieved goals and strategies with

stakeholders’ needs.

- How do you see that impact throughout the whole impact value chain?

Knowledge on:

- Who are stakeholders;

- What are the needs and wishes, and what is their satisfaction of the enterprises’ deliverables?

- External measurement purposes focus on: reporting to stakeholders, building legitimacy and outside communication. - Internal measurement purposes focus on: making managerial

decisions, improving performance and measuring progress of mission accomplishment.

-

1 Balance

social-financial factors

Creating insights in how these two factors are related; - How do SE’s report on both factors?

(25)

Page 24 of 72

3. Method section

The aim of this research is to achieve an in-depth understanding on the process of measuring impact by impact investors, where impact measurement is the central unit of analysis. This chapter explains the (motivation for the chosen) research design. The chapter discusses the sampling, the cases and the data collection and -analysis. The end of the chapter emphasizes important methodological issues within this research design.

3.1. Research design

To give a clear explanation for which research design is chosen, it is needed to first state the main research question:

How do impact investors measure the impact created in their collaboration with social enterprises?

To answer the main question, it is important to know what elements of measurement are valuable and important, as perceived by the participants. Creating insights into the meanings behind a certain measurement rationale becomes even more relevant in this study, since little scientific knowledge exists yet that could explain those meanings. The most appropriate research design is therefore a qualitative research strategy.

The above-mentioned statement is confirmed by Strauss & Corbin (1990): if there is little knowledge on a subject, the qualitative research methods are most suitable to explore the underlying nature and factors of the phenomenon in question (Strauss & Corbin, 1990). Qualitative research can be explained as often having an inductive view on how theory and research are related, where theory is in many cases generated out of qualitative (data)

(26)

Page 25 of 72

research (Bryman, 2012). This research incorporates the more inductive focus due to the lack of scientific literature on the subject. Theory is used to structure this study, where empirical research is used to build the theory in the impact investment field. Theory is in that sense supportive, but not leading in this study: the insights rely on the interpretations of the participants.

Several methods exist when conducting qualitative research. The most used method within this type of research is the interview (Bryman, 2012). How and why this method is used is explained in the section ‘Data collection’.

This study uses a multiple case study approach to answer the main question. The motivation of this approach is to represent a diverse group of participants while achieving in-depth insights on what is valued within the measurement process.

The next section elaborates on how the cases in this research are sampled and what their core characteristics are.

3.2. Sampling

The participants within this research are impact investment organizations which in most cases are investing funds (called asset managers (Jackson, 2012) in the literature review): organizations that receive and manage capital that is used to invest in businesses. The participants that were interviewed were the Investment Manager or Impact Director/Officer of that impact investment organization.

Participants were sampled through purposive and referral methods. Most of the participants were accessed through the network of Oxfam Novib, since Oxfam Novib is a partner of this research. This purposive sampling was done to balance the participants on

(27)

Page 26 of 72

certain characteristics, in order to create a group that represents the total group of impact investors, which now consist of 2,644 investors globally (The GIIN, 2017). The characteristics were chosen from an online desk research on the websites of the impact investors themselves. Eventually five characteristics were chosen. These characteristics were emphasized as factors where they distinguish themselves on, which was later confirmed in the conducted interviews. The five characteristics are: i) independent fund versus fund connected to a larger organization (e.g. bank, NGO), ii) investing in Western social enterprises versus investing in developing countries, iii) financial size of the investment, iv) shareholder (equity) or loan (debt) provider, v) stage of enterprise when investing: startup, early stage, growth.

Besides purposive sampling, a few participants were selected by using the referral sampling method. As a preparation for this thesis, two industry specialists were interviewed in order to gain their vision on impact measurement among impact investments. These interviews provided valuable insights in a field wherein little scientific knowledge exists. This interview also functioned to get access to their network of impact investors. One of these industry specialists, an organization named Sinzer, referred participants from its network to this research.

The eventual sample of 13 participating organizations is illustrated in Figure 3, divided between the characteristics.

(28)

Page 27 of 72

Figure 3. Participants table

3.3. Data collection

The first step in conducting the qualitative cross case study among thirteen impact investors (i.e. cases) is collecting the data. To conduct data of all of the cases, semi-structured

Characteristic Participant Fund is connected or independent Investing in Western businesses Investing in businesses in development countries Size of investment Stage of funding Equity or debt Social Impact Ventures

Independent Focus on the Netherlands X 0,5-3mln Early stage Equity Noaber Ventures

Independent Focus on the Netherlands X 100K-1,5mln Growth, some startups Equity (minority) KNHM Participaties

Independent Only the Netherlands

X - Startup Both

AquaSpark Independent Global Global 250K-5mln

Early stage

Equity

DOB Equity Independent X Africa 1mln Growth Equity

(minority) Root Capital Independent X Africa 50K-2mln Small

farmers

Both

EWB Independent No Africa 10K-100K Startups Both

PYMWYMIC Fund Independent Europe X 250K-4,5mln Early stage - Growth Equity

Goodwell Independent Yes Africa & India 250K-5mln Early stage Equity (minority) Triodos Invesment Management

Connected Global Global - - Both

AgDevCo Independent X Africa 1-2mln Startup Both

I&P Independent X Africa

300K-3mln Growth, some startups Equity (minority)

Sovec Independent X Yes

300K-600K

(29)

Page 28 of 72

interviews with open-ended questions were used to obtain deep insights into the process of measuring impact. Semi-structured interviews as a qualitative method was chosen for two reasons. This type of interviewing is a proper way to structure the interviews, using an

interview guide (Bryman, 2012) as a reflection of theoretical themes that came out of the

literature review. Simultaneously, the open-ended questions made sure that participants have a great deal of leeway when responding (Bryman, 2012), which left space for new topics to emerge. These two characteristics are relevant when creating an understanding on a new topic or phenomenon, which is the case for this research.

As mentioned, the participants were questioned on pre-established topics to gather their understanding on the execution and valuation of the impact measurement process. The topic list (i.e. interview guide) was based on five important themes that came out of the literature review, which are: i) Relationship social-financial factors, ii) Collaboration investor-enterprise, iii) Measure all level impact, iv) Stakeholder involvement and v) Goal of measurement. The topics functioned as a guidance for the interview, not as a fixed structure, in order to be flexible on emerging topics. Creating a research setting where new topics could emerge was relevant due to the lack of (prior) scientific knowledge.

The first three collected interviews were used to make the topic list more developed: the topic list was pre-tested within these first interviews, so some themes needed to be adjusted. This was necessary and relevant since no previous research existed to which the topic list could be compared or copied from. The final interview guide can be found in Appendix II.

(30)

Page 29 of 72 3.4. Data coding and analysis

The respondents gave permission to record the interview. The recorded interviews were translated in transcriptions by the researcher. A code list (see Appendix III) helped to systematically organize the transcripts by coding the pre-established and emerged themes into main codes and sub-codes of this list. The main codes are the same as the main themes, whereas the sub-codes included some emerged topics (illustrated in Appendix III). This approach shows a combination of deductive and inductive coding: new codes were added and some codes were found to be less (or more) important.

After coding the transcriptions, the first step of a cross-case analysis was executed. A word table (see Appendix IV) was created that displays the data for each case to reflect the results (Yin, 2009) on the five main themes. The table with first results based on the interview transcripts created the opportunity to find possible patterns among the three cases (Yin, 2009). The first patterns were translated into a summary that was sent to all the participants to check if the data was well interpreted by the researcher.

The several themes, and possible patterns, were then reviewed by using the data (word table and coded transcripts) to report in detail on the results of that theme and the theme-patterns.

3.5. Ensure qualitative research quality

The overall critique on qualitative research is fourfold: i) too subjective, ii) hard to replicate, iii) problems of generalization and iv) lack of transparency. How this study deals with these issues is discussed here.

(31)

Page 30 of 72

The first issue seen in qualitative research is that it is too subjective (Bryman, 2012). The argumentation for this critique is that the findings rely too much on the view of the researcher about what is significant or important (Bryman, 2012). This issue is solved in this research by having the participants check the most important findings of their own case. The summarized data of individual main findings was sent to the participants, whereon the participants could give feedback and approval. This approach aims to decrease the subjective character of the main findings.

The second issue in qualitative research describes the degree in which the study can be replicated by a different researcher. It is difficult for qualitative research in general to meet these criteria, since it mainly focuses on social settings that cannot be ‘frozen’ for other researchers to investigate (Bryman, 2012). The external reliability (i.e. replication) is negatively influenced in this research, since the researcher partners with Oxfam Novib. Besides a specific social role that each researcher has, the role of working for a company with which impact investors can possibly do business, can influence the external reliability. However, since Oxfam is not a competitor or rival player in the industry, no barriers are expected from impact investors to give honest answers.

The third issue is formally called external validity. This concept is about generalizing findings to other social settings. This concept is usually a problem for qualitative researchers, since they mainly use (single) case studies or small samples (Bryman, 2012). This research aims to tackle this problem by finding a representative group of impact investors, selected on several characteristics2_{. The representation of different types of impact investors creates}

(32)

Page 31 of 72

the expectation that the findings are more suitable for generalization than is possible with other case studies.

The last critique on qualitative researches involves transparency issues. It mainly criticizes on the indistinctness in how choices are made by the researcher. Bryman (2012) mentioned that most often the sampling and analyzing choices show a lack of transparency. The method section aims to tackle the transparency issue by describing in detail how the research is conducted.

(33)

Page 32 of 72

4. Results

The analysis of the interview data resulted in case-patterns within themes and across themes. Within the themes of ‘Relationship social-financial factors’, ‘Collaboration’, ‘Measurement of all level impact’, ‘Stakeholder involvement’ and ‘Goal of measurement’

several insights and new concepts emerged among the cases3_{. Patterns that came out of the}

analysis are now discussed with a theme-by-theme approach. Interesting patterns across the five different themes are discussed in the next chapter, ‘Discussion’.

4.1. Relationship Social and Financial factors

Based on the literature review, the achievement of dual goals was expected to influence the simultaneous measurement of the two factors (i.e. social and financial) (Smith et al., 2013). Social results are here defined as results that have a positive effect on societal systems, so for example an increase in the amount of jobs in a community, accessible and qualitative healthcare, more and affordable schools and so on. Financial results in the case of impact investment are results made visible in balances and statements on costs and revenues of the enterprise.

The cases in this research differ on how they view this conflict: one group of impact investors argues that there is no trade off on achieving and measuring social and financial results, whereas the other group believes that it is indeed about two different types that cannot be measured at the same time. These different views of the two groups are based on how they relate the two factors (i.e. social and financial) in their measurement process: on

(34)

Page 33 of 72

an incorporate basis (i.e. no conflicts; social and financial factors are measured in the same way) and separate basis (i.e. both results are measured on separate moments in time) basis. This section is therefore divided into those two categories: cases that have a separate relationship and cases that have an incorporated relationship when measuring social and financial factors.

4.1.1. Incorporate relationship in measuring

Most of the cases in this research argue that social and financial factors can be measured simultaneously and that no conflicts in measurement need to exist. These cases are therefore placed in the category of incorporate measurement relationship. The cases that measure both factors at the same time are:

 Sovec (Dutch)

 PYMWYMIC (Dutch)

 DOB Equity (Dutch)

 Social Impact Ventures (Dutch)

 Noaber Ventures (Dutch)

 Goodwell (Dutch)

 AquaSpark (Dutch)

 EWB (Canadian)

Societal reporting among these cases consists of collecting quantitative numbers that mainly focus on outreach (e.g. number of (female) employees, number of new customers). The quantitative character of social results enables the enterprise to easily and regularly report towards the impact investor. Social results are therefore in all cases within this category

(35)

Page 34 of 72

mutually measured with financial results, which means on a quarterly basis (or for some; monthly).

Investors in this category take a more pragmatic approach in measuring impact: they do not conduct impact studies themselves. The reason for this approach is the following:

“With starting enterprises, we do want to have a pragmatic approach towards our entrepreneurs: we do not want to ask them to send out a survey every week. It is not realistic to conduct deep impact studies as a starting enterprise, due to the lack of capacities.” (Social Impact Ventures)

This research found other ways in which the cases measure the qualitative meaning behind the social results. As Noaber Ventures explains:

“Correlations between output and outcome4_{are based on studies, so the enterprise can focus on} extracting output information from its patients. This relatively pragmatic approach [aims to make impact] easily measurable.”(Noaber Ventures)

The studies, mentioned in the last quote, are established to prove the correlation between the output that is being measured and the outcome on beneficiaries. The cases of Noaber Ventures and Social Impact Ventures stimulate enterprises to conduct these kinds of studies, where the execution of the study is done by the enterprise itself. The cases argue that by

(36)

Page 35 of 72

stimulating the enterprises to do the research themselves and to see the benefits they create, it eventually helps them to (independently) improve the results of the enterprise.

Two cases in this category measure the more qualitative ‘meaning’ behind the quantitative numbers through storytelling studies. EWB and AquaSpark both collect these studies to receive more in-depth reviews of internal stakeholders once a year. Participants in this study are thus employees from both investor and investee who represent the effects or an established change due to the enterprise. The reason for conducting these storytelling studies is to grasp the meaning behind the results.

4.1.2. Separate measurement relationship

The second category consists of a group that measures financial and social results separately, which creates similar characteristics of the measurement process within this category. The cases that are included here are:

 KNHM Participaties B.V. (Dutch)

 AgDevCo (English)

 Root Capital (American)

 I&P (French)

 Triodos Investment Management (Dutch)

As seen in the first category, financial reports are collected on a quarterly or monthly basis here, mostly performed by the investment manager. The process of reporting on social results is executed on a separate basis; established impact targets or metrics are annually monitored by the investor. Due to low internal available capacity and the high amount of

(37)

Page 36 of 72

enterprises to monitor, societal monitoring appeared to be more significant to measure on a yearly basis within these cases.

In addition to the yearly social reports, all of the cases conduct impact studies themselves. Impact studies in this category range from base and end-line studies to impact studies during the investment period. These cases take a small sample of their complete investment portfolio to conduct impact studies with, by researching the enterprise’s beneficiaries. Due to capacity issues, the sample ranges from 1 to 10 enterprises per year. The reason for conducting these impact studies is stated below:

“Impact studies are something we really do to deepen the impact of one or two companies every year. But because of our resources it is only a small share of our total investments. It is also to build knowledge in specific sectors, to learn about specific investments, which is useful for all other investments. Not something we link to financial reports, it is quite independent.” (I&P)

In addition to these two ways of measuring, one interesting theme emerged throughout several cases in the two categories. This insight involves receiving a reward for both financial and social results. This model makes sure the investment managers only receive a certain reward when a specific percentage of the impact target is achieved, next to the financial targets. PYMWYMIC and Social Impact Ventures both have this incentive model in place, ‘to

make impact results an inherent part of our normal business processes.’ (PYMWYMIC).

Two other cases, I&P and KNHM, are currently piloting this model. I&P has the same model as the two cases above mentioned, where KNHM stimulates the achievement of impact targets on the enterprises’ side; the social enterprise receives a bonus when

(38)

Page 37 of 72

achieving a certain percentage of the impact targets. The reasoning behind this incentive model is to raise awareness on the impact targets and to drive the enterprise or investor to report on the created impact results.

This section showed two ways in which impact investors measure social and financial results. Furthermore, within this theme a new topic emerged: the rewarding on both financial and impact results to raise awareness and stimulate measurement. The next section elaborates on the collaboration between both the investor and investee.

4.2. Collaboration

The findings within this theme give insight in how impact investors and social enterprises collaborate. From the interviews, all the cases label their role as ‘active’. ‘Active’ as a concept can be described for all cases as ‘active involvement’: every impact investor that has participated in this research is in frequent contact with their investee (directly or through local partners or local teams). Frequent means on a range from once a week to once a month. Another element that shows their active involvement is the board position that most of the impact investors take as well; only Root Capital and EWB are not involved as a board member since they have been working mostly with convertible debt. The purpose of this board position is to realize control on important decisions within the enterprise.

Some degree distinctions can be made in the formal (i.e. board position) and informal (i.e. frequent contact) involvement between impact investor and social enterprise. Most of the cases view their role as being supportive towards social enterprises, without getting involved in operational, day-to-day issues. DOB Equity explains why they chose to merely support on a strategic, management level:

(39)

Page 38 of 72 “We have tried to be operationally involved by hiring something we called ‘the traveling project manager’. Every time issues occurred with one of our enterprises, this project manager was able and available to work for a couple of months within that company. We noticed that we had created our own problem by having this model of hands-on co-working with the enterprise: if there are problems within the management team of the enterprise, we need to find sustainable solutions to strengthen that team. They have to do that themselves, not us, because then the team will become too passive.” (DOB Equity)

Several other impact investors underline the importance of leaving day-to-day issues to the social enterprise itself, because ‘we have a lot of faith in the enterprises and teams we invest

in’ (Goodwell) and ‘since we have to believe, from the beginning onwards, in the enterprise’s management team’ (PYMWYMIC).

Two cases in this research show that their role can evolve into hands-on, operational co-working within the enterprise. AgDevCo and EWB both have the capability to employ (AgDevCo) or recruit (EWB) people for a specific task within the social enterprise. Co-working tasks consist of: helping to set up management accounts and develop a business plan (AgDevCo) and working on operational, financial or technical tasks (EWB). Both cases have this co-working model in place because it is necessary sometimes due to the challenging conditions under which the startups are working and because enterprises benefit from the expertise that the co-workers bring.

The levels of involvement are reflected in the financial and business support that the impact investors provide. Financial and business support in this collaboration can be divided

(40)

Page 39 of 72

into four dimensions, in addition to the evident capital support that is provided. Figure 4 showcases the four dimensions.

Figure 4. Financial and business support dimensions

Dimension Sub-dimension Quote # of

investors5 i. Advisory .. with a

coach

‘KNHM has a large pool of business coaches available for every social enterprise we invest in [..], whom are all experienced as an entrepreneur.’ (KNHM)

3

.. with experts or consultants

“Within Triodos, it regularly happens that a consultant is sent to the business to improve their environmental and social management system, for example.’ (Triodos Investment Management)

5 .. with informal advises from investment manager

“My duty is to make sure that the entrepreneur gives me a call when he or she is struggling with certain issues: to build this trustworthy relationship.” (Noaber Ventures)

10

ii. Delivering

talented employees

“This [co-working] is needed since our enterprises operate under challenging conditions. The

co-workers that bring in agriculture or business expertise are beneficial to the enterprise in building a sustainable business.” (AgDevCo)

2

iii. Training or

workshops

“The most important support that is provided by Root Capital are the loan- and financial management trainings.” (Root Capital)

2

iv. Access to

network

“With our community, we have a lot of expertise that we can bring in to make sure we can add value to young enterprises who prioritize impact.” (PYMWYMIC)

All

Where the financial and business support provision is evident among all cases in this research, the support on impact (achievement and/or measurement) is more ambiguous. Five of the thirteen impact investors have no clear impact support system in place. Others

(41)

Page 40 of 72

stated that it is automatically a part of what they do to support enterprises (AquaSpark), or that it is provided by establishing and tracking the impact targets (PYMWYMIC) that came out of a Theory of Change model (Noaber, Social Impact Ventures). This latter point is comparable to the support provided by AgDevCo, KNHM and EWB, who help to set up an impact system by respectively sharing their own practices, through the use of available coaches or through portfolio officers who are in frequent contact with the enterprises.

What stands out here is that the impact support consists of much more diffusion, where no questions at all exist around the financial/business support within the different cases. This can be linked to the point made by Smith et al. (2010) in the literature review: impact or social factors are more difficult to measure, which, proven here, makes it also more difficult to support.

The last result in this theme relates to the phases of impact measurement. In addition to the monitoring phase, referred to in the above stated results as ‘reporting’, the phase of

planning impact was emphasized as relevant by So & Capanyola (2016). The phase of

agreeing on impact plans beforehand is remarkably similar among the cases. The process of planning consists of the establishment of impact targets or metrics, which is in many cases done in collaboration with the two parties. All cases emphasized the importance of the alignment of impact targets with their own impact strategy. In addition, some cases show that the estimated impact targets are formally anchored in the contract between impact investor and social enterprise.

(42)

Page 41 of 72 4.3. Measure all levels of impact

Measure on all levels of impact, the theme based on the researches of Rosenzweig (2004) and Bagnoli & Megali (2009), deals with measuring the results on each level of the impact value chain (as illustrated in Figure 16_{). Five of the thirteen cases conduct impact studies in}

order to measure towards end-level impact among beneficiaries. The earlier mentioned

impact studies7_{aim to achieve qualitative insights into the outcomes on beneficial groups.}

Conducting impact studies is therefore one way to measure towards end-level impact. Other methods that are used by cases to measure towards the end-level impact are: i) storytelling: interviewing stakeholders within the investment organization or investee (EWB, AquaSpark) and ii) using studies among beneficiaries conducted by the enterprise to connect outcomes to the measurable outputs (Noaber, Social Impact Ventures). Some cases do not measure towards end-level impact, since they claim it to be the responsibility of the enterprise or because they have not established a strategy on impact studies yet (Goodwell, DOB Equity, PYMWYMIC).

In addition to what impact investors do to measure on several levels of the impact value chain, new interesting topics emerged within the main theme. These topics mainly stress the relationship between impact investor, social enterprise and beneficiary. Depending on the relationship in question, different types of impact can be distinguished. Figure 5 illustrates these relationships and types of impact.

6_{Figure 1 can be found on page 14 in this research.}

(43)

Page 42 of 72

Figure 5. Direct and indirect impact

The concepts mentioned in Figure 5 were found in the interview data. A practical article (from Brest & Born (2013) that was found after these themes emerged) gave insights in the different types of impact. The two types of impact appear to be divided into three levels: i) direct impact as investment impact, ii) direct impact as enterprise impact and iii) indirect impact on beneficiaries.

Investment impact is about achieving the goal of additionality: an investment made by

impact investors increases or enables the enterprise’s social output to grow in quality or quantity (Brest & Born, 2013). From the interview data, it was found that this type of impact is something that several cases are aware of: five cases have an evaluation system in place (i.e. surveys sent out to social enterprises) to understand their influence on the enterprise. As stated by EWB:

“This type of evaluation is done with the purpose to achieve our mission and vision: to see if the organization's strategy is effective or not.” (EWB)

Impact investor Direct impact - Investment impact - Enterprise impact Indirect impact Social enterprise Beneficiari es

(44)

Page 43 of 72

Investment impact is, as pointed out by EWB, I&P and Sovec, as well achieved by having a certain investment focus. A focus that lies on investing in startups or early stage enterprises in which commercial or other impact investors do not invest. This is part of their mission and impact strategy. Achieving impact with the investment focus is formally included in the impact measurement process as an important criterion when selecting incoming enterprises.

The other component of direct impact is the so-called enterprise impact. Enterprise impact is defined as product impact – the impact of the goods or services that the enterprises deliver – and operational impact – the impact of management practices on the employees and the effects on job creation in the community or other environmental effect of its business (Brest & Born, 2013). This direct impact is found extremely important by impact investors, since their overall goal is to maximize the growth of the enterprise – which is reflected in this type of impact. All cases receive insights into the achieved enterprise impact from the impact targets, which are measured with the monthly or yearly impact monitoring reports. A different type of measuring enterprise impact is seen in the case of KNHM, which measures the impact of the enterprises through the sessions that coaches have with the enterprise. Two other cases, EWB and AquaSpark, add a more qualitative approach to measuring enterprise impact by collecting storytelling of involved stakeholders on investee level on an annual basis. The reason for conducting storytelling studies is explained within the concept of indirect impact.

The next level after measuring enterprise impact is to measure how the output (enterprise impact) contributes to the intended outcome (Brest & Born, 2013); the impact and social value of the created goods and services on beneficial groups or systems. This type

(45)

Page 44 of 72

of impact emerged out of the interview data as indirect impact: the concept of ‘indirect’ means that the sphere of influence of the impact investor lies with the enterprise, not with the beneficiary. To be able to grasp the created indirect impact, five cases conduct impact studies themselves (as discussed in the previous section) to see actual outcomes on beneficial groups or systems. Another way to prove the correlation between output and actual outcome is to stimulate the enterprise to conduct a study that proves this correlation, as seen in the cases of Noaber Ventures and Social Impact Ventures. If it is impossible for the enterprise to conduct such a study, a more general study on the same subject is used to represent the intended outcome on beneficiaries. The last insights on how the intended outcome is represented is illustrated by EWB and AquaSpark. These impact investors collect more in-depth information on the change through storytelling studies with the investees: representing their vision on what the intended outcome is. The reason for conducting these stories is explained by AquaSpark:

“If a real change has occurred due to our involvement with the social enterprise, we would like to show how this change has happened. Our main goal is to have an indirect impact on this sector [of fish feed], but that is difficult to measure. Therefore, we use stories and examples [that we collect from people that are involved in the project].” (AquaSpark)

Other cases mention that it is not measured (yet) or that they do not feel the need or responsibility to conduct this study. An explanation was giving by PYMWYMIC for not conducting indirect impact studies:

How impact is measured when impact investors invest in mission driven enterprises