Impact Investing ALL PDF

What is Impact? Session 1 Impact Investing (7.170), Fall Term 2024, University of St. Gallen Prof. Julian Kölbel 1. Defining Impact 2 Impact in a nutshell: 3 A definition of Impact Impact is the chan...

What is Impact? Session 1 Impact Investing (7.170), Fall Term 2024, University of St. Gallen Prof. Julian Kölbel 1. Defining Impact 2 Impact in a nutshell: 3 A definition of Impact Impact is the change in a parameter Y caused by activity X 3 Ingredients: Parameter Y that I want to change Action X that I take A claim that X causes a change in Y (“aka model or theory of change”) 4 Impact = Outcome - Counterfactual 5 Think of an example where you had impact… (2min) What was the Y? What was the X? Why did X cause Y? Are you sure Y would not have happened without you? 6 Why is causality essential? If X does not cause Y, no amount of X will help We waste time and money on X1 that we could spend on X2 7  We do not measure impact! Causal effect? X Y We can measure X and Y. We can infer the effect of X on Y. And based on that, we can predict the impact of different Xs on Y, to help us decide. As a result, there is always some uncertainty about impact. 8 Impact attribution is often impossible Milk and Espresso are both necessary to have cappuccino Both have impact (Y being the cappuccino) How much of the outcome can we attribute to milk? 9 2. Investor Impact 10 Learning Goals 1. I can define impact in the context of investing 2. I can compare and contrast different mechanisms of investor impact 3. I can qualitatively assess the impact of concrete investment products 11  Why invest sustainably? 3 Motivations ”I don’t want to profit ”I want to help from coal mining.” fighting climate Values Impact change” Performance ”Considering climate risk will yield better returns” 12 Investor Impact and Company Impact 13 Investor Impact and Company Impact 14 Key Concept #3: You can grow green companies or improve brown companies Grow green companies: Grow green companies Companies with a positive impact. Companies whose growth is limited by external financing conditions. Improve brown companies Small and young companies Impact per Size Immature financial markets Companies with intangible assets Improve brown companies: E.g., voting, engagement & conduct-based screening Companies that have room to improve (also large ones) Size Restricted to incremental change 3. Examples 16  Example: Root Capital financing the Musasa Coffee Cooperative (Rwanda) Sub-Saharan Africa: Low value-add for farmers selling raw coffee locally Global specialty coffee markets offer much higher revenue to farmers Farmers often have no access to financial markets for the required capital 17 OWL Ventures ED-Tech focused VC fund 600 million learners in 240 countries Affordable quality education at scale Key value and impact proposition: navigating procurement in the education sector 18  Example: Hermes engaging Sinopec on Climate Change Methane leakage form oil production causes high GHG emissions. Hermes initiated a dialog with the management of Sinopec on climate change. Sinopec launched a methane reduction program, saving 3 million tons of CO2. Market Signals Does exclusion affect the market price of corporate debt and equity? Does it change the effective cost of capital? How dependent is the company on external financing? Non-Market Signals Markets are embedded in political contexts Exclusions can be a visible statement and influence the debate. Evidence Level for different Investment Approaches SI Approach Evidence Level Shareholder Engagement B: Empirical Evidence Thematic Investing, Impact Investing B: Empirical Evidence ESG Integration, Best-in-Class, C: Model-based prediction Norms-based-screening Industry Exclusion D: Narrative Key Requirements for the different Investment Approaches to be effective SI Approach Requirements Limitations Industry Exclusion Publicity Depends on political process ESG Integration, Best-in- 1. Deviation from Benchmark Works better for low-cost Class, Norms-based- 2. Clearly defined criteria improvements screening 3. Used by many investors Shareholder Engagement Influential Institution Works better for low-cost improvements Thematic Investing, Impact 1. Portfolio companies have net Business model viability Investing positive company impact often depends on policy 2. Growth of portfolio companies is limited by financing constraints How to maximize Investor Impact -> Key recommendations. Step 1: Enable impactful growth with alternative investments Allocate capital to young impactful companies that struggle for funding, as much as your risk tolerance allows. Step 2: Encourage improvement across the portfolio Pursue an engagement and voting strategy across your entire portfolio. If you screen, screen both equity and debt, make criteria clear, focus on broadly supported ESG practices that can be implemented at reasonable cost. Step 3: Support systemic change Be vocal about your investment decisions. Emphasize the need for political change. Assignment Read the “Investors Guide to Impact” Answer the quiz Why are you in this course? 3 Comprehension questions 25  Microfinance and Experiments Session 2 Impact Investing (7.170), Fall Term 2024, University of St. Gallen Prof. Julian Kölbel (Today, Prof. Stefano Ramelli) Learning Goals I have an overview of the microfinance market I can articulate the theory of change of microfinance I evaluate experimental results on the impact of microfinance 2 Who am I? - Stefano Ramelli - Assistant Professor at HSG since August 2022. - MSc in Economics at the University of Edinburgh, PhD at the University of Zurich. - Before the PhD, > 5 years of experience in sustainable/ESG. - Main courses: Corporate Valuation (Spring), CFA Challenge (Fall) - Research in corporate finance, climate finance, and behavioral finance: https://sites.google.com/view/stefanoramelli/about 3 “Long time ago in a galaxy far, far away...." In 2010-2011, I contributed to setting up a microfinance project in Malawi, financed by a fair-trade company. My little “secret” microfinance background. I like microfinance and I like experiments. 4 Banks as traditional fuelers of development «The process of Creative Destruction is the essential fact about capitalism.» Steady-state economy: No development. By extending credit, banks create purchasing power, enabling new entrepreneurs to innovate. The entry of new entrepreneurs fosters “impact”. 5 But what’s the problem with trading banks? Adverse Selection: Banks face difficulty distinguishing between high-risk and low-risk borrowers due to incomplete information. This uncertainty can lead banks to restrict credit, even for creditworthy individuals! Moral Hazard: Borrowers may take riskier actions after obtaining loans, which the bank cannot monitor effectively, increasing the likelihood of default. Hence, require collaterals. Lack of credit history or collateral: Many individuals or small businesses, particularly in developing regions, lack formal credit histories or sufficient collateral to secure loans. Mohammed Yunus Pioneer of Microfinance Founded Grameen Bank in 1983 Nobel Peace Price in 2006 7 Original Idea: Lift borrowers out of poverty by giving them access to credit Traditional banks: Lend primarily against physical collateral and guarantees. Microfinance Institutions: Lend based on trust and social collateral. Main strategies: Group Liability: Replaces individual liability with group responsibility. Positive Assortative Matching: Encourages self-selection into groups with similar risk profiles. Peer Monitoring: Leverages group oversight to ensure repayment. Dynamic Incentives: Increasing benefits for good repayment behavior. Social Collateral: Relies on social ties and community reputation. Women are a primary focus due to: Limited access to traditional financial services. Higher likelihood of using loans to support families. Reliability in repayment. 8 Microfinance before microfinance Rotating Savings and Credit Associations (ROSCA) or Accumulating Savings and Credit Associations (ASCAs): The true predecessors of modern microfinance. Existed for centuries all over world, addressing the principle: “You cannot save alone.” Operate on a rotation system where each member receives a loan in turn, promoting shared financial access. Rely solely on self-monitoring and mutual support, without the need for external enforcement. Modern microfinance leverages on these informal organizations. 9 Saving groups: How they usually work Regular Meetings and Shared Contributions: Members attend regular meetings and purchase "shares" to build the group’s collective savings, which are securely stored. Commitment and Mutual Support: Active participation strengthens individual responsibility while fostering mutual assistance and collective saving. Establishing a Social Fund: Members contribute to a shared fund to support unforeseen expenses, promoting financial resilience within the group. Enabling Entrepreneurial Growth: Over time, the pooled savings serve as a resource for loans, empowering members to pursue entrepreneurial ventures. 10 Example: Saving groups in rural Malawi 11 Microfinance in practice Village Selection: Based on population, poverty level, road access, political stability, and safety. Community Integration: MFIs introduce their mission and services after village selection. Group Formation: Peer pressure and collective responsibility encourage repayment discipline. Training: Educational program on financial procedures, business skills, and credit discipline. Village Center: Central hub for group coordination and financial transactions. Weekly meetings for loan applications, utilization discussions, and community issues. 12 Forms of Financial Inclusion Access to credit Focus on those that are not Access to savings served by commercial banks Access to insurance Bottom of the Pyramid (BoP) population Digital payment services i.e. those without collateral, Financial literacy formal documentation, etc. Marginalized groups, remote locations 13  Ex-Ante: Theory of Change Input Activity Output Outcome Impact 14 Theory of Change Microfinance Input Activity Output Outcome Impact Capital Lending to Business Increased Long-term people investment sales & profits Increase in without household collateral income (poverty trap) 15 Poverty Trap Generalized Poverty traps are situations where income reverts back to some lower level. Income tomorrow Can you think of examples? x1 x0 Income today 17 The poverty trap Income does not cover basic needs No money to invest, e.g. in Farmer has very low yield fertilizer 18  Poverty Trap Generalized Poverty traps are situations where income reverts back to some lower level. Lack of money to invest in Income tomorrow technology Lack of sufficient food to be productive Lack of access to healthcare or jobs (too far away) x1 x0 Income today If a person in the poverty trap receives a sufficient cash shock, then this person can grow its future income. 19  Microfinance Market Overview Total outstanding loans about USD 200 billion 20 Source: https://symbioticsgroup.com/wp-content/uploads/2020/02/symbiotics-symbiotics-2019-miv-survey.pdf  Microfinance Market Overview Main (Swiss) players: Symbiotics BlueOrchard Responsability 21 Source: https://symbioticsgroup.com/wp-content/uploads/2020/02/symbiotics-symbiotics-2019-miv-survey.pdf  Microfinance Market Overview Investing without currency hedge allows to invest in more countries as some currency pairs are not hedgable in practice 22 Source: https://symbioticsgroup.com/wp-content/uploads/2020/02/symbiotics-symbiotics-2019-miv-survey.pdf Microfinance asset have diversification potential https://www.blueorchard.com/wp-content/uploads/report/bomf/BOMF_EN_P.pdf 23 Controversies in 2010 “the industry [has] become no Microfinance boomed, especially better than the widely despised after the Nobel Prize village loan sharks it was intended IPO of SKS Microfinance, India’s to replace.... The money lender largest MFI in 2010 lives in the community. At least you can burn down his house.” Increasing reports of predatory lending, too many MFIs, and over-indebtedness Question marks about impact India banned Microlending for some time Reading: Help microfinance, don’t kill it 24 Attention to microfinance 25 Experiments for Microfinance (aka randomly controlled trials) 26 Esther Duflo & Abhijit Banerjee Spearheaded the use of randomized controlled experiments in development economics Founded J-PAL (Jameel Poverty Action Lab), which applies RCTs worldwide to understand better how to help the poor Nobel Prize in Economics 2019 (with Michael Kremer) 27 Example: Impact of Microfinance 5 A new microlender has set up 4.5 business in Guatemala. They have so 4 far 1000 clients. To estimate their 3.5 impact, they have tasked a 3 consultant to collect data on a 2.5 control group. 2 Treatment = person has taken a 1.5 1 microloan of $100 0.5 Outcome: Change in household 0 income after 1 year in % Treatment group (T) Control group (C) Results look promising, but what Outcome (Y) could be a problem? 28 Ex-Post: Experiment 29 Treatment Effect Treatment Effect: YiT – YiC Read: Outcome for unit i when treated minus outcome for the same unit i when not treated We never observe this! You can Two roads diverged in a yellow wood, And sorry I could not travel both only take one path. And be one traveler, long I stood And looked down one as far as I could To where it bent in the undergrowth. (full poem here) 30 Instead: Average Treatment Effect (ATE) 5 We can analyze the difference in group means: 4.5 4 E[YiT – YiC] 3.5 Read: The average outcome (Y) of those units (i) that got 3 2.5 treated (T) minus the average outcome of those units 2 that did not get treated. 1.5 This is an average effect (it can be very different for 1 individual units). 0.5 0 Difference in group means = causal effect + selection bias Treatment group (T) Control group (C) If the treatment is assigned randomly, selection bias Outcome (Y) disappears. 31 Implementation Partners For real-world experiments, you need real-world partners Ideal time: during pilot phase, many organizations are eager to learn themselves whether their program (i.e. treatment) is working, before they scale Experiments are costly and intrusive, so the implementing partner needs a strong rationale to do it (for example making a valid impact claim) Often, complete randomization is unfeasible Start-up cannot decline clients at random It is unethical to withhold help from those that need it most But there are alternatives: 32 Alternatives to complete randomization Approach Issues to keep in mind Oversubscription Treatment is constrained by capacity, It can be unethical to assign treatment at so access to treatment can be random and not based on needs. assigned by a lottery. Phase-in Instead of randomly assigning It is important that the treatment effect plays treatment, the starting point of out before treatment 2. Long-term effects treatment is randomly assigned. cannot be estimated. A problem can arise, Compare treatment cohort 1 to when cohort 2 changes behaviour due to treatment cohort 2, before anticipation of treatment. treatment 2) Encouragement Design Rather than randomly assigning the This only affects the probability of treatment, treatment itself, randomly chosen but when used as an instrumental variable, subjects receive an encouragement still allows to estimate a causal treatment to select into treatment effect. 33 Limitations of RCTs Spillovers and general equilibrium effects are hard to capture External validity: to what extent can impact in setting A be assumed to hold in setting B? For this, we need theory. RCTs can be very restrictive. One may want to adapt the intervention quickly based on interviews and qualitative methods, rather than keep everything fixed until the RCT results are in 37 So --- does Microfinance work? A review of 7 RCT studies conducted in the early 2000s is rather discouraging There is evidence that people are more flexible in how and when they spend money. But the boost to entrepreneurship that gets people out of the poverty trap is not there. Also, no effects on female empowerment visible in the experiments. See Reading: “where credit is due” 38  Example: The Miracle of Microfinance? Method Results Expansion of Spandana Microfinance Neighborhoods with a new Spandana in slums in Hyderabad, India branch had more new businesses and 52 of 104 slums were randomly more profitable existing businesses selected to receive a branch of than neighborhoods without Spandana in 2005 Spandana. Group loans exclusively to women. Shift on spending on more durable Only loan (no package with financial goods, less “temptation good”, such as literacy), and freedom to use the tobacco money as they see fit. No visible impact on total household Endline 12 to 18 months after spending or women empowerment treatment for 6850 households Banerjee, Abhijit, Esther Duflo, Rachel Glennerster, and Cynthia Kinnan. "The miracle of microfinance? Evidence from a randomized evaluation." American Economic Journal: Applied Economics 7, no. 1 (2015): 3922-53.  Example: Crop Price Insurance in Ghana Method Results Maize and Eggplant Producers in Take-up very similar in all groups, Ghana face price volatility. no clear treatment effect on Does crop price insurance increase borrowing their willingness to invest? Treatment group spends a bit more Treatments: on fertilizer Loan Loan plus insurance (for free) Effects are small and uncertain, Loan plus financial literacy training probably an underpowered Loan plus financial literacy training experiment plus insurance (for free) Some evidence that packages (loan 169 farmers (about 40 per plus insurance) are a good idea treatment) https://www.poverty-action.org/study/examining-effects-crop-price-insurance-farmers-ghana, Karlan, Dean, Edward Kutsoati, Margaret McMillan, and Christopher Urdy. "Crop Price Indemnified Loans for Farmers: A Pilot Experiment in Rural Ghana." The Journal of Risk and Insurance 78, no. 1 40 (2011): 37-55.  Example: Timing of Microinsurance Method Results Even though insurance seems a rational choice for poor farmers, they Large effect of the delayed tend not to take it premium payment. 5% choose Does a delayed payment of the insurance in Treatment 1, similar insurance premium at harvest time increase insurance take-up? in Treatment 2, but 74% in 605 sugarcane farmers in Kenya Treatment 3. Treatments: Farmers don’t have cash to spare 1 Standard offer at market rate 2 Discounted standard offer when they plant their fields. 3 Harvest deduction offer: market rate is Timing of payments can make a deducted from revenues at harvest time huge difference in practice. https://www.povertyactionlab.org/evaluation/agricultural-microinsurance-sugar-cane-farmers-kenya Casaburi, Lorenzo, and Jack Willis. 2018. “Time versus State in Insurance: Experimental Evidence from Contract Farming in Kenya.” American Economic Review 108 (12): 3778–3813. 41 https://doi.org/10.1257/aer.20171526. Example: Universal Basic Income (UBI) Method Results Does UBI encourage laziness or UBI in rural Kenya leads to economic improve the economic activities of growth with an increase in the low-income households? number of businesses, revenues, costs Rural communities in Kenya get short- and net income, especially in the non- term (ST), long-term (LT) and lump- agricultural sector. sum (LS) payments to cover their basic Labour supply remains similar, but needs. wage labour is replaced by self- Treatment employment. LT: 44 villages, 5’000 people, for 12 years. Communities receiving LS transfers ST: 80 villages, 8’800 people, for 2 years. experience economic growth LS: 71 villages, 8’800 people, in two comparable to or higher than those instalments. receiving LT transfers. ST payments lead to lower growth in total output. Banerjee, Abhijit, Michael Faye, Alan Krueger, Paul Niehaus, and Tavneet Suri. 2023. “Universal Basic Income: Short-Term Results from a Long-Term Experiment in Kenya.” 42  Can microfinance unlock the poverty trap for some entrepreneurs? Method Results Return to the Spandana data set Gung-ho show a strong increase used in 2005, evaluation of in business assets (25%), and longer term outcomes 6 years revenues (200%), as well as after treatment profits. Exploration of heterogeneous Their improvements dominate treatment effects: gung-ho the effect of the overall sample entrepreneurs (i.e. people who Access to microfinance must be already owned a business before paired with a talent for treatment) vs. reluctant/new entrepreneurship to facilitate an entrepreneurs escape from the poverty trap. https://voxdev.org/topic/finance/can-microfinance-unlock-poverty-trap-some-entrepreneurs 43 Banerjee, Abhijit, Emily Breza, Esther Duflo, and Cynthia Kinnan. 2019. “Can Microfinance Unlock a Poverty Trap for Some Entrepreneurs?” w26346. Cambridge, MA: National Bureau of Economic Research. https://doi.org/10.3386/w26346. Conclusions Microfinance has an appealing theory of change. But in practice, its effects are more limited than the theory suggests. One sized approach (small term loan) does not work for all. However, there is also no evidence of clear negative effects, and some benefits are there, like greater financial flexibility (while staying poor). Many experiments since then have focused on a variety of financial services and their specific terms, in order to find the right fit for the constituents. Transactions Savings Insurance Ultra poor programs Focus on heterogeneity: The same treatment may have different effects on sub-groups Microfinance has a valid theory of change, there is a basis to claim impact, but a lot of value lies in appropriate design. 44 Assignment Read 3 “Where credit is due” Take Quiz 45  Shareholder Engagement and Matching Session 3 Impact Investing (7.170), Fall Term 2024, University of St. Gallen Prof. Julian Kölbel In a nutshell: E.g. Where leaving is costly: job The Clash, 1982 “this song is soo good my neighbor threw a brick through my window to hear it better” https://www.youtube.com/watch?v=BN1WwnEDWAM 2 Learning Goals Evaluate shareholder engagement strategies Critically assess the impact of shareholder engagement Interpret a matching study 3 Exit, Voice, and Loyalty by Albert O. Hirschmann, 1970 Two fundamental responses to organizational decline Exit = economic action, binary, unclear about why Voice = political action, gradual, more specific about why Loyalty generally delays exit, favours voice, can be irrational Applied to: Emigration Employment Schools General considerations Cost of exit (and re-entry) Cost of voice (and gain from improvement) Combinations of voice and exit (e.g. threat of exit) 4  Example: Hermes engaging Sinopec on Climate Change Methane leakage form oil production causes high GHG emissions. Hermes initiated a dialog with the management of Sinopec on climate change. Sinopec launched a methane reduction program, saving 3 million tons of CO 2. ! 5 Shareholder Engagement: Requirements & Limitations See page 14, Heeb, Florian and Kölbel, Julian F. 2020. “The Investor’s Guide to Impact.” https://www.csp.uzh.ch/en/research/Academic-Research/Investor-Impact.html. 6 The tools of shareholder engagement Sending letters to voice a concern Meeting with management or board to discuss (behind closed doors) Sending an open letter (seek media attention) Voting on a shareholder resolution Filing a shareholder resolution Voting against Board members Successful engagements are often settled behind closed doors. 7 One of the most important initiatives: Climate Action 100+ https://www.climateaction100.org 8 Climate Action 100+ The three asks Escalation Steps Implement a strong governance framework on climate 1. Holding meetings with companies change; 2. Conducting investor roundtables Take action to reduce greenhouse gas emissions across the value chain and; 3. Asking a question at a company earnings call or Annual General Meeting (AGM) Provide enhanced corporate disclosure. 4. Making a statement at a company AGM 5. Writing a public letter to the company Focus 6. Supporting shareholder resolutions on climate change risks and opportunities The 160 largest CO2 emitting companies 7. Voting for the removal of directors who have failed in their accountability of climate change risk 8. Voting against reports, accounts and company led resolutions 9. Making joint statements with the company 9 Legal Background matters: US In the United States shareholder proposals are not binding Proposals need to pertain to corporate governance, not ordinary business (ordinary business operations exception) Trinity church filed a shareholder proposal at Walmart that the board should oversee the But shareholder votes on CSR sale of ”products that threaten public safety” – meaning assault weapons – it was not allowed still seem to have an effect for interfering in ordinary business (Flammer, 2015) https://www.forbes.com/sites/clareoconnor/2015/04/15/walmart-beats-out-church-in- court-over-gun-sales/ Flammer, Caroline. 2015. “Does Corporate Social Responsibility Lead to Superior Financial Performance? A Regression Discontin uity Approach.” Management Science 61 (11): 2549–68. 10 https://doi.org/10.1287/mnsc.2014.2038. Legal Background matters: Switzerland 3. März 2013: Minder Initiative passed with 68% yes votes (!!) “Die Generalversammlung stimmt jährlich über die Gesamtsumme aller Vergütungen (Geld und Wert der Sachleistungen) des Verwaltungsrates, der Geschäftsleitung und des Beirates ab.” Legally binding “say on pay” for all listed Swiss companies Other items, unrelated to compensation, are still non-binding Flammer, Caroline. 2015. “Does Corporate Social Responsibility Lead to Superior Financial Performance? A Regression Discontin uity Approach.” Management Science 61 (11): 2549–68. 11 https://doi.org/10.1287/mnsc.2014.2038. Does “say on pay” work? There is evidence that say-on-pay votes had an effect on compensation levels and the degree to which compensation is linked to performance. But there are also concerns: Shareholders may not have time to study compensation packages carefully. Risk is a focus on “headline” measures, such as total compensation, which are not optimal Given that votes are public, shareholders may be under pressure to vote in socially desirable ways, which can put appearance before substance. Edmans, Alex, Xavier Gabaix, and Dirk Jenter. 2017. “Chapter 7 - Executive Compensation: A Survey of Theory and Evidence.” In The Handbook of the Economics of Corporate Governance, edited by Benjamin E. Hermalin and Michael S. Weisbach, 1:383–539. https://doi.org/10.1016/bs.hecg.2017.11.010. 12 Say on Climate Lafarge Holcim has set a “Science-Based Climate Target” in 2020 Announced to publish a transition report in 2022 and submit it to a shareholder vote Approved with 89.85% in favour on 2022 AGM (and scheduled again for 2023) The idea is to get investor approval for an ambitious, long-term program The vote is non-binding Nestle held a say on climate vote in 2021 with 95% approval 13 Potential problems of Say on Climate companies putting forward unambitious transition plans; investors rubber stamping plans that are unfit to limit global warming to 1.5°C; possible limitation of companies’ climate ambition beyond what is set by the approved plan; reduction of investors’ influence in company engagements upon approval of a plan; investors diverting their focus and resources from more-targeted stewardship actions. Does it work? Empirical question that has not been answered (thesis topic) https://www.forbes.com/sites/bobeccles/2021/01/05/here-is-my-say-on-say-on-climate/?sh=3660fc205c49 14 Democratizing Shareholder Voting: Tumelo Interesting trend: digital tools allow retail investors to vote their own shares Likely to make shareholder votes more democratic But how much time do individual investors have to think about the issues? Podcast with Tumelo Founder Georgia Parker 15 The Matching Method 16 Success Rate of Engagements Kölbel, Julian F., Florian Heeb, Falko Paetzold, and Timo Busch. 2020. “Can Sustainable Investing Save the World? Reviewing the Mechanisms of Investor Impact.” Organization & Environment 33 (4): 554–74. https://doi.org/10.1177/1086026620919202. 17 Set-up 50 47 X: The engagement firm issues 45 engagement requests, such as: Please 40 report on climate risks following the 35 TCFD accounting conventions 30 25 Y: Does the firm issue a TCFD report? 20 15 A simple regression: Y = a + bX + e 10 7 5 0 What are the concerns? Engaged Not engaged % TCFD 18 Matching: Treatment is “random” within groups 50 47 Suppose firms in Europe have a 45 higher likelihood of issuing a TCFD 40 40 report 35 Engagement occurred more with EU 30 firms 25 But being a EU firm is the ONLY 20 variable that influenced targetting 15 Then, we get a valid estimate from: Y 10 5.4 4.4 = a + bX + yRegion + e 5 0 Because firms from the same region Engaged Not engaged are compared «within» that group. % TCFD EU % TCFD US 19 Matching generalized Often, there are many variables Zi that influence selection into treatment (size, industry, country, current ESG rating, profitability, brand value etc.) We can build a control group on a multivariate Z using “propensity score matching” Key assumption: selection into treatment is ONLY determined by Z, and Z is fully observable If we can find a group of firms that is equal with regard to Z, treatment is as good as random 20 The real effects Environmental Activist Investing Working paper, studies the impact of engagement on the level of toxic chemicals released Existing engagement program by the New York City Pension System Forms a matched control group, based on the targeting analysis to the right Predictors of being an engagement target Naaraayanan, S Lakshmi, Kunal Sachdeva, and Varun Sharma. 2022. “The Real Effects of Environmental Activist Investing.” Working Paper. https://doi.org/10.2139/ssrn.3483692. 21  The key table for any matching approach: equality of means 1. Are the means of relevant variables equal across the treatment and the matched group? 2. Are all the relevant variables reported? --> You are good to go The difference does not need to be zero, but it should not be statistically different from zero. This table should include standard deviations. 22 Naaraayanan, S Lakshmi, Kunal Sachdeva, and Varun Sharma. 2021. “The Real Effects of Environmental Activist Investing.” SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3483692. The real effects Environmental Activist Investing Release of toxic chemicals of the engaged companies is lower compared to the matched control Matching is combined with differences-in-differences, but the idea about creating a valid counterfactual remains the same The engagement activities of the NY caused a reduction in toxic chemicals release Naaraayanan, S Lakshmi, Kunal Sachdeva, and Varun Sharma. 2022. “The Real Effects of Environmental Activist Investing.” Working Paper. https://doi.org/10.2139/ssrn.3483692. 23  Some key findings based on matching Engagement is more likely to succeed, when the engaging investors holds a large fraction (or the engagement is collaborative with other shareholders that together hold a large fraction) when the lead engager is culturally close to the engaged firm when the engager itself is a large and/or reputable institution When the requested change is of a low-cost nature, i.e. reporting of existing information, rather than changing the production process Engagement leads to (Small) Reductions in emissions (Naraayanan et al., 2022) (Small) Improvements in ESG scores (Barko et al., 2021) Barko, Tamas, Martijn Cremers, and Luc Renneboog. 2021. “Shareholder Engagement on Environmental, Social, and Governance Performance.” Journal of Business Ethics https://doi.org/10.1007/s10551-021-04850-z. Dimson, Elroy, Oğuzhan Karakaş, and Xi Li. 2015. “Active Ownership.” Review of Financial Studies 28 (12): 3225–68. https://doi.org/10.1093/rfs/hhv044. Dimson, Elroy, Oğuzhan Karakaş, and Xi Li. 2018. “Coordinated Engagements.” Working Paper. https://papers.ssrn.com/abstract=3209072. 24 Do you think there could still be selection bias in studies that match firms on size, industry, country? 25 An engagement experiment 26 The Problem Engagement Request Change in firm behavior Confounder: Overall prominence of the topic Scientific concern: does engagement really make a difference? Practical concern: how can I identify good engagement services and how much should I pay for them? 27 Our solution: random assignment 0.20 Fraction of firms with SBT commitment 0.15 Group Engaged: No 0.10 Engaged: Yes 0.05 0.00 Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Month (2023) 2022 2023 28 Treatment Letter from index provider*’s CEO to the company’s chairman Ask: Commit to an SBT with the Science-based target initiative Threat of exit: SBT status affects constituent weights in Climate indices Mentions that the index provider itself is setting a target Treatment period 2022-02-01 to 2023-01-31 The experiment was pre-registered: https://www.socialscienceregistry.org/trials/8924 *anonymous by request 29 Outcome SBT Status provide by Science-based target initiative (SBTi) online https://sciencebasedtargets.org/companies-taking-action/ Public registry, independent of the companies and the investor We count both “Targets set” and ”Committed” as a successful outcome (=1) If the company is NOT in the registry, the outcome is unsuccessful (=0) Companies that commit and do not follow through remain listed as “commitment removed” 30 Example: TUI Group Total CO2 Emissions 2021: about 7 million tCO2eq (sample average 1.5 million tCO2eq) Homepage SBTI Filing Committed: 2022-7-22 Targets verified: 2023-3-01 TUI commits to reduce absolute scope 1 and 2 GHG emissions from hotels 46.2% by FY2030 from a FY2019 base year. TUI commits to reduce scope 1 and scope 3 well to wake emissions from jet fuel 24% per revenue passenger kilometer by FY2030 from a FY2019 base year. TUI commits to reduce absolute scope 1 and scope 3 well to wake emissions from marine cruise fuel 27.5% by FY2030 from a FY2019 base year. 31 Sampling Target Population: 1227 international corporations Starting from 1964 firms in any of the climate-related indices, as of 2021-12- 20 Science-based Target (SBT): NO (as of 2022-01-01) Engaged by Climate 100+: NO (as of December 2021) We selected 300 randomly for treatment in a blocked random design Blocking on 27 countries (p=0.239) Why? Bolton & Kacpercyk (2022) find country affects targets 32 Results Fraction of firms with Science−based climate Target 0.20 Control Treatment 0.15 SBT No 781 237 0.10 SBT Yes 146 63 Sum 927 300 0.05 Fraction Yes 15.7% 21.0% 0.00 control treatment Pearson’s Chi-Square Test: 4.42, p=0.036** 33  Commitment vs. “Targets Set” Fraction of firms with Science−based climate Target The treatment effect is driven by 0.20 targets set committed commitments to set targets, not targets themselves. 0.15 This is in line with the treatment (request to commit). 0.10 Future work: treatment effect on targets 24 months later 0.05 0.00 control treatment 34 Implications 1. Engagement by financial institutions can affect corporate policies when a feasible request is combined with a credible threat of exit. 2. Financial institutions that already apply ESG screens should proactively inform companies about these screens. 3. The EU CTB/PAB index regulation provides a basis for effective shareholder engagement. 35 A broader strategy: NBIM https://www.nbim.no/contentassets/2a7c78b9185b4a21986b09f85b854e81/2025-climate-action-plan.pdf 36 Assignment Until next week: Read Engine No. 1 Case Essay Question 37  Private Equity and the Additionality of Capital Session 4 Impact Investing (7.170), Fall Term 2024, University of St. Gallen Prof. Julian Kölbel What is the limiting factor?  Environmental benefits of plant-based meat are well established MP20 Scenario = 20% of per capital beef consumption is replaced with Microbial Protein (lab-grown meat) Percentages indicate the relative change of emissions of the beef food production system, compared to MP0 globally until 2050 Humpenöder, F. et al. Projected environmental benefits of replacing beef with microbial protein. Nature 605, 90–96 (2022). Planted Foods ETH spin-off founded in 2019 125 Million CHF raised (another 70mn on Sept. 1, 2022) Co-lead of Series A: Blue Horizon Venture Yann Sommer is involved The market for alternative protein is growing Products in retailer shelves (Coop, Migros, Lidl) These guys are heros! How much of this success is due to IMPACT investors? https://www.crunchbase.com/organization/planted-foods How is planted doing? Source: Pitchbook Enter: Blue Horizon Zurich-based private equity General Partner (GP) with a focus on food tech Seed, Venture, Growth, and Platform Development funds Participated in several rounds of Planted fundraising 9.2.2021 – Series A (amount undisclosed) 8.3.2021 – Series A (17mn with others) … Blue Horion Overview Source: Pitchbook  What about Impact? “We underwrite all our investments with our Dual Mandate in mind, targeting top-tier financial and impact returns. Therefore, the aggregated impact of our portfolio companies makes up the impact of our capital.” I would say: Owning impact is not the same thing as causing impact. For the outcome it does not matter who provided the capital. Having said that, there are some things that BlueHorizon does really well. The potential investees must have net positive impact Underlying transformation thesis is convincing Convincing case that sector expertise contributes to investees success Distribution, expertise, and mission support Impact Report: https://bluehorizon.com/wp-content/uploads/2022/08/BH_Annual_Impact_ESG_Report.pdf What about governance? Unfortunately, Blue Horizon is not developing well: IPO was canceled Limited partners forced take-over of the first (and only) PE fund Second fund has not closed 4 board members left Most employees left (or were let go) https://www.tippinpoint.ch/artikel/76942/umstrittener_vegan-unternehmer_enttaeuscht_schweizer_investoren-elite.html Recap: Differentiation of Impact Investors Not for profit For profit Investor Investee Deep dive investors I/III Impact Focus: Focusing on absolute poverty (>2USD/day) Investing in the development of companies and organizations in developing countries whose activities provide poor people with new or additional sources of income Investment Strategy: Financing start-ups and growth phase (investing horizon up to 10 years) Investing predominantly in the form of shareholding and loans (fund providers are high net worth individuals) Impact Measurement, Reporting and Communication: Evaluating the organization and business model of the company in which elea plans to invest, risk factors, and the additional benefit (“elea Impact Measurement Methodology”) Quantifying the impact created by measuring the number of people positively impacted and the amount of money saved or gained per person impacted by elea's invested enterprise Innovation and Scalability: Focusing on the creation of public added value as a tax-exempt charitable foundation Dealing with limited scalability due to high investment costs, time commitment, and limited funding availability Deep dive investors II/III Impact Focus: Focusing on the food system: Sustainable food products, sustainable food production, sustainable food packaging and delivery, and business models that address food waste and recycling. Investing in the transition to a sustainable food system that delivers both financial and impact return Investment Strategy: Focusing on early-stage companies and helping to scale the companies Investing predominantly in the form of seed or Series A investments in the form of shareholdings Impact Measurement, Reporting and Communication: Aiming to transform the Sustainable Food System Extending beyond consumer food products, ranging from better crop practices to sustainable packaging and innovative distribution. Innovation and Scalability: Trying to scale each company AND the market as a whole Deep dive investors III/III Impact Focus: Focusing on companies driving measurable social and environmental impact alongside business performance and strong returns: The Rise Funds, TPG Rise Climate, and the Evercare Health Fund Investment Strategy: Investing their funds in growth-stage, high potential, mission-driven companies Offering investment resources, business-building skills and a global network Managing $18 billion of assets across The Rise Funds, TPG Rise Climate, and the Evercare Health Fund Impact Measurement, Reporting and Communication : Deploying Y Analytics, a public benefit corporation dedicated to helping capital allocators better understand, value and manage social and environmental impact Partnering with Y Analytics and Bridgespan Social Impact, The Rise Fund created a methodology to assess potential impact (Impact Multiple of Money (IMM)) Innovation and Scalability: Later stage compared to Blue Horizon No focus on investor impact Deep dive investees I/III Business model: Hasiru Dala provides a livelihood for informal waste pickers and waste workers in India by integrating them into the city's formal waste management system, creating a positive environmental impact. Aiming for-purpose, not-for-loss Investment Characteristics: Ability to lift people out of poverty and offering improved livelihoods Profitability of business model in line with impact model – Little risk of mission drift Target Sector Impact: Informal waste pickers in Bangalore, India Business: Total waste management, event waste management, plastic collection for recycling Deep dive investees II/III Business model: Planted specializes in developing and producing plant-based meat alternatives with a focus on creating sustainable and flavorful food products. For-profit oriented business model Investment Characteristics: Impact inherent in business model – little risk of mission drift Target Sector Impact: Reduce the environmental impact of the traditional meat industry Business: Alternative meat industry Deep dive investees III/III Business model: Dodla Dairy sources, processes and distributes a variety of dairy products, while increasing the income of India's rural dairy farmers by enabling them to access lower-risk, higher-priced milk for sale For-profit oriented business model Investment Characteristics: Profitability could affect mission focus of business model – high risk of mission drift Target Sector Impact: Improving livelihoods of rural farmers Business: Dairy industry in India Bigger picture Many PE funds focus on industries such as health tech, education tech, agri-tech, real estate It is relatively easy to say that investments in these industries have positive impact (that is, company impact) The real question: How would these sectors look like without impact funds, and just with “normal PE funds”. Additionality of Capital Sufficient vs. necessary support Markets are competitive Additionality of capital hinges on limits to arbitrage/inefficiencies Ideal: overlooked but underfunded It is easier to claim investor impact with sector expertise than with capital Capital Sector Expertise Markets are competitive. When they are Whereas investments are the same efficient, good entrepreneurs will find funding. from any investor, expertise You can do concessionary returns depends on who provides it Give an explicit subsidy to the Takes time to build up, hard to copy. entrepreneur at your own loss It makes a difference whether Operate in inefficient, high-friction company is funded by fund A or B. markets, where it is costly to find good entrepreneurs (overlooked but underfunded) Paired with a credible impact mission, expertise can enhance both The fact that returns are concessionary does returns and impact not guarantee additional growth For example, non-concessionary prior Providing a lot of expert resources, investors might soak it up as extra return however, may also be a form of In my experience very rare concessionary return (higher fees). Mission Drift The idea that a start-up with a mission to save the world over time replaces that mission with simply making profits Often associated with replacement of founders PE investors can play an important role in accelerating or avoiding mission drift (this is basically engagement) The conundrum In order to have impact, you probably have to give something up (return expectations, higher cost, provide expertise). Extreme form is simply a donation But giving up returns does not guarantee impact, so you need to measure it somehow Impact Measurement and Management Take-away: There is not one accepted accounting standard for impact. The state of impact management and measurement in practice. Second Edition. The GIIN. 2020.  Thinking about company impact This is one possible framework by the Impact Management Project A guide to classifying the impact of an investment. The impact management project. 2021. Thinking about portfolio impact (IMP) The main point: it makes sense to think of company impact and investor contribution (investor impact) as two different dimensions. This tool is quite useful in practice, because you can sort your portfolio along these categories  The Impact Efficient Frontier McCreless, Mike. 2017. “Toward the Efficient Impact Frontier.” Stanford Social Innovation Review.  No data, no deal? Proposition: there may be a trade-off between measurability and impact magnitude Impact Management & Measurement (IMM) is a developing Healthcare practice Systems Software Impact Uncertainty Often, investors want to document Innovation impact ex-post, for accountability For investment decisions, however, Selling Malaria we need ex-ante estimates of impact Bednets Lack of credible ex-post data, or noisy ex-post indicators, can Expected Impact Magnitude sometimes hinder promising impact investments Source: No data, no deal? By George Carew-Jones and Alex Money. Second order effects 10 investors invest in 10 startups that do similar things. 9 go bust, but one becomes a unicorn. Who had impact? Potentially, investor appetite for alternative protein start-ups has motivated entrepreneurs to experiment all around the world with technologies and business models Even a loss-making investment might contribute to the development of the sector as a whole Assignment Please prepare a PowerPoint presentation with 3-5 slides in which you present an investment vehicle (e.g. a fund, an impact investment company, an ETF, a bank, etc.) that makes an impact claim: What is the Name, ISIN (if available), and URL of the vehicle? What claim does the vehicle make? Is it the impact of the company or the investors? What mechanism(s) does the vehicle use? Does the vehicle formulate a theory of change? What type of reporting does the vehicle provide? Overall, how would you rate the impact potential of this vehicle? Would you recommend this vehicle to an impact investor? To help you get started you can find a list for an investment vehicle here. Deadline: Monday, Dec. 16. You are invited to present during the last lecture (when there are too many submissions, we might make a selection).  Special Instruments Session 5 Impact Investing (7.170), Fall Term 2024, University of St. Gallen Prof. Julian Kölbel Overview Green Bonds Sustainability-Linked Bonds Litigation Finance Blended Finance Green Bonds  Green Bond Market Green Bond issuance per year (USD bn) For reference: total bond issuance is about 20 tn USD per year  Ford Motor USD 2.5bn green bond (Nov. ‘22) Proceeds of the bond go into electric vehicle (EV) production Largest green bond issue in the US at the time Ford plans to spend bn 30 USD on EV until 2025 https://www.wsj.com/articles/ford-motor-offers-2-5-billion-green-bond-to-investors-11636413890 Green Bonds “Green Bonds are any type of bond instrument where the proceeds or an equivalent amount will be exclusively applied to finance or re-finance, in part or in full, new and/or existing eligible Green Projects” Key components to specify: 1. Use of Proceeds clause in the bond contract (proceeds will only finance or re-finance ”green” projects) 2. Process for Project Evaluation and Selection (which projects are ”green”?) 3. Reporting how the proceeds were used https://www.icmagroup.org/assets/documents/Sustainable-finance/2021-updates/Green-Bond-Principles-June-2021-100621.pdf Use of proceeds reporting Once the bond is issued, the issuer must demonstrate a balance of financed project to match the borrowed funds The projects do not need to be new, proceeds can be used to re-finance existing projects For instance, from co- investments in railway infrastructure, savings of 15 million t CO2 p.a. are expected. Projects are not collateral for the bond https://www.deutsche-finanzagentur.de/fileadmin/user_upload/Institutionelle-investoren/green/reports/Green_bond_impact_report_2020.pdf https://www.deutsche-finanzagentur.de/fileadmin/user_upload/Institutionelle-investoren/green/reports/GreenBondAllocationReport_2020_en.pdf Green bonds in the balance sheet Assets Liabilities Plant Property & Equipment Bonds Green Assets Green bonds Cash Equity Etc. Etc. Green bonds need to be balanced by green assets Is there a Greenium? Twin Bunds Two sovereign bunds that are nearly the same in their characteristics (except size) The green twin has a use of proceeds clause Allows to determine the ”greenium” and a yield curve for green EUR securities https://www.deutsche-finanzagentur.de/en/federal-securities/types-of-federal-securities/green-federal-securities/twin-bond-concept (including a slick video with beautiful German Accent)  The Greenium https://www.deutsche-finanzagentur.de/en/federal-securities/factsheet/isin/DE0001030716?cHash=b427cab3cc3a5f84f936a4f23ecbbfb7  Do green bonds have impact? Green bonds often refinance existing green projects, The greenium is small comparing to other factors or finance projects that were already planned determining the cost of capital https://2degrees-investing.org/wp-content/uploads/2020/01/2018-Green-bonds-updated-paper.pdf  Do green bonds have impact? Using a matching approach, Flammer 2021 finds that firms which issue green bonds improve their ESG rating (Refinitiv) reduce carbon emissions Best explanation: green bonds are a signalling device Setting up, and certifying a green bond is costly This costly signal is a way to convince investors that you are serious about going green The greenium seems to be very small for corporates, if it exists at all Flammer, Caroline. 2021. “Corporate Green Bonds.” Journal of Financial Economics (Forthcoming). https://doi.org/10.1016/j.jfineco.2021.01.010. Zerbib, Olivier David. 2019. “The Effect of Pro-Environmental Preferences on Bond Prices: Evidence from Green Bonds.” Journal of Banking & Finance 98: 39–60. https://doi.org/10.1016/j.jbankfin.2018.10.012. Larcker, David F., and Edward M. Watts. 2020. “Where’s the Greenium?” Journal of Accounting and Economics 69 (2): 101312. https://doi.org/10.1016/j.jacceco.2020.101312. Sustainability-Linked Bonds What are Sustainability-Linked Bonds (SLBs)? Two ingredients: SPT: a predetermined sustainability performance target with a date ESG rating level, CO2 emissions, number of female board members, etc. Coupon ratchet: the bonds coupon changes conditional on target status Step-up when target is not reached Step-down when it is reached (less common) SLB Market growth, regional and sector breakdown By year By GICS sector By region Kölbel, Julian F, and Adrien-Paul Lambillon. 2022. “Who Pays for Sustainability? An Analysis of Sustainability-Linked Bonds.” Working Paper. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4007629.  Example: Holcim SLB Sustainability Performance Target (SPT): 475 kilograms of net carbon dioxide per ton of cementitious material per Key Figures: reporting year Holcim Ltd Chart Title (Swiss building materials producer) 700 600 Notional: EUR 850 million 500 Issue Date: 23 November 2020 400 300 Maturity Date: 23 April 2031 200 Coupon: 0.5% p.a. 100 0 Coupon step-up: 0.75% p.a. SPT Date: 31 December 2030 Potential penalty: ~EUR 2.125 million (Notional*step-up*time to maturity, i.e. 850mn * 0.75% * 1/3 year) https://www.holcim.com/sites/holcim/files/documents/lafargeholcim_sustainability-linked_notes_prospectus_19_nov_2020.pdf Do SLBs have impact? Remains to be seen – most of the targets lie in the future. From the corporate side, there is a clear incentive (even though it might be small). Investors sacrifice returns for an agreed-upon impact, or get their money back. The target needs to be ambitious, i.e. beyond what the firm would have done anyways. There are second-party opinions on this matter.  Are SLBs relevant? Three conditions Size of coupon adjustment (e.g. Size of coupon adjustment 25bp) Remaining time to maturity after the adjustment (e.g. 1 year) Ambition level of the target (e.g. reach carbon neutrality by 2030) Remaining time to maturity If one of the three conditions is close to zero, the SLB contract becomes irrelevant. The SLB Premium Based on a matched sample (SLBs versus regular bonds issued by the same company, similar maturity and size) we find an average premium of about 13 bp for the yield on SLBs. The maximum penalty (step-up coupon * remaining maturity/maturity) is on average 10bp It sees that SLB investors “like” SLBs for reasons other than the extra coupon they might gain. Kölbel, Julian F, and Adrien-Paul Lambillon. 2022. “Who Pays for Sustainability? An Analysis of Sustainability-Linked Bonds.” Working Paper. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4007629. SLBs in practice The CFO of Axpo told me that the company first decided on their renewable energy strategy, and then issued SLBs on that basis. In practice, SLBs are also a signaling device, but one that is more oriented to the future and more suited to “hard-to-abate” sectors. Litigation Finance Litigation Finance Investor provides funds to claimant to pursue a legal interest After subtracting costs for lawyers, investor receive a share of the payout Can be efficient, as claimant may not have the financial resources to pursue claim, even if the case is won Litigation Finance https://www.europarl.europa.eu/RegData/etudes/STUD/2021/662612/EPRS_STU(2021)662612_EN.pdf Impact and litigation finance 15,000 Indonesian seaweed farmers are suing an Australian oil company for allegedly causing an oil spill that destroyed their livelihoods. Litigation fund provides funding to register and bring a class action suit on behalf of these farmers If the case is successful, it might generate tens of millions AUD in compensation (part of which will go to the fund) Immediate impact is on the seaweed farmers. They receive compensation that they may not have had received without the investment. In addition, the company looses resources that might otherwise have had destructive effects elsewhere. There could be an even wider impact by setting a legal precedent, that could serve to avoid future destruction of livelihoods, by other companies. One of a few funds to pursue this is https://www.aristata.co.uk Obviously, the devil is in the details with such deals!!! Blended Finance Blended finance Blended finance is the strategic use of development finance for the mobilisation of additional finance towards sustainable development in developing countries. https://issuu.com/oecd.publishing/docs/amounts-mobilsed-from-the-private-sector-by-dev-fi Blended finance is about structuring https://www.usaid.gov/sites/default/files/documents/1865/BlendedFinanceStarterKit1.pdf The Allianz SDG Loan Fund Mobilization rate 1:9 FMO (Dutch Development Bank) originates loans to developing markets firms in agribusiness, financials, and renewable energy. Fund participates in 100-120 loans from the balance sheet. FMO IM is an independent manager. Proceeds from the loans go to Class A shares first, then to Class B. The 25 million loan guarantee by MacArthur allows FMO to bear greater risk on their 111 million share in the fund. Expected impact: 60k supported jobs and 450kt avoided carbon emissions. SFDR Article 8 Fund More Details: https://www.convergence.finance/resource/SDG-Loan-Fund/view Challenges One big issue for blended finance is that it development banks may be seen as subsidising private investors Another challenge is that development banks may crowd out private investors Nonetheless, the concept holds promise and offers endless opportunities for smart financial engineers…  Features Toward the Efficient Impact Frontier By Michael McCreless Stanford Social Innovation Review Winter 2017 Copyright  2016 by Leland Stanford Jr. University All Rights Reserved Stanford Social Innovation Review www.ssir.org Email: [email protected]  Stanford Social Innovation Review / Winter 2017 49 who join Furaha, by contrast, not only gain a route to a At Root Capital, leaders are using tools from , mainstream financial analysis to calibrate the safer and more reliable market, but also receive a price premium that Furaha has negotiated with foreign cof- role that subsidies play in their investing practice. fee buyers. In addition, Furaha provides clean water and electricity to farmers. Unlike the loans to UCC and Toward the GADC, Root Capital’s loan to Furaha required a subsidy: The cost of lending to the cooperative was greater than the interest that it would pay to Root Capital. For Tugume, building a high-impact, financially Efficient sustainable loan portfolio requires a delicate balanc- ing act. “Each year, I try to make five or six big loans to large, well-established businesses,” he says. “These loans provide revenue to Root Capital, and the busi- nesses meet our social and environmental criteria: They purchase crops from local farmers and often pro- Impact vide services like agronomic training and farm inputs. Then, in the rest of my portfolio, I make much smaller loans to earlier-stage businesses that have a harder time getting loans but show potential for growth.” Tugume, in short, has developed an intuitive ­approach to creating a portfolio that generates both Frontier impact and revenue. His intuition is a powerful tool— but intuition is fallible, and it isn’t scalable. To achieve impact on a large scale and to do so efficiently, he and other portfolio managers need analytical tools that will equip them to make lending decisions in a quan- titative and holistic way. My colleagues and I at Root Capital are implementing such tools, and we believe that other practitioners in the broad field of impact By Michael McCreless investing may find them useful. DEVELOPING A TOOL KIT R At Root Capital, we practice a specific kind of impact investing. We operate in a space between traditional ichard Tugume is a portfolio manager at philanthropy (in which donors expect no return on grants that they Root Capital, a nonprofit lender to agricultural make) and mainstream financial markets (in which investors expect ­enterprises that connect small-scale ­farmers in market-rate returns). We draw on both grant funding and private Africa and Latin America to ­markets for their capital, and our fund is “concessionary,” in that our investors forgo crops. His portfolio includes about 20 loans the chance to earn the highest possible financial return on their to businesses in Uganda and the ­Democratic money. Most of our loans fall into one of two categories: Either they Republic of Congo (DRC). Those loans, like yield a negative financial return to Root Capital and thus require a all loans made by Root ­Capital, are designed subsidy, or they yield a positive but below-market return. Critical to to enable borrowers to create positive social our investment strategy is our ability to subsidize loans in the first or environmental impact. category with revenue from loans in the second category. We also In 2015, Tugume’s portfolio included large borrowers such rely on grant funding to support the subsidized loans. as Uganda Cocoa & Commodities (UCC) and Gulu Agricultural But are we allocating our philanthropic and investment funding ­Development Company (GADC). UCC connects more than 5,000 in the best possible way? For most of our history, we have not had cocoa farmers to export markets. GADC sources cotton, sesame, a way to answer that question with a high degree of rigor. In that chilies, and sunflower from 60,000 smallholder farmers in Gulu, regard, we are not alone. Until recently, intuitive methods have had a district in Uganda that is recovering from 25 years of conflict. to suffice as impact investors’ primary means of balancing financial One of the smaller businesses in Tugume’s portfolio in 2015 was return with impact. From rural Uganda to Wall Street, investors Furaha, a coffee cooperative in war-torn eastern DRC. Back in 2013, have lacked tools that quantitatively and comprehensively take into when Tugume first conducted due diligence on the company, he had account all of the factors—financial, social, and environmental— to plan his visits carefully to avoid local militias that were active in the that define the performance of both individual investments and in- region. Many Congolese farmers had no option but to smuggle coffee vestment portfolios. As a result, the process of allocating capital to into Rwanda, where they bartered it for food and supplies. Farmers achieve impact has been inefficient at best and inaccurate at worst. 50 Stanford Social Innovation Review / Winter 2017 We set out to create a tool kit that would support decision ­making Michael McCRELESS is senior director of strategy and impact at Root Capital. at the level of either a single investment or an entire portfolio. To do so, we first had to develop a way to integrate data on the finan- cial, social, and environmental (FSE) performance of our loans. We needed a way to view FSE data as part of a single picture— a way to analyze how different FSE goals relate to each other and to result, the full cost of serving them is high. Yet few of them are able identify where trade-offs between impact goals and financial goals to bear that cost: The interest payments on an unsubsidized loan might be necessary. would cripple their operations. We therefore charge interest rates A breakthrough came when we plotted our loans on a graph that and fees that are on par with those of local commercial banks, and measures expected return on one axis and expected impact on the we seek philanthropic support to cover the rest of our expenses. other axis. Determining the expected return of each loan was a rela- Consider the loan that Root Capital made to Furaha in 2015. It tively straightforward task. But to measure expected impact, we had to was a one-year loan for $75,000, with interest and fees that came develop a tool that synthesizes data on the social and environmental to 11 percent of that sum. The loan would thus generate $8,250 in performance of each borrower—together with an estimation of the revenue for us. (I have rounded and simplified some of the numbers investment impact of each loan—into a single metric. By joining that in this and other examples.) Given the early stage of this business metric to our financial return metric on a standard graph, we were and its hard-to-reach location, the fully loaded cost to underwrite able to generate an integrated picture of FSE performance. and monitor the loan was about $23,600. (As Root Capital grows and In working to make sense of those data, we have found inspira- achieves economies of scale, these operational costs are declining.) tion in concepts used by investors in mainstream capital markets. The sum that we pay to investors for use of their capital typically One especially powerful concept is that of an efficient frontier. Simply comes to 2.5 percent of the loan amount; in the case of the Furaha put, a portfolio that lies on the efficient frontier offers the greatest loan, that cost was $1,875. Add in a risk premium of $6,300, and the possible return for a given level of risk and for a given set of invest- total cost of this loan was $31,775. Even if Furaha were to repay the ment opportunities. We broadened this concept so that it would loan in full, in other words, that revenue would cover only about encompass not just risk and return, but also impact. one-third of our costs. The expected “return” on the loan, there- A portfolio of investments that lies on what we call the “efficient fore, was a net loss (or subsidy) of $23,525. impact frontier” would offer the highest level of overall impact, We funded more than half of that loss ($13,025) through a ­relative to the cumulative financial return of those investments. cross-subsidy from larger, more profitable loans, and we drew on (In the case of Root Capital, the return on our portfolio is usually ­philanthropic support to fund the remainder ($10,500). So for this negative; it takes the form of a required subsidy.) The idea of apply- investment in Furaha, the value that Root Capital, our investors, ing the “efficient frontier” concept to impact investing isn’t new. and our donors create lies in transforming a $10,500 “grant” into But we are putting this idea into practice. a $75,000 loan. By adopting and adapting concepts like the efficient frontier, Our loans to GADC and UCC, by contrast, had more favorable ­impact investors can make better-informed decisions about indi- expected returns. We expected GADC to borrow $1.1 million and to vidual investments, set more comprehensive goals for their port- pay $109,800 in interest and fees. We estimated our operational costs folio as a whole, and col