Prepared with A Sandbox for Innovation: My Exploration of Pay for Success in K-12 Education
Advisor Social Finance
Start Date 2018-00-00
Notes A Sandbox for Innovation: My Exploration of Pay for Success in K-12 Education The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters Citable link Terms of Use This article was downloaded from Harvard University’s DASH repository, and is made available under the terms and conditions applicable to Other Posted Material, as set forth at http:// use#LAA  A Sandbox for Innovation: My Exploration of Pay for Success in K-12 Education Doctor of Education Leadership (Ed.L.D.) Capstone Submitted by Teresa Anne Knickman Plancher To the Harvard Graduate School of Education in partial fulfillment of the requirements for the degree of Doctor of Education Leadership. April 2018 1 For my dad, whose unwavering belief in the importance of a doctorate led me to this work. Thank you for teaching me that data-driven approaches are not reserved for companies (you can use them in your life too!). For my mom, whose unwavering passion for K-12 education also led me to this work. Thank you for teaching me that data is only useful to the extent that it is improving the world. For my husband, whose unwavering support led me to complete this work. Thank you for teaching me that data is irrelevant in questions of love, the pursuit of gotchu-ing, and the purchasing of puppies. 2 Acknowledgements To Social Finance: Thank you for joining me in this exploration, and for the amazing year of learning, impact, and friendship. Tracy, Jeff, Lara, and Navjeet, thank you for taking a bet on having a resident, and for your leadership and guidance throughout the year. All of my project teams, thank you for being my partners in crime. Without your willingness to roll up your sleeves with me, your incredible insights, and your deep commitment to impact, this Capstone and the work behind it would not have been possible To my Capstone Committee: Thank you for your insightful feedback and guidance throughout this Capstone process. Monica, I am eternally grateful for all of your support as my advisor this year, and throughout the Ed.L.D. program. You have pushed my thinking and helped me grow academically, professionally, and personally. Thank you for your counsel and for your friendship. Dutch, I knew that taking your class would lead to great things! Thank you for introducing me to Social Finance, to the strategy audit framework (which has become a guiding principle in my approach to all things), and to an expanded view of what it means to be social enterprise. Jeff, you have been a constant sounding board for my sometimes wonky ideas and have supported me ferociously throughout the past year (even in Oklahoma...). Thank you for being the ultimate boss, and for helping me continue to grow as a leader and thinker through this work. To C6: Thank you for sharing this journey with me! I have learned more than I can express from each and every one of you about passion, persistence, and the importance of being a warrior for change. I can’t wait to see all of “the good” that you will lead in your lives. Bachelor crew, thank you for the moments of pause and frivolity amidst the crazy, and for making Mondays my favorite day of the week. KEO and RJ, thank you for everything, but in particular, for not judging my skorts. 3 Table of Contents I. Abstract............................................................................................................................ 5 II. Introduction .................................................................................................................... 6 III. Review of Knowledge for Action ............................................................................... 11 The Field(s): PFS and Education .......................................................................... 11 The Firm: Organizational Context ........................................................................ 19 The Work: Managing Change............................................................................... 21 Theory of Change ................................................................................................. 25 IV. Description, Evidence, and Analysis of the Strategic Project .................................... 26 The What and How: Overview of the Strategic Project ....................................... 26 Evidence To-Date ................................................................................................. 42 The Why: Analysis of What Happened ................................................................ 50 V. Implications for Sector................................................................................................. 64 Implications for the Pay for Success Sector ......................................................... 64 Implications for the K-12 Education Sector ......................................................... 66 Cross-sector Implications...................................................................................... 71 VI. Implications for Site ................................................................................................... 75 Social Finance in Education.................................................................................. 75 Social Finance beyond Education ......................................................................... 84 VIII. Conclusion ............................................................................................................... 88 Bibliography ..................................................................................................................... 90 Appendix A. Evolution of “Intro to PFS” Materials......................................................... 93 Appendix B. Summary of Takeaways from February CTE Project Convening............... 94 Appendix C. PFS in Education Working Session: Key Takeaways ................................. 96 Appendix D. Summary of Responses from “Formal” Data Collection Efforts ................ 99 4 I. Abstract Pay for Success is an innovative financing tool through which private investors provide upfront capital for high quality, evidence-based social services programs, and are paid back by government once pre-determined outcomes are reached. This tool enables governments to explore new programs without risk, and connects dollars directly to outcomes, rather than the delivery of services. Social Finance, a Boston-based nonprofit, launched the Pay for Success field in the U.S. in 2011, and is a leading practitioner in developing and implementing Pay for Success projects. Although 20 projects have launched in the U.S., none have been in the field of K-12 education due to the complexity of the tool and challenges associated with applying it in education contexts. This Capstone is focused on the strategic project that I led as a resident at Social Finance, through which I sought to assess the relevance of the Pay for Success model in the K-12 education sector, and pursue adaptations of the model to increase its potential for impact in the space. The foundation of my work was a set of six education projects that Social Finance launched in the second half of 2017, each intended to assess the feasibility of Pay for Success to support an education service provider’s K-12 intervention. My action plan for adaptation, which required both mindset and model changes to Social Finance’s typical approach, was grounded in change management literature and centered on engaging across the six specific projects to understand and address the barriers that Pay for Success presented across the projects’ education contexts. Through my work on and across these six projects, I found that Pay for Success, as it is traditionally defined, does have narrow relevance in the K-12 education space. The two major challenges that we identified included a lack of rigorous evidence for particular interventions, and a fundamental disconnect between stakeholders who pay for and implement education and those who benefit (or suffer) from the outcomes of education. To address these challenges, we developed four adapted models, each based on the traditional Pay for Success approach, but each with higher potential for impact in the education space. My strategic project has significant implications for Social Finance and for the Pay for Success field. By broadening the definition of Pay for Success beyond a single model, to include a range of outcomes-based financing tools such as the four adaptations that we explored in my work, there is immense opportunity to expand the reach, relevance, and impact of this innovative concept in education and beyond. 5 II. Introduction Pay for Success (PFS) is an innovative, outcomes-based financing tool that is intended to help high quality social services programs make the jump to scale. By using this tool, government jurisdictions are able to fund new programs without risking precious budget dollars if the programs fail to deliver results. The traditional PFS model, also called a “Social Impact Bond” (SIB), is a partnership between a jurisdiction, a service provider, and private investors. Instead of paying directly for services or programs, the jurisdiction defines the outcomes it is trying to improve – and how those outcomes will be measured. Private funders provide long-term, up-front working capital to support service providers in delivering the programs, and government only repays the upfront investment to the extent that programs achieve predetermined goals. Figure 1 depicts an example of a SIB focused on increasing post-secondary and career outcomes for immigrant and refugee populations in Massachusetts, providing a more detailed explanation of the key stakeholders and funding flows. My residency focused on the question of how PFS could be utilized in K-12 education to help high quality programs scale and improve long-term student outcomes. My residency site was a nonprofit called Social Finance, which brought the SIB – and the PFS field – to the U.S. from the U.K. in 2011, launching a market that has now enacted 20 deals across a variety of issue areas including homelessness, health, and criminal justice. Although Social Finance has been thinking about education since its conception, prior to 2017 they had not designed or implemented a PFS project in the sector; in fact, only two PFS projects have launched in U.S. education, both focused on Pre-K. My strategic project was focused on understanding why PFS has been slow to enter the education sector, and helping Social Finance think about whether and how they might expand in the space. 6 Figure 1. SIB Example: Massachusetts Pathways to Economic Advancement Project This figure depicts the stakeholders and funding flows involved in the traditional PFS model, the SIB. In this project, which was launched in 2017 by Social Finance, a group of 40 impact investors provided upfront capital to a service provider, Jewish Vocational Services (JVS). JVS used the capital to support 2,000 individuals through four program tracks, including English Language, employment, and post-secondary bridging skills trainings. The Commonwealth of Massachusetts agreed to pay back the impact investors – with a modest return – if pre- determined outcomes targets for educational attainment, employment status, and wages are met. Through this deal, the impact investors expect to benefit through a financial and social return on their investment, JVS benefits from a guaranteed pool of growth capital, and the Commonwealth of Massachusetts benefits by exploring the impact of these services on critical populations – without taking on the financial risk of the programs not working. Expansion capital ($) 40 Impact Investors Repayment ($) Intermediary Intervention Jewish Vocational Services Implementing services through four program tracks, in a tiered service offering 1. Rapid Employment 2. English for Advancement 3. Occupational Skills Training 4. Bridges to College Outcome Payor MA Executive Office for Administration and Finance as outcome payor, with project support from Office of Labor and Workforce Development and Executive Office of Education Evaluator Target Population 2,000 immigrants and refugees seeking to transition to employment, higher wage jobs, and higher education Social Finance’s mission is focused on “mobilizing capital towards social progress.” Education is one the most important levers towards social progress; having a good education is associated with higher future earnings, reduced criminal justice engagement, reduced reliance on social welfare, and many other indications of progress towards a fulfilling journey through life (Social Impact Research, 2012). Thus, any impact that Social Finance has in the education sector is catalytic in pursuing its broader social sector mission. 7 Social Finance is not alone in its desire to invest in education. Education is the second largest destination for philanthropic dollars1, receiving $60 billion annually from foundations, corporations, and individuals (Giving USA, 2016). This significant influx of funds is dwarfed by the amount of government resources dedicated to education; in 2013, the U.S. Department of Education (DOE) estimated that a total of $1.15 trillion was spent on education across the Federal, State, and local government levels (U.S. Department of Education, 2017). These vast amounts of resources highlight the value that we place on ensuring that US youth have access to high quality education, but they do not always lead to results that reflect those same values. In spite of the significant investment, a huge gap remains in the performance of the education sector relative to our aspirations. The US labor market increasingly requires post- secondary degrees: although required by only 28% of jobs in 1973, they will likely be required by 47% by 2020 (Georgetown Center on Education and the Workfoce, 2014). Meanwhile, only 53% of students who start high school enroll in post-secondary, and only 26% complete those degrees (National Center for Education Statistics, 2017). More problematic is the inequity of these outcomes for students of color and from low income families; only 27-36% of these underserved students enroll in post-secondary, and less than 15% earn these increasingly critical degrees. These outcomes have impact: students with post-secondary degrees have $1 million higher average lifetime earnings than those with only high school diplomas (and $1.4 million higher vs. no degree) (Hanley et al., 2012). Governments also feel the consequences of these gaps; adults who do not have a high school diploma are more than twice as likely to be eligible for food stamps, and eleven times more likely to receive housing 1 Charitable giving (in dollars) to education is topped only by giving to religion, which has been the largest destination of giving since 1970 when it was first reliably tracked. 8 subsidies than those who do, costing the U.S. government over $300,000 per high school drop out in lost taxes, welfare, transfers, and institutionalization expenses (Hanley et. al, 2012). Good solutions exist that support students in their journeys through K-12, post- secondary, and career education. For example, programs like Dual Enrollment and Early College High School enable students to access college experiences and credits, leading to better preparation and faster completion. Career Academies and internships expose students to career options and support them in building skillsets required in the labor market. Other programs such as Back on Track focus on ensuring that students stay in – or return to – education, decreasing negative outcomes such as absenteeism and suspension that pull students away from their learning. Many of these programs have strong histories of success – some demonstrated through rigorous Randomized Control Trials (RCTs) – in improving long- term outcomes such as post-secondary persistence, employment, and earnings. However, few of these programs operate at scale. For example, there are currently about 7,000 Career Academies, one of the most rigorously evaluated and wide-spread programs, serving 1 million high school students (only 6-7% of the public high school population) (National Career Academy Coalition, 2017). For the vast majority of students, education continues to look similar to the models employed decades ago; some districts are able to fund new programs, but most face tight budgets and minimal flexibility to test or implement new approaches. PFS could offer a way for students to experience these evidence-based programs without requiring districts to find upfront funding within their budgets. By using private capital to support initial implementation, schools, districts, and States can wait to ensure that student outcomes are improving from the programs before redirecting their limited resources. In this way, PFS funding could act as experimentation capital for education leaders, enabling 9 them to bring evidence-based innovations to student learning with limited financial risk. Education sector leaders have begun to see the potential value of PFS; in fact, PFS was explicitly named as an option for the deployment of Federal dollars in the 2015 Every Student Succeeds Act (ESSA). Thus, as I began my residency, the idea of using PFS in education was top of mind not just for Social Finance, but for the broader education community as well. During my residency, Social Finance launched its first project work in the public K- 12 education setting with four projects focused on Career and Technical Education (CTE) in secondary school, one on middle school programs to improve ninth grade readiness, and one on teacher retention across grade levels (of note, the four CTE projects were launched in a cohort model, as they were all supported by the same funding source: moving forward, I will refer to these as “the CTE projects”). These six projects formed the foundation of my strategic project; through them, I was able to explore how PFS might work across six different education interventions and contexts. Given the field’s delayed entrance to education, and some of the inherent challenges in applying PFS in the space (which I will discuss in the following Section), I approached the work with a lens of adaptation; as I will describe, the SIB is complex and specific in its application, and I suspected that Social Finance would need to adapt its typical approach in order to be successful in the sector. Further, although my primary goal was to answer the question of how Social Finance should approach the space, there was a more fundamental, underlying question that remained at the top of my mind throughout the project: is education a sector that Social Finance should aggressively pursue moving forward? In the following sections, I describe and analyze the outcomes of my project. In Section III, I review context on the PFS field (and its history in education), Social Finance as 10 an organization, and the academic literature on adaptive change management. These contexts lead to my theory of action on how to help Social Finance consider how – and whether – to adapt its model in education. In Section IV, I describe the strategic project in more detail, including a review of the problem of practice, an overview of my approach, and an analysis of what happened. Finally, in Sections V, VI, and VII, I discuss the implications of my project on two critical questions: First, how does PFS apply in K-12 education, and what might need to change in order to expand the tool’s relevance and impact in the space? Second, what role should Social Finance play in education: is there a feasible and sustainable line of work that the firm should pursue? III. Review of Knowledge for Action The Field(s): PFS and Education The traditional PFS model, or SIB, was introduced globally in the U.K. in 2010 and in the U.S. in 2012 (Dear, et al., 2016). Through this innovative financing tool, private investors provide upfront capital for social services, and are paid back by government once pre-determined outcomes are reached. This allows governments to explore new programs without risk, and enables them to purchase outcomes rather than the delivery of services. PFS emerged at the intersection of three powerful trends in financing for social services: 1. Increased Government Focus on Evidence and Accountability. The most established foundation for PFS is the concept of performance-based contracting (PBC), in which performance requirements are incorporated into statements of work or contracts, and used to influence the selection and accountability of government contractors. This practice was first introduced in the Federal Government through the Department of Defense, and spread across Federal agencies through a series of letters and policy statements by the Office of Federal Procurement and the Office of Management and 11 Budget throughout the 1990s, leading to a General Services Administration goal of 50% PBCs in 2005 (Office of Federal Procurement Policy, Office of Management and Budget, Executive Office of the President, 1998; Boykin, 2005). Many early PBCs’ requirements focused on quantity or basic quality metrics that were used as grounds for stopping payment, rather than as contingencies that had to be met prior to payment. The disconnect between expenditures and outcomes continued; in 2013, Peter Orszag and Jim Nussle – former Budget Directors under Presidents Obama and George W. Bush – found that less than 1% of government spending is backed by evidence of effectiveness or impact (Orszag & Bridgeland, 2013). Since then, government has seen a significant increase in accountability standards, evidence-based decision-making, and outcomes- based (rather than quantity or basic quality) performance metrics (Ray, 2016; Rosenberg, 2012; Palandjian & Shumway, 2015). PFS takes this trend to new rigor, making government payment contingent upon outcomes. 2. Rapidly Growing Impact Investing Field. In 2007, the Rockefeller Foundation coined the term “impact investing,” to refer to investments seeking both a financial and social impact return (Clarkin & Cangioni, 2015). Since then, impact investing has expanded rapidly, with engagement from traditional philanthropies and traditional investors (e.g., venture capital firms, private family offices, and corporate investors). Impact investors, like traditional investors, place “bets” in the form of financial investment based on the level of social and financial returns they believe an organization will achieve, but the two forms of return are accrued independently; an organization may make an investor a significant financial return but not achieve its social outcomes, or may achieve its social outcomes but not achieve financial returns at all. PFS also takes this trend to new rigor, 12 as financial returns are contingent upon social return such that investors either get both forms of return, or no returns at all (Palandjian & Shumway, 2015). 3. Opportunity for Transformative Scale. Although there are a vast number of nonprofits in existence, few are operating with scale or evidence of effectiveness. In fact, almost two-thirds of the roughly 1.1 million registered public charities operate with annual revenues under $500,000, and “roughly 1 percent...garner over 85 percent of [the] $1.5 trillion that comes into the sector each year” (Berger, 2013, para. 1; McKeever & Pettijohn, 2014). Although some nonprofits have strong evidence of their effectiveness, many do not rigorously evaluate their impact, leading to “a general fatigue with social welfare programs that never seem to work” (Ray, 2016, The Emergence of SIBs section, para. 2). The nonprofit sector, characterized by small organizations and fragmented evidence of impact, faces an urgent need to support programs that work in achieving scale. PFS offers a remedy: by providing nonprofits with upfront growth capital, and governments with a risk-free way to explore new solutions, the tool has the potential to bring programs that work into standard government practice, transforming both the nonprofit and public sectors (Palandjian & Shumway, 2015). Emerging at the convergence of these trends, PFS has seen rapid expansion since its founding. In the U.S., the tool garnered early interest from nonprofits, governments, and funders. In 2011, “Social Impact Bond” was one of the top ten philanthropy buzzwords, and in 2016 SIBs were listed as a top legislative issue to watch by Governing Magazine (Belinsky, 2012; Pettus, 2013; Ray, 2016). Currently, there are 20 SIBs in the U.S., and over 100 around the world in the U.K., Europe, India, and Australia. These 13 deals total $200 million of invested capital, a number which is expected to increase to $400 million - $1 trillion by 2020 (Rangan & Chase, 2015; Saltman, 2017). PFS has been slow to expand into education; to-date there are only 5 PFS deals in education worldwide – 2 in the U.S. focused on Pre-K, 1 in Portugal focused on computer science in high schools, and 1 each in Canada and Israel focused on higher education (Gustafsson-Wright & Atinc, 2014). In the U.S., the tool has gained interest since its inclusion in ESSA (Barrett, 2015; Saltman, 2017). PFS builds on the sector’s increasing emphasis on accountability, first with No Child Left Behind’s focus on assessment and compliance and then with ESSA, which takes a more flexible approach to the same concept of high-quality and consistent standards (Gustafsson-Wright & Atinc, 2014). Through PFS, the DOE aims to promote outcomes, broaden the set of partners involved in the delivery of education, coordinate services based on student need, and use data to create performance-based feedback loops (Gustafsson-Wright & Atinc, 2014). Although there is limited academic literature on PFS given the young nature of the field, a few pieces have highlighted key benefits (predominantly in news, blogs, and practitioner-facing journals), many of which apply in education. Part of the appeal of PFS in the U.S. is its bipartisan support; both Democratic and Republican controlled legislatures have launched PFS initiatives, and the Obama administration even created a specific PFS branch within its Social Innovation Fund (Ray, 2016; Boggild & Hlady, 2016). Benefits of PFS for governments include the ability to pursue creative programs without incurring financial risk, and access to new tools that support monitoring, evaluation, and quantification of the costs and benefits of social welfare (Pettus, 2013; Rangan & Chase, 2015; Saltman, 2017). Thought leaders also highlight system-wide 14 benefits of the tool, including the opportunity to tap new flows of capital to plug funding gaps in the government and nonprofit sectors, and increased coordination between government, nonprofits, and funders that could transform connectedness within the system (Saltman, 2017; Rangan & Chase, 2015; Kohli, 2010). Ultimately, the tool creates a “learning tool for society,” and enables governments to “buy” outcomes, instead of gambling on service models (Saltman, 2017, p. 2). In education, where budgets (at the Federal, state, and local levels) are “almost exclusively designed to pay for inputs rather than achieving outcomes,” PFS “encourage[s] schools to experiment with new methods while guaranteeing that investors are paid only if children succeed” (Mesecar & Soifer, 2016, Performance-based funding section, para. 2; Downey, 2017, para. 6). By expanding capital for programs that work, PFS creates a pathway for improvement of student outcomes across the country (Barrett, 2015). PFS, however, has not emerged without controversy; in fact, much of the literature that does exist highlights the field’s challenges. These critiques fall into three categories:  Perceived Privatization and Profit Motives. The loudest criticism of PFS comes from those who believe that its fundamental impact will be to turn public goods into opportunities for private profit. These critics perceive PFS as a “cloak of respectability” for banks and investors (Cohen, 2014; Ray, 2016). “Let’s call it what it is,” writes Mark Rosenman in Making Sense, “private profit crowding out a public good” (Rosenman, 2014, para. 7). For those who see privatization of public services as a problematic and recent trend, PFS is another tool that is being promoted by the same people and institutions who stand to benefit financially (Saltman, 2017; Boggild & Hlady, 2016). 15 In education, this critique is particularly poignant following the first Pre-K project, launched in Utah in 2013. Although the project’s evaluators found the effort to be successful, a New York Times investigation questioned the results, disagreeing with the evaluation methodology that used prior program data, rather than a live counter- factual, as a comparison group (Boggild & Hlady, 2016). In this deal, Goldman Sachs – the primary investor – was paid back with public funds, launching a series of criticisms across the sector calling PFS a “privatization scheme dressed as innovation, accountability, and Corporate Social Responsibility,” an “outright threat”, and likely to “lead to scandal” (Saltman, 2017, p. 4; McIntyre, 2015). These critiques have likely driven the lag in expanding PFS to education, and highlight the necessity of constructing deals that minimize perverse incentives.  Complexity. Even proponents of PFS worry about the complexity of the tool. “It seems straightforward enough,” writes the Nonprofit Finance Fund (a strong supporter of the field), “but between the start and finish are layers of complicated details involving expensive intermediaries, detailed financial stipulations, and a new type of ‘quant’ who evaluates the program’s progress” (Ray, 2016, para. 5). Critics note that this complexity increases transaction costs, making services more expensive than if government simply procured them directly (Saltman, 2017; Rosenberg, 2012). The complexity is amplified in education, where funding flows are dispersed across Federal, State, and local jurisdictions, and transactions may be even more complicated than in other sectors. Leading practitioners, such as Social Finance, acknowledge the complexity and costs of the transactions, but believe that they are necessary: “Of course we would rather that the government procured the services directly,” our CEO said when asked about this 16 challenge, “but they aren’t, and our role is to prove the value of the services, so that hopefully they will direct their funds more appropriately in the future” (T. Palandjian, personal communication, September 1, 2017). If viewed as rapid government reform – rather than a discreet financing transaction –the perceived complexity of PFS is actually the capacity building, stakeholder alignment, knowledge building around performance and performance drivers, and other activities (and costs) required to support governments in a new way of thinking.  Narrow Relevance. The third category of critique may be the most challenging for the education sector, and was the driver of my strategic project. PFS, in its traditional structure of the SIB, may only be appropriate for a small set of nonprofits (Rangan & Chase, 2015). Rangan and Chase outline two key requirements, the ability to effectively (and reliably) deliver and measure impact, and to translate that impact into quantified benefits and savings. They argue that these requirements only apply to a select group of programs and nonprofits – often those already contracting with the government. Nonprofits without an evidence-based model, outcomes tracking capabilities, and target outcomes that meaningfully translate to value on a government’s budget are unable to take advantage of the tool, and there are high costs associated with building those necessary capabilities. Other limitations include: (1) financial benefits may be incurred across multiple jurisdictions, diluting the deal’s value to any single one; (2) most government budgeting processes use the prior year’s budget as a base, limiting room to direct funds towards innovative schemes; and (3) PFS requires a direct link between the intervention and impact, that can be measured within a reasonable timeframe of a deal (Ray, 2016; Boggild & Hlady, 2016). 17 All of these limitations apply in education, where experimental evidence is limited, tracking long-term outcomes is challenging, and benefits accrue after many years and across many jurisdictions in and outside of the sector (Downey, 2017). In fact, when I arrived at Social Finance, my first observation was the vast disconnect between the education sector and the firm’s criteria for PFS-ready programs (depicted in Figure 2). While the controversial and complex natures of PFS are certainly challenges in the sector, the narrow relevance of the traditional model is the most significant barrier. Figure 2. Education Sector Fit with Social Finance’s “Criteria” for PFS-Readiness The left side of this figure is a summary of the criteria that Social Finance shared with me at the start of my residency, which they use to determine whether PFS is feasible in a particular project context. On the right side, I outline the challenges with applying this framework in the education sector. These challenges have limited PFS’s application in the sector to-date, and adapting to address them was the focus of my work. Characteristics of PFS-ready programs: Challenges in the public education setting: Defined Target Population Underserved, large-scale population, with adequate demand for intervention  In public school setting, typically not permissible to create “inclusion criteria” to limit an intervention to a “target populations” of a particular sub-group of students Codified Program Model Well-codified program model with fidelity monitoring  Most education “interventions” – often whole-school approaches to multiple school years – are broader and more complex than typical PFS projects  Programs typically involve a range of teachers, students, parents, and nonprofit staff; identical implementation from one setting to another is challenging  Strict codification is often illogical, given the varied nature of student needs Rigorously Evaluated Evidence base studied versus rigorous comparison group  Prior research on educational interventions is limited, particularly research using RCTs or other experimental methods  Randomization at the student level is often impossible for equity and implementation reasons (e.g., if students are in the same classroom, there will be spillover effects); teacher-level randomization may also face spillover effects Scalable Service Provider High-quality provider with capacity to scale  Scale of target population determined by district and school size; expansion within a school limited to its geographic coverage Positive Value Link to public-sector benefits (economic and community) in reasonable timeframe  Clear value for students, government, and employers, but often accrued over very long time horizon (particularly programs focused on earlier grades)  Majority of value comes in positive societal benefits, rather than fiscal ones  Value dispersed across multiple stakeholders and levels of government, diluting incentive for any single stakeholder to sign up as a payor Payor and Investor Interest Outcomes attract civic and/or commercial support  Complex funding flows, with significant variation in practice from state to state  Existing funding formulas predominantly based on per pupil calculation; limited existence of flexible funding pools  Value dispersed across multiple stakeholders and levels of government, diluting incentive for any single stakeholder to take on payor role alone With all of the challenges facing PFS in the education sector, one may wonder why we should bother to try to apply the tool in the space. The answer lies in the critical 18 importance of improving outcomes in this sector. An article published by Brookings captured the urgency of innovative funding in one paragraph: Is the annual expenditure on U.S. public elementary and secondary education appropriate and sustainable? Reasonable people can disagree whether the current $600+ billion - 5.2% of the nation’s GDP - is too much or not enough, especially when considering the different federal, state, and local jurisdictions involved. However, there are increasing signs that it likely not sustainable at its present, relatively low level of productivity. (Mesecar & Soifer, 2016, para. 1) PFS could have significant impact for students and fundamentally transform the sector to focus on outcomes – like post-secondary persistence and future employment – rather than standards and compliance. Rangan and Chase critiqued PFS’s applicability, saying that “the application of the PFS model for programs that fall outside of this set of criteria will be challenging and their success will require significant adaptations in financing and measurement” (Rangan & Chase, 2015, p. 28). Reflecting on the urgent need in the space, and the many benefits of PFS, I viewed this critique as a path forward; in order to help Social Finance maximize their impact in education, I would have to focus on what adaptations to the model would be required to apply the tool in this critical sector. The Firm: Organizational Context As I entered my residency, I knew that I was not operating in a vacuum, but within the specific organizational context of Social Finance – a relatively new nonprofit, working in a very new field of financing. In order to better understand this context, and its implications for my work, I conducted a strategy audit of Social Finance’s value, capacity, and support for its broad mission of mobilizing capital towards social progress, and for its work in education 19 specifically (Leonard, 2002). Key findings that are critical context points for my strategic project are discussed below. First, innovation is embedded in Social Finance’s organizational DNA. In the past seven years, the organization has not only launched itself, but has also launched an entire field. Exploring new ideas is a regular concept for the organization; when I started the residency there were multiple active “internal projects” focused on innovation within the firm. One project that I will reference throughout the Capstone, called the Public Impact Initiative (or “PII”), was focused on shifting from a comprehensive offering (i.e., of PFS feasibility or transaction structuring) for government partners to a broader set of more targeted supports. This effort aimed to address the fact that not all jurisdictions are ready to implement a full PFS transaction, but may be interested in setting up the foundational elements of PFS such as data tracking systems and evidence-reviews of high potential programs. The innovative mindset underlying PII and the other internal projects attracts interest from a wide range of stakeholders and provides flexibility to continue to shape the PFS model and the firm’s offerings around it. In the education space, with a new set of stakeholders and challenges, this bias towards innovation was a critical asset. Second, the organization operates, by design, in project-based teams. Cross-project learning occurs through individuals, who often bring insights from similar past or active projects they have experienced. However, when I began my residency, there were few formal examples of teams actively working together, potentially leading to some missed opportunities for learning across projects. Third, the organization is rapidly growing its reach and impact. Within my first six months of residency Social Finance began a large number of new projects, which meant that 20 staff were operating at full – or overfull – capacity in terms of project load. This growth is positive for sustainability and field advancement, but it also meant that at the start of my strategic project, staff had limited bandwidth to support work that fell outside of their projects. Finally, the organization has intentionally taken an “issue area agnostic” approach to its work. This approach has enabled the firm to have impact across a broad range of programs and contexts, and to remain flexible as the new PFS field developed. However, the idea of pursuing an explicit issue area strategy was new to the firm. Even my presence as an education-focused Director was odd in this context; although many staff have deep experience in one or more issue areas, I was the only team member with an explicit focus. Understanding this organizational context was critical as I contemplated the changes that may be required for PFS to work in education. As I navigated through the change management literature, this context helped me adapt my approach – or theory of action – to leading change in a way that would be accepted and have impact in the organization. In the next section, I review my learnings from the literature on change management, and discuss how I applied them given the context described above. The Work: Managing Change Organizational change is hard. This is certainly not news intuitively, and writers and researchers have spent many years exploring why change is so difficult and what strategies support successful organizational change. Their reflections give caution to the leader seeking change. Beer and Nohria find that 70% of organizational change initiatives fail (Beer & Nohria, 2000). Kotter writes about the 100+ companies he has seen attempt to change that: “A few...have been very successful. A few have been utter failures. Most fall somewhere in between, with a distinct tilt toward the lower end of the scale” (Kotter, 1995, p. 96). In helping 21 Social Finance think about change in the context of its young field of PFS, and the particular difficulties involved with PFS in education, I was aware of the potential challenges ahead and eager to ground my strategy in evidence around what change tactics work. As a starting point, I sought out Harvard Business Review’s “10 Must Reads” series focused on change management, a 2011 compendium of HBR’s 10 seminal articles focused on the topic. Each of these 10 papers offered strategies and tactics for change leaders from a slightly different angle, such as focusing on mindsets or the specifics of leading a new team through change. Despite the different frames, five key themes emerged emphatically about the drivers of the change challenge and how to address them: Urgency. Kotter highlights eight errors that change leaders make in Leading Change: Why Transformation Efforts Fail, the first of which is urgency: “without motivation,” he observes, “people won’t help, and the effort goes nowhere” (Kotter, 1995, p. 97). He finds that lack of urgency can be driven by a range of factors such as fear and resistance to change, but is most often accompanied by a failure to communicate the need for change. Without understanding the rationale or value of the change, moving forward is arduous. Kim and Mauborgne echo this observation in Tipping Point Leadership, writing that often, “the hardest battle is simply getting people to agree on the causes of current problems and the need for change” (Kim & Mauborgne, 2003, p. 62). In order to establish this urgency, change leaders should avoid attempts to “tell” the reasons for change, but instead should “show” or demonstrate the need (Kim & Mauborgne, 2003, p. 62). In Why Change Programs Don’t Produce Change, Beer, Eisenstat, and Spector argue that urgency can only be established through joint diagnosis of the issue, and Garvin and Roberto suggest in Change through 22 Persuasion that employees need to understand the value of change logically, emotionally, and through direct experience (Beer, Eisenstat, & Spector, 1990; Garvin & Roberto, 2005). Shared Vision. A second theme is the need to establish a shared vision for success, or the desired outcome of change. Kotter’s third and fourth errors focus on establishing and communicating this vision, without which, he argues, “a transformation effort can easily dissolve into a list of confusing and incompatible projects that can take the organization in the wrong direction or nowhere at all” (Kotter, 1995, p. 99). Beer, Eisenstat, and Spector suggest that the vision be co-created and shared across key change agents at the organization (Beer, Eisenstat, & Spector, 1990). This shared vision supports consensus around goals, competence in enacting steps to reach them, and organizational cohesion to move the project forward. Strong Coalition. The idea of establishing a coalition of change agents is highlighted explicitly in 7 of HBR’s 10 seminal articles. Kotter suggests establishing a guiding coalition even before establishing a vision, arguing that these change agents can then empower others to enact the vision with them. In Leading the Team You Inherit, Watkins suggests that change teams should seek alignment around four basic questions: “What will we accomplish?,” “Why should we do it?”, “How will we do it?”, and “Who will do what?” (Watkins, 2016, p. 65). Key to the coalition is the inclusion of influencers and like-minded colleagues who can act as champions. Kim and Mauborgne define influencers as connectors and persuaders who can ease or block change; these influencers, they write, “act like kingpins in bowling: when you hit them just right, all the pins topple over” (Kim & Mauborgne, 2003, p. 67). Early Wins. Creating and highlighting early wins is also highlighted in 7 of HBR’s 10 seminal articles. In The Hard Side of Change Management, Sirkin, Keenan, and Jackson suggest that change leaders put frequent milestones in place that “describe major actions or 23 achievements” (Sirkin, Keenan, & Jackson, 2005, p. 111). Ferrazzi, in Managing Change, One Day at a Time, argues that it is critical to “find ways for employees to demonstrate and celebrate incremental achievements” such that change feels achievable (Ferrazzi, 2014, p. 25). Kim and Mauborgne suggest that describing the vision as a series of short term, attainable goals makes change feel more manageable (Kim & Mauborgne, 2003). Two of Kotter’s errors are focused planning for and learning from early wins (Kotter, 1995). There is clear consensus that small wins inform strategy, ensure progress, and build morale, all critical for change. Senior Commitment. Change authors tend to focus either on strategic, top-down, revolutionary change or adaptive, bottom-up, evolutionary change. Depending on their contextual frame, the role that they suggest for senior leadership in the change process looks quite different. For example, the change processes put forward by Kotter and Kim and Mauborgne begin with an urgent vision for change, which drives the formation of a coalition and plan to move forward. Others, such as Meyerson and Beer, Eisenstat, and Spector, outline change processes that occur slowly and adaptively, as people and organizations adjust to meet new challenges. Across these contexts, though, there is agreement that senior level commitment to change is critical in order to create a supportive environment for those who are driving the change throughout the organization. Beer, Eisenstat, and Spector (1990) suggest that the optimal role of leadership is to create a “market” for change, including setting standards and accountability for improvement, elevating employees or units who have led successful change as models, and developing career pathways that encourage and enable leadership development (Beer, Eisenstat, & Spector, 1990). In Cracking the Code of Change, Beer and Nohria argue that leaders should set the direction for change, but engagement must occur throughout the organization (Beer & 24 Nohria, 2000). Similarly, Sirkin, Keenan, and Jackson argue that change requires commitment of senior executives and any staff impacted by the change (Sirkin, Keenan, & Jackson, 2005). As I reflected on the organizational context of Social Finance, I felt that an adaptive, bottoms-up approach would offer a stronger path to success. “Successful transformations” Beer, Eisenstat, and Spector write, “usually started at the periphery of the corporations...far from corporate headquarters” and they typically focused “on the work itself...[as] ad hoc organizational arrangements to solve concrete business problems” (Beer, Eisenstat, & Spector, 1990, pp. 158-159). Building on this guidance, I developed a theory of action that incorporates the five themes from the literature, but is grounded in Social Finance’s work and the idea of addressing tangible, concrete business issues. Theory of Change If I opportunistically explore how to adapt and apply PFS in education by:  Establishing project-based urgency for the need to adapt the PFS model for education, by working with the six K-12 teams to uncover barriers in their contexts;  Identifying and energizing a coalition of champions for both adapting the model and expanding the firm’s future education work, by rejuvenating the voluntary “education issue area team” and engaging thought leaders across education project teams;  Enabling shared experience of early wins, by bringing teams and champions together to collaborate, such that all local progress is felt across the emerging coalition;  Building Senior Team commitment, by engaging Senior Team members individually and the Advisory Leadership Team as a group, as challenges and early wins arise; and finally  Using the foundation of project experiences to build a vision for Social Finance in education, that is the natural end point of the work, rather than the starting point Then I will adapt Social Finance’s model of engagement in K-12 education, leading to:  Successful outcomes for Social Finance’s existing education work, by helping each project team adapt its approach in context; 25  Shared agreement about Social Finance’s role and strategy in education, and the degree to which future education work should be considered;  Expanded business opportunities for Social Finance in education, and a foundation of champions, resources, capabilities, and engagement models to support this work. My goal was to lead incremental, adaptive, business-oriented change – explicitly not large, visible, revolutionary change. In developing the theory of action, I tried to stay true to the nature of the firm, particularly its opportunistic approach that has driven much of its success. IV. Description, Evidence, and Analysis of the Strategic Project The What and How: Overview of the Strategic Project My strategic project was focused on leading Social Finance in adapting its approach to pursue its broad social sector mission within the K-12 education context. The goals of the project were three-fold. First, I aimed to understand what elements of PFS presented barriers in K-12 education, and implement adaptations to overcome those barriers on active projects. Second, I sought to design an adapted model of PFS for education that could serve as a foundation for Social Finance’s future work in the space. Finally, I sought to contribute new knowledge to the PFS field about whether and how it might have impact in education. Strategy. As mentioned, the foundation for my strategic project was a set of six K-12 education projects that launched at Social Finance in my first four month months of residency. These projects created a “sandbox” of six different education contexts, in which I could explore my theory of change. In order to accomplish my goals, I worked with my supervisor to ensure that I was engaged in all of the projects, leading four in the Director role, and playing an advisor role on the other two. Social Finance projects are typically conducted in teams of three: a Director, an Associate Director, and an Associate. The Director is accountable for the overall success of the project, serving to guide the team’s thinking and ensure that the work is on-track towards the best impact for project partners. The Associate Director plays the 26 project manager role, defining key project steps and ensuring that the team moves towards them. The Associate plays the analyst role, taking on key workstreams and driving content development. My role as Director on four of the projects meant that I would have the opportunity to directly shape our approach and work, and would be well positioned to understand the needs of our project partners so that I could adjust the approach as necessary. It also meant that, inherently, I would not be working on my strategic project alone, as I had four different project teams that would be exploring PFS in education with me. One important note is that the four projects on which I played the Director role launched between September and December 2017. Thus, in my first two months of residency, I did not have a traditional role on any projects2. My limited formal authority on these first two projects, compounded by the fact that I was new at Social Finance and hesitant to disagree with the traditional feasibility process, resulted in me having less influence on the approach and outputs of the first two projects. My approach involved six key activities (depicted in Figure 3), each intended to accomplish one or multiple prongs of my theory of action: 1) Project-based work. Participating in project work was necessary for many reasons. In order to adapt the model for education, I had to deeply understand how it traditionally worked, and see first-hand what challenges arose in trying to apply it in K-12 education contexts. I also believed that my project work would support three components of my theory of action. First, I could establish urgency for adapting the model by identifying barriers that arose on each of the six projects. Second, in working with a variety of project 2 My role as “advisor” on the two projects that launched in July – the first two of the four CTE projects – was new to the Social Finance teams, and we designed it so that I could observe the project flow before jumping into the Director role (I was the first person hired immediately into a Director role, so others had helped define the role, or observed it prior to promotion). The side benefit of this non-traditional role was that I was able to be involved across all six K-12 projects that would become the portfolio for my strategic project (a typical Director portfolio only includes four projects). 27 teams, I hoped to cultivate champions for the work. Third, in addressing challenges that arose, I hoped to build the foundation of a vision for the firm’s future work in the space. 2) Cross-Project Collaboration. My second activity was to try to create opportunities for project teams to work together. I hoped to take advantage of the fact that four of the projects stemmed from one funding source, and bring teams together to discuss challenges and ideate potential solutions. In terms of my theory of action, I hoped this activity would elevate the urgency from being project-specific to sector-specific. I also hoped that collaboration would enable teams to experience each other’s early wins, providing fodder for continued pursuit of impact, even if a given team was not experiencing its own wins. Finally, I saw collaboration spaces as opportunities to co-create the vision for PFS in education. Figure 3. Strategy: Six Key Activities and Intended Link to Theory of Action This figure lists the six activities that I hoped would enable me to achieve the five “if” components my theory of action. For each, I show which components of the theory I hoped that action would help me accomplish. Theory of Action Components: Key Activities: Project-based Work Establish Urgency to Adapt Energize a Coalition of Champions Enable Shared Early Wins Build Senior Team Commitment Build Vision for PFS in Ed       Cross-team Collaboration Education Issue Area Team Advisory Leadership Team Senior Team 1:1s PFS-in-Education Workshop 28 3) Participation in Education Issue Area Team (“IAT”). Although Social Finance is “issue area agnostic” in its pursuit of projects and in staffing teams, in 2014 the firm established issue area teams (IATs), each comprised of 4-6 staff, in order to build content knowledge and pursue new opportunities in each area. Participation is not required or considered part of formal staffing, so each team has its own set of norms, and varied engagement based on formal project priorities. When I joined the education IAT, the group met biweekly for 30 minutes, but meetings were often cancelled due to conflicting commitments. I hoped to reinvigorate the team for three reasons. First, I saw the IAT as a natural coalition to lead the work; members had self-selected to participate, demonstrating interest in the sector, and many were staffed to one or more of the education projects. Second, I hoped the IAT could help identify and share “wins” across projects. Third, I hoped that the team would help create the vision for future K-12 education work as we learned from the active projects. 4) Participation in Advisory Leadership Team (“ALT”). My fourth key activity was to participate in weekly business development meetings with the Advisory Leadership Team (ALT), which was comprised of my supervisor, who leads the Advisory practice, and the three Directors on the team (including myself) dedicated to Advisory work. Although these meetings were unrelated to active education projects, I felt it was important to understand the practice’s activities and priorities, as buy-in from this group would be critical in order to implement change at the organizational level. Thus, this activity was intended to help me generate senior level commitment for my project, and provide a sounding board for feedback on the vision as it developed. 29 5) Senior Team 1:1s. Throughout my residency, I maintained weekly 1:1 check-ins with my supervisor and monthly 1:1 check-ins with two other Senior Team (ST) members. I hoped to gain insight into their priorities and provide updates on the education projects as they unfolded, leading to ST commitment to my project and engagement in the emerging vision. 6) PFS in Education Working Session. My last activity was to lead a session with members from across the projects, the IAT, and ALT. My goal was to synthesize learnings across the projects, uncover implications, and develop insights for future education work. I hoped that the resources that emerged from the session would be truly co-created, increasing the potential for my work to have lasting impact beyond my residency. What Happened. Although most of my key activities spanned the duration of my residency, the strategic project unfolded in roughly three phases (Figure 4 depicts the activities and their timing). In the first “assimilate” phase, I focused on understanding and embedding myself into the existing structures of the organization, such as project teams, the IAT, and the ALT. I learned about the PFS model and explored the firm’s history within education. In the second “act” phase, I shifted to more direct engagement with project work and more proactive implementation of all of the key activities within my strategy. Finally, in the third “amplify” phase, I focused on synthesizing and spreading the work and learnings within organization. 30 Figure 4. Timeline of Key Strategic Project Activities This figure shows the timeline for each of my key activities. In addition, I have marked the three phases of my work; each phase involved multiple activities but, as I describe below, my actions and outputs within each activity shifted over time. July September November January March Key Activities: Phase 1: Assimilate Phase 2: Act Phase 3: Amplify Project -based Work Cross-team Collaboration Education Issue Area Team Advisory Leadership Team Senior Team 1:1s PFS-in-Education Workshop Phase 1: Assimilate. In my first 2-3 months of residency, I focused on learning and building a foundation for my change efforts. Through my project work (as an advisor on the first two CTE projects), I focused on three types of learning: First, I learned about the traditional PFS model (and the feasibility assessment around it), both by engaging with the CTE teams and by looking at outputs from past feasibility projects. Second, I shadowed the Director leading the first two CTE projects to understand the responsibilities of the role. Finally, I collected early learnings about the barriers of implementing the traditional PFS model in education. For example, I supported the teams in accessing the research literature on various CTE interventions, uncovering evidence for one of my early hypotheses that there is limited experimental (e.g., through RCTs or quasi-experimental approaches) research on most interventions, with some exceptions such as the Career Academy and Early College High School models that could serve as foundations for the projects. Through these experiences, I 31 gained an understanding of the feasibility approach, and began to create pockets of urgency for adjusting that approach by supporting teams as they encountered barriers in their work. During this phase I began to join IAT meetings (when they occurred). These meetings deepened my understanding of the firm’s history in education and the beliefs on the team about PFS and education. I learned that the IAT’s focus to-date had been on writing proposals and some exploration of public education funding flows. I used one of the meetings to hold a focus group, through which I hoped to establish a baseline of perceptions about Social Finance’s work in education. I asked the group about the firm’s vision in education, our strategy and approach, and what had gone well and not well to-date. It was during this focus group that I learned that there was not actually agreement – even within the IAT – that the PFS model needed to change in education; many in the group felt that every issue area came with a set of challenges, and education should be no different than the others. I also began to lay a foundation for other activities in this phase. For example, when we launched the first two CTE projects, the two teams seemed at first to be operating independently; for example, the teams worked separately on literature reviews despite reviewing quite similar research. Given my advisory role (vs. a formal leadership role) on these projects, I tried to stimulate some collaboration by creating a shared resource folder and suggesting to the Associate Directors on the teams that they connect, but no formal collaboration occurred during this phase. I also began to join ALT meetings and launched regular 1:1 meetings with ST members. Although these spaces were largely places of learning and updates during phase 1, they helped create a foundation for my change work later as I was able to establish some credibility with my peers and the ST. 32 Phase 2: Action. In September and October, we launched the four projects that I would be leading in the Director role. Thus, in phase 2 (from then until December 2017) I was able to take on a more proactive approach across the six activities in my change strategy. Much of my time during phase 2 was spent embedded in project work, through which I was able to deeply explore the limitations of Social Finance’s traditional PFS feasibility approach in education across a variety of contexts. In Figure 5, I show the challenges we faced in assessing the six feasibility criteria. Although many of the barriers that I had anticipated did arise, two stood out across the projects. First, we struggled on almost every team to find ample rigorous evidence of our focus program’s impact. Even on projects focused on Career Academy models, which have been tested by MDRC in a rigorous RCT-based evaluation, our project partners had limited evidence of their own academies’ impacts on student outcomes in high school, post-secondary, and beyond. This created a significant challenge; the crux of the SIB is the certainty that the program will lead to expected outcomes. Without a rigorous evidence base, impact investors would be asked to place a very high risk bet, as the traditional SIB requires a program to reach specific outcomes before any money is returned. Second, in every project we struggled to identify a potential payor. It is important to note that finding a payor is often the most difficult part of setting up a PFS project. However, we found it more difficult than usual in our education projects due to what Social Finance calls the “wrong pockets” problem. On the one hand, the stakeholders who are currently paying for K-12 education out of their “pockets” (e.g., districts, State Education Agencies) have extremely limited flexibility within their budgets to support PFS projects. Further, these stakeholders do not actually realize the benefits of the highest value outcomes of education (that is, post- 33 secondary and career outcomes3). Thus, they are often not only unable, but also have limited incentive, to sign on as a PFS payor. On the other hand, those who do accrue value from improved education – post-secondary, labor, and community stakeholders, whose “pockets” directly benefit from more students persisting through college, being prepared for jobs, and helping to build the prosperity of their communities – often have an indirect association with outcomes from any particular district, and have little precedent of paying for K-12 education. In addition, the benefits that accrue to these stakeholders occur many years after education actually happens (that is, after the time of “service delivery”), further reducing the direct association of those benefits with the costs of delivering high quality education. Thus, these stakeholders also feel limited urgency to sign on as a payor for a K-12 PFS project. Figure 5. Key Challenges Faced Across Projects As outlined in Figure 1, I expected to face challenges in assessing the PFS feasibility criteria across the six education projects. In this figure, I summarize the extent to which each criteria proved challenging, and an explanation of why (or why not) based on my experiences across the six teams. PFS Criteria Difficulty Explanation / What we saw across projects  For all six projects, vast majority of students were naturally in “target population” based on demographic characteristics  Across projects, we were able to take a school-based approach to target populations  Three of six projects were focused on codified program models For the other three projects, teams sought to define programs based on core elements, although there was some challenge in ensuring consistency of implementation Three of six projects were focused on programs with little or no evidence from the national literature Across all six projects, limited evidence existed about the prior performance of the specific service provider or context implementing the model In programs dealing with “core” academics, difficult to think about scale as students would switch from one program to another within a school, rather than receive a new intervention For two of the projects with external nonprofit, the lack of district engagement from before the start of the project led to significant set-back to progress while gaining buy in Although it was possible to articulate clear, long-term societal benefits, they were dispersed across many stakeholders and accrued many years after the intervention  Little to no ability of districts across all six projects to be an outcomes payor or funder, given tight and inflexible budgets We saw only limited interest from State Education Agencies or other jurisdictions across projects; some potential from employers In at least one project, intense legislative budget debates made it challenging to engage any level of government (outside of the district) Defined Target Population Codified Program Model Strong Evidence Base Scalable Service Provider Positive Value Interested Payors 3 Post-secondary and career outcomes can be translated into specific financial benefits such as: increased revenues for institutions of higher education, increased employee productivity, reduced recruitment and training costs, decreased social welfare expenses, and increased tax revenues. 34 These challenges sparked project-based urgency to consider how we might adapt the model in order to support our project partners. Figure 6 highlights a few examples of adaptations that we explored to address the evidence and payor challenges. In all of these examples, we tried to expand the PFS model to fit the context, while remaining true to the underlying philosophy of connecting funding flows with outcomes. Figure 6. Examples of Project-based Adaptations to Address Top Challenges This figure contains a deep dive on the actual content of a selection of my projects. In the figure, I offer three specific examples of how the two most common challenges manifested on different education projects. I then describe how the project teams addressed the challenges, either finding ways to adapt the traditional feasibility approach or developing new PFS-like models that better fit their specific educational contexts. Challenge Project Specific Context Adaptations Explored There was no obvious intervention; although In order to address the lack of evidence, our team proposed a one-year Teacher Retention research exists around drivers of turnover and impact of teacher-focused programs on student outcomes, no research evaluates teacher outcomes of programs. pilot to test a series of high potential programs in the local context, and only move those that met pre-determined outcomes goals into the deal phase. Lack of Rigorous Evidence Base Career Academies (CTE) Although the model is RCT tested, our partners wanted to understand the value of improving quality of academies across a specific district. No evidence existed in the literature or from our partner data to demonstrate the impact of a higher vs. lower quality program We planned to analyze the differential in costs between higher and lower performing academies and any differences in student outcomes between those academies, and triangulate the differential in performance to longer- term outcomes, based on the evidence of the full program in the literature. Our recommendation would then suggest a full research study to confirm our assumptions, but offer a risk-adjusted estimate of expected benefits for improving quality across the district. The “Back on Track” model has not been evaluated, so we had no clear national evidence to draw on. Further, only two of three phases had been implemented to-date in the location (and within the past year), so there was no historical evidence based on local performance. Our team searched the literature for evidence of programs that were similar to each of the three Back on Track phases, and worked with outside content experts to leverage the evidence the informed the model’s development. We are now working on developing an ranged estimate of positive impact based on “adding” the impact across the three phases. Back on Track (CTE) There was a heated, multi-year political battle around the education budget in the local setting; it was therefore unlikely that the state or district would step up as an outcomes payor for a PFS deal. In order to address the lack of a payor, our team proposed a philanthropic outcomes fund, which would aggregate funds from foundations, corporations, and individuals to serve as the collective outcomes payor; after 5 years, districts would determine whether to redirect funds to maintain programs. Lack of Clear High Potential Outcomes Payor Teacher Retention 9th-grade Readiness The district stated in first meeting that they In order to address the payor challenge, our team proposed a redirection had no capacity to be the outcomes payor. of existing funds intended to support whole-school reform towards the program, which is backed by national evidence. The district stated in the first meeting that they had no capacity to be the outcomes payor. We proposed two adaptations: (1) a performance-based contract between the district and nonprofit, so the district pays more if academies improve and less if they do not, and (2) a “prevention fund” in which upfront capital is provided philanthropically; the district would commit to paying to maintain the programmatic changes moving forward, rather than paying back impact investors with a return on their investment Career Academies (CTE) In phase 2, I was able to begin some cross-team collaboration. As I launched the second two CTE projects, I asked the Associate Directors to tweak materials from the first projects as much as possible, enabling the teams to spend less time making kick-off materials and conducting literature reviews. I also revisited my earlier suggestion that we hold cross- 35 CTE meetings with our project teams. With buy in from my supervisor and the Director leading the other two teams, we held one of these sessions in September and one in November. Each featured project updates and a deep dive on a challenging content area or upcoming workstream. Finally, as the first two CTE teams faced one of the biggest challenges in applying PFS to education – quantifying the benefits of the interventions – the Director leading those teams suggested an internal cross-team session focused on that topic. Given my advocacy for collaboration, I was happy to support her, and we held our first internal problem solving across CTE projects in November. The IAT’s engagement with the education projects increased in phase 2, as we launched an effort to aggregate resources across the projects. Building on the shared resources folder I had created in phase 1, we built a template to capture evidence on interventions and on connections between outcomes. We hoped to bring all of the teams’ research into one place that could serve as a foundation for future work. We also made a new “norm” that we would not cancel meetings as long as three people could meet. This led to more frequent meetings, and gave us more time to dig into specific challenges and ideas. One challenging outcome from the increased collaboration was that instead of leading to shared early wins, as I had hoped, teams ended up predominantly sharing early challenges. For example, as the IAT began to dig into the research, we saw a cross-project trend that limited research existed at the level of rigor that Social Finance typically seeks. Although this sparked important conversations about alternative ways to define evidence, the initial experience was of shared frustration (discussed in the evidence section). Further, as four of the projects began to wrap up December, it became evident that the SIB model would not be 36 a likely outcome for any of them. Again, this realization ultimately drove creativity (e.g., as highlighted in Figure 6), but for many team members, this was first perceived as a failure. In order to capture these perceptions more formally, I interviewed all staff members who had experienced education work at Social Finance. Through these interviews, I learned that some team members saw the challenges on education projects as an indication that PFS was an unlikely match for education – for them, our foray seemed to be leading to failure. Others saw the challenges as opportunities to adapt and have impact, even if the approach looked a bit different from typical PFS. Through these interviews I began to collect insights for what might work for PFS in education, and identified potential champions for the work. Finally, in phase 2 I began to use some of my time with the ALT and ST members to elevate the challenges and potential adaptations. One significant step forward for my strategic project occurred when my supervisor, during a meeting with one of my project teams, began to advocate for some of the adapted model options that we were exploring, and suggested that we shift our “Intro to PFS” materials to broaden the PFS definition and incorporate new models. He also suggested that we hold an ALT “re-think” meeting to make sure that we were aligned around the broader framing. This input and buy in helped my teams further refine the adaptations, and created a path for our findings to have impact on the broader organization. Phase 3: Amplify. My actions in phase 2 created a foundation of data points on challenges and potential solutions in applying PFS in the K-12 space. In phase 3 (the first quarter of 2018), I attempted to build on this foundation to synthesize and amplify a vision for Social Finance’s engagement and impact in education in the future. Across the six projects, we were able to develop and explore a range of alternative outcomes-based financing models. Figure 7 outlines the four “new” models that we proposed across the six projects; each of 37 these models addresses the key barriers of the traditional SIB in education, but maintains the underlying intent of PFS in linking dollars directly to outcomes45. These adaptations were initially explored separately by project teams, and in phase 3 we began to realize learnings across projects through two opportunities for collaboration. First, in January we received funding for a cross-site convening for the four CTE projects. During the day-long convening in February, we had the opportunity to learn about each site’s progress, share best practices, and problem solve around key shared challenges. Importantly, we shared the PFS adaptations that the teams had been exploring, and began to reframe our goal as “outcomes-based funding” (vs. looking only to develop SIBs). Another important insight from the convening was that all of teams had experienced significant interest from local employers in the work. In fact, employers were the most likely stakeholders to consider playing a funder or payor role, particularly if projects were connected to their labor needs. The ability for teams to learn about these shared experiences with alternate models and intrigued employers were the first moments of shared “wins,” as teams began to see how PFS thinking could be leveraged in education (full takeaways are included in Appendix B). 4 Many adaptations that we explored were not entirely new ideas. Through one of its internal projects, called the Public Impact Initiative, Social Finance had done very early thinking on some model variations, but had not developed or explored any in project contexts. Through the education projects, we accelerated the development of these models and continued to adapt them to fit our contexts. 5 Given the long time-frame of these projects (feasibility analysis takes 6 months, and implementation takes 6-12), we did not get to the “end point” of structuring adapted models. However, we are planning to move into the structuring period for three of education projects, so some of these adaptations will likely become realities in the near future. 38 Figure 7. Four PFS Adaptations Explored in Education Projects The top graph depicts funding flows within a SIB. The bottom chart provides an overview of the four adaptations that we explored. For each, I show how funding flows (and stakeholders involved) differ from the SIB, and how those changes address the key barriers around evidence and payors. Visual Depiction of Traditional Social Impact Bond Provide upfront capital Impact investors Service Provider Repays capital based on success Payor Adaptation Visual What is adapted Impact on evidence / risk barrier Redirection of Funds Outcomes- based Contract Prevention Fund Outcomes Fund Service Provider Service Provider Redirects existing funds based on evidence Payor Pays differentially based on outcomes achieved Payor Pays to sustain programming Payor Payor Philanthropy Removes the impact investor. Existing funds are deployed directly from the “payor” towards the evidence-based programs and ongoing tracking. Removes the impact investor. Districts pay vendors differentially based on student outcomes achieved. Replaces impact investor with a philanthropic source of upfront capital. If outcomes are met, payor does not return upfront costs, but instead pays for the continuation of the program. Replaces the payor with philanthropy. Impact investors provide upfront capital, and the philanthropic pool of capital is deployed to pay back investors once outcomes are met. Carries minimal risk to districts as programmatic “bets” are based on evidence (rather than existing processes) and no funding is put in jeopardy. Risk stemming from lack of evidence is accrued by the service provider (who is able to control the degree of quality and impact of its services).  Removes the risk associated with the limited evidence base. Philanthropy spends millions of dollars in education annually; this model offers an opportunity to spend the same amount with a sustainable exit strategy if the program works.  Inherently carries the same risk to impact investors as a SIB, but opportunity to reduce risk based on the outcomes tied to payments. Government payors often require outcomes with direct links to cost-savings and rigorous evaluation to ensure that targets are met. Philanthropy is currently spent with no insurance of outcomes, so there may be room to develop intermediate outcomes with less rigorous evaluation, decreasing risk of loss to investors. Does not require a new influx of capital, just a shift in the process of allocating existing resources. Thus, it can be conducted at the district (or potentially state) level, based on internal priorities. District may be required to pay more (for higher outcomes), but additional payments are incremental on existing contracts, rather than large new sums  Does require a payor, but may be simpler as payor does not pay “extra” (a return on investment) if the program works and can potentially redirect funds (approach #1) to pay for ongoing program support. This model may also attract long-term outcomes beneficiaries such as employers and post- secondary stakeholders as they can see the value created with lower perceived commitment.  Removes the traditional payor. However, long-term sustainability is still critical. Interestingly, a traditional SIB has no inherent long- term sustainability; governments only agree to pay (retrospectively) for the cycle of the program within the deal. Goal of outcomes fund is to prove value to government, leading future expenditures to align with that value. Ideally, the outcome fund merges with a prevention fund, and a payor agrees to continue the program if outcomes are met. Provide upfront capital Philanthropy Service Provider Service Provider Provide upfront capital Impact investors Ideally, pays to sustain programming Repays capital based on success 39 Second, following the convening I held a PFS in Education Working Session with 11 team members from across the education projects, the IAT, and the ALT. This working session focused on common themes across projects, outcomes-based funding models that each team explored, key learnings and advice for future education work, and hypotheses around what makes sense for PFS in education (full takeaways are included in Appendix C). The very lively session gave a broader group of staff the opportunity to synthesize learnings across projects, and (in addition to re-enforcing the takeaways from the CTE convening) led to two new insights for Social Finance as it considers its future work in the space:  Value to education partners: We realized that across the education projects, two key pieces of our work brought significant value to our partners. First, our exploration of PFS feasibility required analysis of the specific costs and benefits associated with each program. For most of our partners, particularly the nonprofits, program costs are distributed across the nonprofit and the district, school, and employer partners involved in program delivery, such that few sites had a comprehensive view of the cost of their programs. Further, although all of our partners tracked some student outcomes, few had connected those outcomes to the costs of the programs. Second, in order to identify payors, our feasibility analysis seeks to quantify the value of increased student outcomes and identify which stakeholders accrue that value. This analysis was not only new to our project partners, but also generated ideas about who benefits from – and therefore might support – their work.  Need for a more nuanced education segmentation: During the working session, we were able to articulate a pattern that I had begun to discuss with my supervisor throughout my residency. Our projects that were focused on CTE, which we could directly connect to career and labor market outcomes, offered compelling “ROI” value propositions to 40 employers, post-secondary institutions, and State and Federal beneficiaries of tax revenues. Our other projects, focused on teacher retention and middle school effectiveness, were removed from this career-focused value proposition, diluting the potential ROI to longer- term beneficiaries of improved education. On these projects, we were forced to focus on intermediary outcomes such as attendance and grade retention, which are important to education stakeholders but challenging to quantify in value. Team members with experience on Pre-K projects reflected that the value analysis and relevant stakeholders also looked different in those contexts. Thus, we realized that moving forward we need to think and approach projects differently based on their different direct, tangible outcomes. In spite of all of this progress and learning, I also experienced some challenges in phase 3. First, my two projects that were not part of the CTE cohort were put on pause. In both cases, our initial work was philanthropically funded, and we knew that further funding would be required to continue beyond the exploration phase. In the New Year, it became clear in both cases that our district partners could not fund a continuation of the work themselves or find external funding (despite reflecting great excitement about moving forward if there was funding available). This raised a new critical question, around whether there is a sustainable business model to support future work in education. Second, the engagement of the IAT began to decline in phase 3. As the surged project load continued into the New Year, most meetings were canceled because not even three people could attend. Our efforts at synthesizing research and materials across projects stalled as current projects took priority, and by the end of my residency, we still had not completed any of shared resources that we had envisioned. 41 Finally, in February, I conducted a second round of interviews with education project team members. During the interviews, I learned that the challenges we had encountered across the projects, while sparking urgency to adapt the model, had discouraged many team members from excitement about future education work (particularly relative to other issue areas). In addition, the business model questions that arose on the non-CTE projects left many team members questioning whether Social Finance should be pursuing more education work at all. Thus, as I reached the end of phase 3, and of my residency, I had mixed feelings about the outcomes of my strategic project, and its implications for the field and firm. On the one hand, I felt that we had learned a lot about the challenges of PFS in education, and started to develop some good ideas around what adaptations might work better in the future. On the other hand – in addition to the fact that none of our projects had actually led to an outcomes- based funding transaction (although some still may!) – critical questions remained around the business model, and whether there is sufficient excitement within Social Finance to continue the work. In the next sections, I discuss and analyze the results of my project more deeply, and reflect on their implications for myself, Social Finance, and the education and PFS sectors. Evidence To-Date This section describes the evidence of my results in terms of outputs (the “if” side of my theory of change) and the outcomes (the “then” side of my theory of change). I use a variety of sources of evidence, the majority of which focus on qualitative conversations and observations. To supplement these observations, I also conducted three lines of more formal inquiry (an overview and summary of outputs from each are included in Appendix D).  August 2017 focus group with the IAT (total of four staff members; of note, three were ultimately staffed on at least one of the six education projects). 42  November 2017 1:1 interviews with all staff who had experienced K-12 work at Social Finance (13 in total). Interviews lasted 45 minutes and focused on the firm’s history and approach in education, perceived challenges, and potential ways to address them.  February 2018 1:1 interviews with staff who had participated on the six projects (11 in total). These interviews lasted for 30 minutes and focused on current thinking on the firm’s role, value, approach, and future opportunity in the education sector. Looking across my residency, I had mixed success in pursuing my theory of action, both in terms of outputs and impact. Overall, my project did lead to an opportunistic, project- based exploration of PFS in education. The six education projects served as a sandbox for innovation, and through them we developed and tested a range of adaptations to the traditional PFS model. Ultimately, though, while the project certainly created a foundation for more opportunities in the sector, it did not lead to comprehensive clarity on the firm’s vision for a broader education strategy. Instead, my work helped Social Finance explore a new toolkit that it can use – even if not in education – across all of its work in the future. In terms of output, a summary of the evidence of my progress is included in Figure 8, and synthesized below: Establishing urgency for adapting the model. During the course of my residency, I had significant success in establishing agreement that a new approach was needed for Social Finance’s work in education, and that it was needed immediately for success on our six education projects. In phase 1, there was not agreement that education required a meaningfully different approach. This was evidenced through my August focus group with the IAT and the fact that the first two projects kicked off using traditional “Intro to PFS” materials focused only on SIBs (see Appendix A). By phase 3, through project work and cross-team 43 collaboration, we reached consensus that the SIB was not likely to work in many K-12 education contexts, and that adapted models were needed. This is evidenced by my November and February interviews (see Appendix D), observations from cross-team collaborations, our new “Intro to PFS” materials featuring an expanded view of outcomes-based financing tools (see Appendix A), and the fact that, ultimately, all six projects explored alternative models. Energizing a coalition of champions. Overall, I had mixed success in energizing champions around the change. I was hoping to cultivate champions both for the idea of adapting the model on our current education projects, and also for a longer term vision in K- 12 education. Through project work and cross-team collaborations, I did help to energize champions for the adapted models that we explored in our education projects. This is evidenced by the fact that we explored adapted models across all six projects, and was also articulated in interviews and at cross-team collaborations. For example, one of my proudest moments of residency was during the February CTE convening, during which one of the Associate Directors shared with the full group (including representatives from the funder): "these projects have not only been trail-blazing in that they are the first explorations of PFS in education, but they have been trailblazing for us at Social Finance as we have learned that what we thought was a very cool hammer [i.e., the SIB] is actually a tool-box, and we have been able to take out and try different tools in these education projects." I had less success in energizing champions for future education work. In phases 1 and 2, the IAT’s increased engagement and desire to build cross-project resources suggested that a coalition of champions may be forming. However, as the IAT’s engagement declined in phase 3 and the six projects came to a close, it became clear that a strong coalition did not exist. Strong evidence for this emerged in my February interviews, when 7 out of 10 education 44 project team members said they did not believe that the firm's excitement to do more education work had increased based on the results from the six projects, and only 5 out of 10 said that they would personally be excited to work on future education projects (see Appendix D). Enabling shared experience of early wins. I had limited success in enabling teams to experience early shared wins. Our collaborations were predominantly focused on shared challenges and frustrations; in fact, after our November internal cross-CTE project meeting, one team member joked: “well, at least now I know I am not the only one banging my head against a wall!” I also saw evidence of this frustration in our IAT efforts to create shared resources; for example, one IAT member wrote to me in an email: “I am also finding a lack of rigor in many places as I go through this lit review, which can get a bit demoralizing” (anonymous, personal communication, November 15, 2017). During the February CTE convening and the PFS in Education Working Session, we did experience some (late) shared wins, evidenced by the outputs from both sessions (included in Appendices C and D), which reflect the realization that the adapted models had high potential across project contexts, as well as our new insights about our value and potential future focus in the space. However, given my goal of shared wins coming early, I consider the overall success of this output limited. Building Senior Team Commitment. I had significant success in building ST commitment to adapting the traditional model and to the pursuit of future education work. This is evidenced by the fact that we held an ALT “rethink” meeting in November, and afterwards I shared our expanded “Intro to PFS” slides (included in Appendix A) so that we could be aligned in both our messaging and materials. Further evidence of ST commitment is that the expanded approach was highlighted during our fourth quarter Board of Directors 45 meeting, and at subsequent staff meetings. Finally, the fact that the ST has asked me to stay onboard to continue to pursue education work is strong evidence of their commitment. Building a Vision for Social Finance in Education. I had moderate success in building a vision for our future work in education. My strategic project did not lead to a fully developed strategy for Social Finance in the education sector. In fact, during my February interviews, all of the team members expressed uncertainty around whether we should even create a firm-level education strategy. However, the work did lead to a much clearer vision of how we might engage in education moving forward (if we chose to do so). In fact, in the same interviews, all 10 team members suggested that adopted models or a PII-like approach of stand-alone analyses had higher potential for impact in education than SIBs. Finally, the outputs of our PFS in Education Working Session (included in full in Appendix C) suggest progress towards the starting point of a vision for future focus in education, both in terms of identifying analyses that were particularly valuable to our education partners, and a more nuanced set of education contexts with higher ability to translate to positive ROI across stakeholders. 46 Figure 8: Evidence of Results for “If” Side of Theory of Change This chart ranks my progress towards each of the intended outputs of my project. Green circles suggest significant progress, yellow circles suggest some or mixed progress, and red circles suggest limited or no progress. Key Evidence of Results Output Progress Phase 1 Phase 2 Phase 3 Establ i shi ng Urgency for Adapting the Model Evidence from Project-based Work: --In August focus group, IAT listed challenges in applying PFS to education but believed that "functionally, education has the same challenges we have in every other issue area" so the standard approach should apply (see Appendix D). --First two projects used the traditional "Intro to PFS" materials with project partners, which focus only on SIBs as "PFS" (see Appendix A). Evidence from Cross-team Collaboration: --Although two projects kicked off in phase 1, there were no joint meetings; when I suggested it (in my advisory role), the teams did not see that as a valuable use of time. Evidence from Project-based Work: --In November interviews, 5 of 9 project team members said it was unlikely that their team would recommend traditional PFS (but did not offer alternative solutions aside from consulting-like advising) (see Appendix D). --Next four projects used the new "Intro to PFS" materials with project partners, which focus on a broader range of outcomes-based financing solutions (see Appendix A). Evidence from Cross-team Collaboration: --Series of three cross-team meetings occured between September and November, each focused on problem solving around a specific aspect of the model that was proving challenging on-the-ground. Evidence from Project-based Work: --In February interviews, 8 of 10 project team members believed that adapted models or a PII approach would work better than SIBs in K-12 contexts (see Appendix D). --In five of six K-12 projects, traditional SIB is not being considered as a financing option and adaptations are being or have been explored. Evidence from Cross-team Collaboration: --Key takeaway from February CTE convening was that traditional PFS is not the likely approach across the board, but alternatives are exciting (see Appendix B). Energizing a Coalition of Champi ons Evidence from Education Issue Area Team: --During August focus group, members of the IAT shared excitement to move forward with the six projects, saying, for example, "before, the goal of the issue area team was, let's get some ed projects. The nut of even getting the projects hadn't been cracked. Now, we have finally gotten a run way and it is exciting just to be able to start the work." (See Appendix D). Evidence from Education Issue Area Team: --Increased engagement across the team, including creating new norm in October that we would meet if at least 3 people could join and fact that the majority of meetings in October, November, and December did occur. --Launch of effort to create shared education resources; created template and assigned aspects of work across team. The goal was both to support the existing teams and to offer a leg up for future business development and future project teams. Evidence from Cross-team Collaboration: --At Feruary CTE convening, staff championed adaptations to partners (see quote in body of text). Evidence from Education Issue Area Team: --Biweekly meetings remained challenging; no meetings occured in January or February of 2018. --No output from efforts to create shared resources. Other Evidence: --Increased use of new "Intro to PFS" material: Two ADs used the new slides on other, non-education projects. Head of partnerships integrated the new slides into standard materials. --No increase in excitement about further targeted education exploration: In February interviews, 7 out of 10 team members said that they did not believe that the firm's energy to do more education worked had increased based on the results from the six projects. Enabl i ng Shared Experiences ofEarly Wi ns Evidence from Cross-team Collaboration: --During all three cross-team collaborations in phase 2, multiple team members stated their gratefulness at knowing that they were not alone in their experiences of challenge. -- No evidence of solutions output from collaborations. Evidence from Education Issue Area Team: --Efforts to build shared resources led to realization of more challenges than expected. Evidence from Cross-team Collaboration: --During February CTE convening, two partner teams shared that seeing how the other teams had explored alternative models gave them hope for how our projects could have impact. Evidence from PFS in Education Working Session: --Participants expressed new energy for education work as they saw that alternative approaches had been explored across projects. --Outputs from the session include ideas of best practices for future work (see Appendix C). Evidence from Education Issue Area Team: --Biweekly meetings occured rarely (limiting sharing). Building Senior Team Commitment Evidence from Senior Team 1:1s: --When I asked the ST if I could staff a team to explore an Outcomes Fund, they gave me two months to do so (without revenue coming in). --During 1:1s with my supervisor, I was able to be fully transparent about my vision for adapting the model and we were able to link the thesis to his PII efforts, so that he -- and the other ST members -- had clear rationale for sponsoring my exploration. Evidence from Senior Team 1:1s: --CEO offered to help connect with me education- focused funders in our network. --Two ST members became champions for new applications of PFS in education, pulling me into exploration of two new potential projects Evidence from Advisory Leadership Team: --In November, we held an "Advisory Rethink" to discuss the need for broader PFS language in our materials to capture our exploration of alternative / expanded approaches. Quote from calendar invite: "Guys! We need to start aligning our thinking, language, and materials about the new PFS! It will be fun! (Blame Annie KP) -- Jeff" Afterwards, I distributed new "Intro to PFS" slides so that ALT could work from same materials (see Appendix A). Other Evidence: --In the December Board of Directors meeting, a slide from the new "Intro to PFS" was included. --In an internal update, my supervisor highlighted the alternative models to the full team. Evidence from Senior Team 1:1s: --I was made an offer to stay at Social Finance past my residency to accelerate the education work and continue to define the strategy. --Every member of the ST made at least one introduction between me and someone in the education space who I thought might be able to inform and / or support our education work. Evidence from Advisory Leadership Team: --The Director leading the first two projects asked in January to discuss a reframe of the way we introduce PFS, saying: "We need to frame our goals differently at the start of new feasibility projects; instead of saying we are determining feasibility for PFS, we need to ground teams in the idea that there are a range of outcomes-based financing tools and our goal is to understand which is most helpful in the given context." --The ALT decided to recommend that the first two projects (which had completed feasibility) move into transaction structuring (for alternative models, rather than traditional PFS). Building Visionfor Social Finance in Ed Other Evidence: --During August focus group, members of the issue area team shared that "beyond execute well, not sure there is a broader strategy in education." (See Appendix D). Other Evidence (from November interviews): --8 out of 13 staff members defined the vision for our work in education as opportunistic, and a few explicitly stated that "there is no vision" for Social Finance in education (see Appendix D). --9 out of 13 staff members offered ideas for how we could add value in the education space outside of traditional PFs projects (see Appendix D). Evidence from Project-Based Work: --All 10 project team members identified specific ways that Social Finance could bring (and had brought) value to the education space (see Appendix D). --8 out of 10 project team members suggested that future work should focus the alternative models that we explored through our projects, as well as stand-alone, PII-like work (see Appendix D). --All 6 projects explored alternative models, creating a set of examples to guide future work. Evidence from PFS in Education Working Session: --Outputs included a set of conditions for success for future project, a list of highest potential models, and hypotheses for next steps to continue to explore PFS in education (see Appendix C). 47 This mixed success in achieving my intended outputs led to similar mixed success in my impact (depicted with evidence in Figure 9). On the positive side, we were able to create meaningful value for all of our education partners, enabling our success across the existing projects. This is evidenced by positive feedback from our partners and reflections from team members in interviews. For example, at the February CTE convening, one district partner stated that the project was "the best recommendation and support that we have ever received" and another nonprofit partner stated that "even if we do not end up with an outcomes-based funding model, the cost-benefit analysis alone is a game-changer for us and will fundamentally improve our messaging to our district and employer partners." Project team members also highlighted this success across the six projects: in my February interviews, all 11 team members named one or more ways in their projects had been successful in providing value to partners. This value was further articulated and synthesized during the PFS in Education Working Session (see Appendix C). In addition, we gained significant knowledge and experience in the education space, and built insights and tools that will increase our impact on new education projects. Thus, my strategic project created a foundation for Social Finance for expanded opportunities for business (and impact) in education. Evidence of this success includes the fact that we now have education-focused business-development materials, a significant number of inquiries about our education work from philanthropies and other stakeholders, and a few education business development opportunities already on the horizon. The impact that I made least progress towards was a shared agreement on Social Finance’s role and strategy in the education sector. My work did offer new information on what that role and strategy might look like, as highlighted above. These learnings for the future 48 are reflected in both the models we explored (depicted in Figure 7) and in the outputs from the PFS in Education Working Session (included in Appendix C). Figure 9. Summary of Evidence of Results for Theory of Change This chart provides a recap of my progress towards my intended outputs and adds an overview of my progress towards the three intended impacts (that is, the “then” side of my theory of action). As in Figure 8, each assessment is supported with a list of relevant evidence from my six key activities and broader observations. Theory of Change Progress What Happened If I take an opportunistic, business-driven approach to exploring PFS in education by: Establishing Urgency for Adapting the Model See Figure 8. Energizing a Coalition of Champions Enabling Shared Experiences of Early Wins Building Senior Team Commitment Building a Vision for Social Finance in Ed Then I will adapt Social Finance’s model of engagement in K- 12 education, leading to Successful Outcomes for Existing Projects --Client feedback on project value was outstanding. For example, at the February CTE convening, one district partner stated that the project was "the best recommendation and support that we have ever received" and another nonprofit partner stated that "even if we do not end up with an outcomes-based funding model, the cost-benefit analysis alone is a game-changer for us and will fundamentally improve our messaging to our district and employer partners." --In February interviews, all project team members believed that their projects had been successful in providing value to clients (see Appendix D). --In PFS in Education Working Session, team was able to align on top value added to clients across projects (see Appendix C). Shared Agreement of Role & Offering in Ed --In February interviews, only 5 out of 11 education project team members said they would be very excited to work on another education project. --In February interviews, all project team members said that they did not know if we should have a firm-level education strategy; reasons given included that the firm is issue-area agnostic and that it is unclear that education has more value for the firm as opposed to other issue areas --In my offer to stay at the firm, the ST expressed strong support for me driving forward our education approach, but also suggested that without me on the team, future exploratoin would likely remain opportunistic. Expanded opportunities for Social Finance in Ed --We now have a set of education-focused materials developed for use with potential funders and future project partners. --We have a suite of examples of our education work, and a codified set of learnings and possible model adaptions from our six projects that can be used for future work. --We have had multiple opportunities to speak at conferences on PFS in education, and recently had the opportunity to speak at education specific conference; one of the ST members moderated, and I participated on, a panel at the 2018 LearnLaunch conference called "PFS Comes to Education". --We are currently engaged in multiple business development conversations about potential future education work. Ultimately, though, the decision to move forward with a strategy in K-12 education is connected to a broader decision around the firm’s strategy, and therefore I was unable to see this specific impact in the ten month period of my residency. Interestingly, two pieces of evidence suggest that my work had broader strategic influence, and may ultimately lead to some aspect of an education strategy. First, in our February 2018 Board Meeting, the ST announced that the PII effort will initially take an issue-area approach, and one of the two focus areas is workforce and career development. Thus, my work added to a growing body of 49 knowledge across the firm about the political momentum for projects that relate to career outcomes. This evidence also suggests that my work will support the development of the PII strategy, which will include K-12 career-focused programs (as well as programs supporting adult education and workforce development). Second, in March, Social Finance launched an internal project to reflect on, and possibly redesign, our organizational structure. One consideration on the table is whether it makes sense to align some team resources against issue areas, rather than taking our current, fully issue area agnostic, approach. Thus, while I was unable to develop the comprehensive strategy that I envisioned during my residency, my project may have had some influence on the broader organizational strategy that could support coherent future pursuits in education (and other issue areas). Looking across my impact, I made significant progress in addressing the primary goal of my strategic project: helping Social Finance adapt its model to have impact in education. I made some progress (although less than I had hoped) on coming to shared agreement around the underlying question of whether education is a sector that Social Finance should aggressively pursue. The Why: Analysis of What Happened In this section, I assess why I was successful in some of my efforts, while making less progress in others. In order to synthesize my reflections, I will return to the strategy audit referenced in Section III, which suggests an analysis of the value, capacity, and support for my project to understand why its outcomes occurred (Leonard, 2002). Figure 10 depicts the key components of this framework, which will serve as the pressure points of my analysis. Value. My first consideration is how the value of my work, from Social Finance’s perspective, enabled and hindered my success. Overall, the mission and strategy alignment 50 between my project and the firm enabled my success in creating urgency for adaptation, energizing champions for model adaptations (even if not for future education work), building ST commitment for the work, and launching the exploration of alternative models that fed into the beginnings of a vision for Social Finance in education. On the other hand, two other value-based factors – the lack of financial (i.e., business model) mission alignment and the disconnect between my project and Social Finance’s theory of action – detracted from my progress in solidifying a coalition of champions for future education work, enabling shared early wins, and developing a comprehensive vision for Social Finance in the space Figure 10. Key Elements of Strategy Audit Framework6 Capacity: Support:  Funders  Competitors  Policy Landscape  Broader Social Impact Financing Field  Leadership  People and Partners  Organization and Culture  Process and Infrastructure Value:  Mission  Vision and impact goals  Theory of action My efforts aligned strongly with Social Finance’s mission. As I learned in my November interviews, most staff at the firm perceived significant value in Social Finance pursuing work in the K-12 education space from a mission perspective. One staff member 6 Adapted from: Leonard, H. B. (2002). A Short Note on Public Sector Strategy-Building. Harvard Business School, John F. Kennedy School of Government. Cambridge, MA: Herman B. Leonard. Accessed from class with Professor Leonard in Fall 2017; possible to access online at: strategy-building/. 51 reflected: “Education affects all of the issue areas where we do work now. To not be involved in education would indicate that we aren’t serious about a lot of issues. How do you argue about hitting prevention points in other areas if you aren’t in education?” Another similarly argued that “if we are an organization whose main goal is system change in the US, education is one of the most important ways to do that. Social Finance needs to be in education in order to realize its mission at ambitious scale.” At the beginning of my project, this passionate belief in the necessity of Social Finance exploring work in education as part of its mission “towards social progress” created a strong foundation of excitement and anticipation about the work. Further, although these six K-12 projects were the first at the firm, they launched on the heels of many years of internal thinking on how to enter the space. Thus, before we even started the work, there was even more significant excitement from staff across the firm to see where the projects landed. This strong perceived mission-oriented value fostered some urgency to adapt the typical approach as project teams faced barriers. Looking back to the change management literature, this aligns well with the argument put forward by Beer, Eisenstat, and Spector that urgency is often generated from an immediate business need (Beer, Eisenstat, & Spector, 1990). The mission alignment also helped me to energize champions for the adaptations early on. Because most team members saw the projects as important to Social Finance’s mission, they were eager to do what it took to create impact for our education partners. Finally, the alignment enabled my project to launch on a strong foundation of commitment from the ST, who saw the exploration as highly strategic for the firm. As the projects progressed through phases 2 and 3, the alignment of my project with the mission and strategy of the firm became clearer in two distinct ways. First, as more 52 team members saw the value that we were generating for our partners, the imperative for Social Finance to have impact on our projects became stronger. This led to an increased urgency and willingness to adapt, and teams began to actively explore the adaptations that would ultimately serve as inputs as we started to define a vision for Social Finance in education. Second, as the urgency to adapt and the work of adapting progressed, the strategic project began to more squarely align with another strategic priority: sustainability. As the firm reached seven years of age, the ST and Board of Directors had made it an imperative to increase the firm’s sustainability (both in terms of quantity of work and ability to generate revenue). Thus, the fact that my projects were exploring not only a new issue area but also new types of potential offerings generated even more ST support for the work as we progressed. In fact, although my projects did not begin as part of the PII effort (focused on developing a broader suite of services for government clients), the outputs from my projects – examples of how stand-alone cost-benefit analysis added value to our partners, or how we helped a district think about redirecting funds towards evidence- based programs – will serve as templates and examples for future PII work. At the same time as mission and strategy alignment enabled my success, key aspects of my project’s value were disconnected from the firm’s approach, limiting my success. First, as mentioned in Section IV, the business model value to Social Finance of education work came into question during my strategic project. Although we gained clarity on the value we could bring to the space, our projects left us with less clarity around how Social Finance can generate revenue from future education work. This skepticism around the business value was expressed by many team members in November interviews, and likely led to some decreased engagement and support from the IAT and project teams, detracting 53 from my ability to energize a coalition of champions for future education work. The business model question also played significantly into the perceived lack of shared wins; in fact, there is a chance that the increased perceived moral value of being in the space, combined with the increased perceived lack of business model value, exacerbated the perception of failure on some of the projects. Another detractor of my success, from a value perspective, was related to the firm’s theory of action. Social Finance intentionally takes an issue area agnostic approach to its work, and many staff members believe in the necessity of this approach. Insofar as my project was perceived as an exploration of alternative models, it aligned with the opportunistic and sustainability-focused nature of the firm. However, when my work was perceived as a push for the firm to develop a specific commitment to and strategy for increasing its work in the education sector, it inherently clashed with the firm’s theory of action, as well as the issue area agnostic self-image shared by many team members. This misalignment limited my ability to build a coalition of champions for future education work, and ultimately contributed to my limited progress in creating a comprehensive and co-created vision for Social Finance in education. A final reflection on my success, from the lens of the value my project was generating for Social Finance, is that during the time of my residency Social Finance was (and still is) questioning and clarifying its mission, impact goals, theory of action, and strategy. During a recent organizational health survey, two of the top three development areas highlighted by the staff were “mission, vision, and strategy” and “internal communications.” These areas resonated with my observations. For example, when I asked project team members in November about how our education projects fit with Social 54 Finance’s mission, the majority said they were not sure7. When I asked the same question in February, the majority of team members focused on the expansion of the toolset and how that effort aligned with our PII and sustainability efforts. In terms of whether they believed a targeted expansion of education efforts was warranted, most said that before launching into that work, they would need more clarity on which direction the firm was headed. One person said: “Right now it feels like we are exploring everything, and it is unclear what the vision is for what our work looks like in the next five years, or what business(es) we want to be in. I think we need to get more clarity on the broader strategy before deciding whether education is a strategic priority or not.” This lack of alignment, clarity, and communication around the mission – and how each project fits within it – was a detractor for me, particularly in terms of building a coalition and vision for future work in education. Capacity. Four capacity-related factors enabled my work: my own leadership, the people and partners on my project teams, Social Finance’s culture of hard work and innovation, and the cohort model structure of the CTE projects. These enablers supported me in developing urgency, early champions for adapting the model, and the foundation of a vision based on the new tools and insights we developed across the projects. I also faced capacity constraints that hindered my success, including gaps in my leadership, the lack of capacity of project teams given the high project load, and the project-based culture of the firm. These detractors slowed my progress in developing early urgency, highlighting shared wins, energizing champions for future education work, and, ultimately, completing a shared vision for Social Finance in education. 7 Given that the organizational health survey found that the firm’s broader mission and strategy (beyond education) feels unclear to staff, it is not surprising that staff were not sure how education work fit into it! 55 In terms of my leadership, strong sources of power, combined with an entry strategy focused on learning and adding early value, enabled my success. From a positional perspective, I played the Director role on four of the six education projects, so was naturally able to (and, in fact, had the responsibility to) shape the direction of the projects. Further, given my resident status and specific charge to lead a “change project,” staff across the organization knew that I approached the work with an eye towards change, and that I actually “had to” explore the idea of adaptation. These positional factors enabled me to espouse a learning-focused, experimental mindset across my project teams. From a personal perspective, I was aided by the fact that I entered as an “education content expert” in a firm of issue area agnostic staff. This expert status gave me a seat at the table in all conversations related to education at the firm, and helped me in building credibility with my teams such that they were willing to help me explore non-traditional ideas. In addition, my former professional experience as a management consultant was another source of personal power for me at Social Finance. Much of the firm’s work, particularly in the Advisory Practice where I sat, relies on similar skillsets to consulting such as problem solving, analysis, and communication. My ability to utilize these skillsets early on gave me further credibility, and also enabled me to add value to the firm early on, even as I was still learning about the field and my role. Finally, from a relational perspective, I was able to build strong relationships with all of my project team members from two angles. First, part of my leadership style includes an emphasis on skills and career development. I learned this approach at McKinsey, where the theory of action around people development centered on individual project managers creating opportunities and providing strong support for team members. I brought this 56 development lens into all of my project teams, trying to deeply understand my teammates’ goals and which skillsets they sought to flex or build. Second, my leadership style is naturally casual and non-hierarchical (in fact, it has been described in past as friendly). My simultaneous efforts to support my teammates’ professional development, and to engage in and support their lives personally, was instrumental in my success. As mentioned in Section III, Kotter writes that if “people won’t help...the effort goes nowhere”; all of my teams were happy to help me think about the individual projects, and across the projects, in a new way, and without their help any progress would have been impossible. Another aspect of my leadership that enabled my success was my entry approach. In my early months of residency, particularly in July through October, I did not have formal positional authority to pursue my strategic project (the four projects on which I played a director role began in October). During this time period, I leveraged my other sources of power – my resident status, education expertise, and consulting experience – to immediately roll up my sleeves to help on other projects and broader firm development work. In doing this, I was able to both absorb a large amount of knowledge and experience, and also to build relationships with and add value to many teams and leaders at the firm. In an action and achievement oriented firm, where delivering high quality work is the top priority, my willingness to jump right in set me up for success in building my own leadership credibility and reinforcing my sources of power as I moved into formal leadership roles in October. All of these aspects of my leadership gave me a platform and credibility for my call for change, helping to create the urgency that I needed. They also enabled me to energize champions (at least for adapting the model), and to explore adaptations and insights that formed the foundation of the vision for future work. 57 The people and partners across my project teams were also significant enablers of my success. Social Finance is fortunate in its ability to attract an exceptionally talented staff. The Associate Directors and Associates on each of my teams had significant prior experience across the consulting, finance, and policy fields, and brought strong analytic, problem-solving, and communication skills to our work. The high capacity of each project team enabled us to identify and address barriers quickly and creatively, accelerating the establishment of urgency and development of adapted tools that could inform our vision. Relatedly, Social Finance’s culture of working hard, driving value for clients, and having an innovative mindset was also an enabler of my success. “No two SIBs are the same,” said one team member, “so in a way, every single one is innovative.” This mindset led team members to roll up their sleeves to ensure that we were adding value for our project partners and explore adaptations, which helped me in leading the development of new tools and starting to define a future vision. It is important to note that this innovative mindset was also, at times, a barrier to getting buy-in for my project as a change project. On the one hand, there is a belief that every project is innovative and looks different – for example having different types of payors, outcomes targets, and methods of tracking those outcomes – and this fodders innovation within the model. However, this also limited the initial perceived need to innovate on the model itself, which is ultimately what we did through our adaptations. The final capacity-related driver of my success was the fact that the four CTE projects were naturally structured as a cohort supported by the same funding source. All four projects were explicitly focused on similar programs and, therefore, similar evidence bases, cost analyses, value assessments, and potential payors. Thus, although we were 58 working on four separate projects, there was clear opportunity to identify shared challenges and co-create solutions given the similarity of the work. The presence of a cohort created a natural forum for cross-team collaboration. The two non-CTE projects did not participate in any cross-team collaboration until the PFS in Education Working Session. Thus, the fact that there was a cohort element to four of the projects was helpful in ensuring that some cross-team collaboration occurred. Although these cross-team sessions initially focused on shared challenges, they increased the rationale and urgency for adapting the model, and ultimately enabled the teams to share insights and progress (that could perhaps be called “shared late wins”) that formed the foundation for a vision for Social Finance in education. Although capacity-related factors led to much progress in my strategic project, I also faced critical capacity-related constraints. In terms of my leadership, I had limitations in two key sources of power. From the positional perspective, I was not leading the first two education projects that launched. From an organizational perspective, this was a logical decision; I was new and had never played the Director role at Social Finance, or conducted a feasibility analysis using Social Finance’s method. From the perspective of my strategic project, however, this delayed my ability to directly implement my strategic project until October when my projects began to launch. Even in my advisor role, I felt I needed to balance my incoming belief that adaptation was needed with my desire to integrate smoothly into the firm and my recognition that I was inexperienced with the traditional approach that I wanted to adapt. Relatedly, from the personal perspective, although I was an education expert, I was explicitly not a PFS expert. This also contributed to my hesitance, as I did not feel I had full credibility to suggest change without that expertise. In terms of direct impact on my strategic project, these positional and personal constraints in 59 my leadership caused me to move slower in suggested and pursuing change. Thus, if I had been asserted my views earlier, I may have been able to establish urgency faster, and therefore made more progress along all aspects of my theory of change. On the other hand, these constraints reinforced my entry approach that was focused on learning and adding early value. If I had entered the residency declaring that we needed to change the model – instead of spending my first months building knowledge and credibility – I may have alienated many of the team members who ended up helping me explore the adaptations later on. Thus, although these constraints slowed my progress, without them I may have made no progress at all. Two additional capacity-related detractors from my success were the busy staffing period that occurred throughout my residency combined with the project-based culture of the firm. As mentioned in Section III, Social Finance entered a very busy period in October during which all staff were working at or above capacity in their project loads. I identified this issue as a potential detractor early in residency, and earlier in this Capstone. I made a concerted effort to accommodate this issue by asking my teams for insight into their workloads on other projects, working with the Associate Directors to make sure we were focused on highest priority activities, leveraging existing work from other education projects where possible, and taking on direct pieces of work myself to try to lighten workloads. These efforts were helpful in my direct project work, as my teams were able to successfully navigate the challenges that arose in our PFS feasibility assessments. Some of the work associated with my strategic project, though, was not part of the typical project work, such as cross-team collaborations, developing shared resources, and synthesizing across projects. Although we talk about our work at the issue area level often (particularly 60 in, for example, funding or business development contexts), on the ground, there is a very project-based nature to the culture of the firm. Thus, the incremental pieces of work that fell across projects were uncommon, and had no natural forum in which to occur. There was certainly willingness – even eagerness – from IAT and team members to engage in these activities, but the busy staffing period made it difficult (even for me) to prioritize “extra” activities over more critical project work. This limited my progress in creating a coalition of champions – because I was unable to bring potential champions together with frequency – and also limited the sector-level learning activities needed to complete a shared vision. Support. In terms of support, my success was enabled by the existing funding for our education projects, the strong interest and excitement of additional funders around innovative financing for education, and the competitive context that Social Finance was working on what could be the first PFS projects ever in U.S. K-12 education. These factors increased urgency for teams to adapt (once the need for change was clear), because there was increased pressure to be successful given the strong funder interest. They also served as an initial foundation for ST commitment to my exploration, as funders are an important (and necessary) priority at the firm. Despite this early support, though, the complex reality of financing structures in education was a core challenge in reaching the full impact that I sought because it led to fewer shared wins, and a less developed vision than I had intended. Part of the rationale for my joining Social Finance for residency was that the firm had just received funding for the four CTE projects. Thus, from the beginning, there was inherent support for my work from the financial perspective. In addition to that built-in support, we uncovered much interest in exploring PFS in education from stakeholders 61 beyond that funder. As we kicked off the CTE projects, we launched an advisory council of funders interested in innovative financing for youth career outcomes. Initially, we hoped that a handful of funders would be willing to serve on the council to provide feedback and problem solve with us across the projects. Ultimately, we brought together a group of 11 different funder organizations. This interest from potential funders created new support for my exploration particularly in terms of ST commitment, but also in establishing urgency and champions for adapting the model (in order to ensure project success). Another driver of support for my work was that the CTE projects were the first formal explorations of PFS in U.S. K-12 education. Given the young nature of the field, and the relatively small number of competitors, being a “first mover” in an issue area is critical to developing a competitive edge for future work in that area. Thus, the ST cared about the outcomes of the education projects, the learnings across them, and the potential to expand in the sector moving forward, further increasing ST commitment to my project. Although the funding, political, and competitive landscapes did support my work, some key contextual realities challenged that support. First of all, all six of the projects were funding by one-time grants. This created a sandbox for exploration, but it also meant that although local jurisdictions and nonprofits were sufficiently interested in PFS to apply to participate in the work, none had actually “demanded” the work or committed to it financially. This put each project at a disadvantage, because we were in a position of “selling” our ideas to local jurisdictions, rather than advising those jurisdictions on how they could address their most critical issues through outcomes-based financing. The need to establish buy in from local jurisdictions slowed our work across the projects, such that instead of making specific financing recommendations, we identified a set of high potential 62 financing structures for each site, which we plan to explore in subsequent phases of work. This delayed progress limited our ability to have full confidence in a vision of what will work best in the space in the future. Looking across value, capacity, and support, the reason for my progress in establishing urgency, establishing champions for adapting the model, building ST commitment, and developing the starting point for a vision becomes clear. As I hypothesized in my theory of change, these outputs were critical in leading to two of my intended impacts: successful outcomes of the individual projects and the creation of a foundation for Social Finance for future opportunities in the space. Further, understanding some of the detractors that I experienced in terms of value, capacity, and support for my project also helps unpack the reasons that I made less progress in other areas, such as establishing a coalition of champions for future education work, finding opportunities for shared early wins, and developing a comprehensive and co-created vision for Social Finance in education. Again, looking back to my theory of action, these areas where I made less progress are largely responsible for my limited progress in my third intended impact, of creating a clear strategy for the firm in its future education work. In the Introduction of this Capstone, I outlined two critical questions that I hoped my work would begin to address. In the next sections, I reflect on the implications of my strategic project on each of those questions. My first question was: how does PFS apply in K-12 education, and what might need to change in order to expand the tool’s relevance and potential impact in the space? In Section 5, I address this question from the lenses of both the PFS field and the K-12 education sector. My second question was: what role should Social Finance play in education: is there a feasible and sustainable line of work that the firm 63 should pursue? In Section 6, I discuss the implications of strategic project for Social Finance, addressing this question and also reflecting on what the firm can learn from my project even outside of its education work. V. Implications for Sector Implications for the Pay for Success Sector My strategic project has two primary implications for the PFS field. Although grounded in my education-related learning, these implications are likely relevant to a range of other issue areas and contexts. Application of PFS to Education. The results of my strategic project imply that the traditional PFS model of the SIB has limited reach within education contexts. In general, education interventions are backed by too little evidence (which would make potential projects very high risk to investors), too great a time-lag between the intervention and the value-generating outcomes (which would require long deal times, and dilute the present value of the deal for all stakeholders), and too much disconnect between stakeholders who pay for and implement education and those that benefit (or suffer) from the outcomes of education (which limits incentive for any stakeholder to sign on as an outcomes payor). As a result of these challenges, I found that a SIB would not be feasible across any of the six projects that were the focus of my residency. This implication does not suggest that a SIB will never work in the K-12 education setting. In fact, one key insight from this work is that we need to take a more nuanced view of education when thinking about how outcomes-based financing might apply. Education interventions with a clear link to post-secondary and labor market outcomes – both in terms of having demonstrated impact on those outcomes and a short time proximity between the intervention and students achieving those outcomes – may be 64 well-suited for a traditional SIB, with Departments of Higher Education and Labor, post- secondary institutions (e.g., public community colleges), or employers participating in the outcomes payor role. However, overall my project did confirm the key critique outlined in Section III, that the traditional SIB has narrow relevance across the full field of K-12 education (and, likely, other issue areas with similar challenges). How PFS can adapt for increased relevance. In spite of my finding that the SIB has a narrow application in K-12 education, my work suggests that there is opportunity for the PFS field to have far more reach and impact in education by expanding from the SIB to additional outcomes-based funding models. At the heart of the SIB are two core principles. First, the tool is meant to decrease the risk to government of trying new social services programs. Second, the tool is meant to directly connect dollars to outcomes, rather than delivery of services. In Figure 7, I outlined four adaptations to the SIB that we explored. All four of these adaptations maintain these two fundamental principles of PFS, and have a higher potential for impact in education, and likely in other sectors as well. A critical question for the field, then, is how its definition will evolve. If PFS continues to be defined by the SIB, my strategic project supports some of the critiques, and suggests that the impact of the field is truly limited. However, with a broader definition that gets to the intent of the SIB, but incorporates a range of tools that can be deployed based on a given political and programmatic context, my strategic project suggests great opportunity for the field to continue to expand. The clarification and expansion of this definition is becoming increasingly important as PFS is included in more Federal and State policies. For example, on February 9, 2018, the Federal government passed the Social Impact Partnerships to Pay for Results Act, a huge 65 milestone for the PFS field. This bill provides $100 million in funding to support PFS projects for State and local governments, including support for feasibility, evaluation and outcome payments. The language in the policy, though, is highly reflective of the SIB. If interpreted narrowly, this policy will only be used in contexts that fit the stringent requirements for the SIB, limiting its relevance and potential impact in education and beyond. As the field moves forward, critical leadership will be required by Social Finance and other field-builders to help other stakeholders (particularly policymakers) accept this broader definition by demonstrating its expanded potential for impact. Implications for the K-12 Education Sector Although my strategic project implies a need for the PFS field to adapt, my exploration also uncovered sector-level barriers that will continue to limit the application of PFS in education, even with an expanded definition. The two most significant barriers that I encountered in my strategic project (described in Section IV and depicted in Figures 5 and 6) were the limited rigorous evidence base and the disconnect (in time and priorities) between funders and beneficiaries of high quality education (i.e., the “wrong pockets” problem). These challenges do not just limit the education sector from benefiting from PFS; they are also broader systemic issues that limit our overall impact on student outcomes. My strategic project highlighted critical system-level conversations required to increase the PFS-readiness of the field, and to optimize its impact. At the same time, our projects demonstrated that in spite of these systemic challenges, the analyses behind PFS can bring significant value to the sector. This leads to a more optimistic implication of the work, which suggests key tactics that service providers and districts can take to adopt an outcomes-based approach in their daily activities. 66 Broad Systemic Implications. My project highlighted opportunities to increase impact by addressing two key gaps in the sector:  Evidence. We need more research to understand what drives impact for students. The vast majority of education interventions have no formal evaluation to demonstrate their impact. As nonprofits and schools continue to innovate, it is critical that we as a sector identify which of these innovations are effective and what drives their impact (whether that be particular programs, interventions, or tactics; characteristics of school or classroom settings; qualities of educators; or other factors), so that we can bring solutions that work to more students, and so that we don’t waste resources on ineffective efforts. Rigorous research methods such as RCTs, and even alternative evaluation methods such as propensity matching and pre/post analyses, are expensive, complex, and often take years to complete. Although in many cases our students cannot wait for this research to spark change, this should not deter the sector from investing in that research alongside the innovation. By doing this, we can both support students today and determine which supports will have the most impact for students tomorrow. From a PFS lens, this also opens more opportunity in the long run to expand programs that work through innovative, risk-based, financing mechanisms. As I have discussed, some education programs have been tested through rigorous research. For these interventions, the sector would benefit from better dissemination of that evidence, and more ways to expand those programs (PFS is one way to expand evidence-based programs, but there are many alternative pathways to replication that do not require PFS-like transactions). Some organizations are working to address this issue – for example, Mission Measurement is focused on aggregating and synthesizing 67 social sector data to guide decision makers in investing in evidence-based programs – but significant opportunity remains to elevate programs that have been proven to work. It is important to note, in discussing the need for evidence, that there is a vast range of definitions of the intended impact of education programs for students (e.g., improved citizenship, post-secondary success, labor market outcomes). This definition of impact is a critical pre-requisite to ensure that we are building evidence in a coherent and meaningful way.  Stakeholder Alignment. A critical challenge in aligning on a definition of impact is that the benefits or value of high quality education are dispersed across a range of stakeholders (e.g., post-secondary, labor), each focused on different educational outcomes (i.e., the ones that are most meaningful for their own institutions)8. The result of this dispersed value is that these stakeholders engage in education either not at all or in very narrow ways. Perhaps in response, the education sector has developed its own set of outcomes (in the form of standards and assessments) that, although increasingly intended to address post-secondary and career needs, shifts often and does not directly link to any external stakeholder’s priority. Thus, when we talk about building evidence of impact in the education sector with a cross-sector audience (such as in PFS conversations), we are often talking about a diverse range of outcomes that we seek to achieve through educational programs. The education sector would benefit from working together with external stakeholders who are impacted by education – such as post-secondary, criminal 8 For example, a higher education stakeholder may focus only on interventions increase post-secondary persistence or industry credentialing, whereas a justice department stakeholder may focus only in those that directly decrease students’ engagement with the juvenile justice system. 68 justice, labor, and even healthcare stakeholders – to align on a shared vision and set of impact goals for U.S. K-12 education. Right now, the education sector does a lot of measuring, but without focus. By defining this “North Star” of the impact we hope to achieve through education – and the outcomes that are associated with that impact – education providers could focus their strategies on reaching those outcomes, and by being more intentional, would likely have more success in reaching them. This approach may also attract more stakeholders – who reap the benefits of our job well done – to engage and support our work. Tactics for Districts and Nonprofits. In spite of the challenging system-level implications raised above, my work suggests that districts and nonprofits can improve their impact in the near term with two tactics that emerged across all six projects:  Track costs and outcomes more often and more rigorously. Earlier in this section, I discussed the need for more rigorous evaluation. A first step in that direction (which is a large undertaking in itself) is finding ways to better track what we are already doing in the sector, both in terms of costs and outcomes. On the district side, this information is more accessible than for nonprofits. Taking a PFS lens, the key shift in thinking that can help a district make evidence-driven decisions is connecting information on costs and student outcomes to specific programs, interventions, or student experiences. Although this is complex – for example, it is difficult to disentangle the cost of a teacher’s salary to specific parts of a student’s experience – any efforts to make those connections will help districts understand the link between the dollars they are spending, the experiences they are creating, and the outcomes they are helping their 69 students achieve9. Further, to the extent possible, finding ways to track student outcomes once they leave the district (e.g., through National Clearinghouse Data or partnerships with public post-secondary institutions), would also support districts in using data to drive decision making. For nonprofits, tracking is even more difficult, because most nonprofits do not have easy access to a complete view of the data. On the cost side, we found that many of our nonprofit partners had limited insight into the financial implications of their programs for schools or districts (outside of the charges paid directly to the nonprofit). Further, there was a large range in their abilities to understand how their services were actually impacting students; some had their own tracking systems or some degree of data-sharing with district partners, but most had very limited information on outcomes for their students at and beyond secondary school. Thus, a key implication of my work for nonprofits hoping to engage in PFS projects is that they should set data-sharing partnerships in place as early as possible, and seek to gain as much insight as possible into the full costs and benefits of their programs10.  Proactively bring together community stakeholders. Our clients in one district reflected to the team that our project created a reason for the community to come together. The project team, on the other hand, reflected that all of the community stakeholders that did come together were already close contacts of the district. This has huge implications for districts in particular: many districts already have 9 Particularly in situations where schools implement discreet programs (such as career academies), it is likely possible to estimate what share of teacher time (and other resources) are focused on that program vs. other more general school activities. 10 An alternative way for nonprofits to collect this information is to conduct formal evaluations of their programs, and include cost-benefit analyses within those (although more costly, this would also support research-building in the field!). 70 relationships across community stakeholders who could engage in funding or help to shape high impact programs. Ultimately, there is no inherent reason that a PFS project is needed to bring those stakeholders together. Rather than engage employers, higher education, county and city level stakeholders, and others separately, districts should consider opportunities to bring the full group of stakeholders together around shared goals. In the case of our district partner, the impact of having this group together was that the district was able to better tailor its agriculture pathways program to labor market needs, and was able to engage a set of local employers to fund the upfront costs of the new program. I would conjecture that this engagement could have happened without our team, and with no outcomes-based financing lens. However, the status quo for our partner (and in many educational settings) was that these conversations happened infrequently and in a fragmented way. Districts and nonprofits need to find ways to trigger a new, more collaborative model of engagement. PFS and other cross-sector partnership models (for example, collective action efforts) can often catalyze this shift, because they explicitly define a shared vision and enforce collaboration. A key question for districts and nonprofits to consider in their own contexts, then, is whether it is possible to trigger this shift internally, or what external forces might support this shift. Cross-sector Implications In addition to having implications for the Pay for Success field and the K-12 education sector, my strategic project has two implications that apply across sectors. Fixing the Entity Problem. Perhaps the most significant challenge that arose across my projects was the lack of a government stakeholder who felt the benefits of 71 improved educational quality to a degree that sparked willingness to pay for improvements. Throughout the Capstone, I have discussed the disconnect between stakeholders who pay for education and those who reap the long-benefits of it; one cross- sector implication that emerges in considering this issue is the need for an overarching entity that connects these stakeholders and takes on responsibility for driving improvements. There is a key question around whether this entity would focus only on coordinating across public sector bodies (and, therefore, perhaps be a government or public body itself), or whether it would focus on coordinating across the public, private, and nonprofit stakeholders involved in this work. If structured with a public-sector focus, one could imagine this role being played at the County level or through a Special Office of the Mayor or Governor. If structured with a focus across sectors, one could imagine this entity looking more similar to a backbone organization in a collective action effort (Turner, Merchant, Kania, & Martin, 2012). I imagine that a public-sector focused entity would be more influential in driving PFS and other outcomes-oriented financing structures. A public sector body, particularly one that is empowered through budget and staff resources by a local leader, is likely to have more influence over other public sector stakeholders than a backbone organization, which is typically a nonprofit. A public sector entity, particularly one sitting at the central level, could convene stakeholders across departments around a shared agenda, and may have more insight into the potential for shared data and financial decision making than a non-governmental body would have. There is a question around where this entity would be anchored in its mission. For example, the entity could have a charge of improving the local education system and 72 focus on identifying evidence-based education programs that accrue value across public sector stakeholders in the community. Alternatively, the entity might have a charge of increasing economic mobility or labor market supply, and focus on identifying programs – in and beyond the K-12 education system – with evidence of improving those long-term outcomes. I believe that an entity with a long-term outcome mission would be better able to successfully activate PFS-like projects. An entity grounded in an education-specific mission would likely face the same challenges that we faced across our education projects in connecting a given K-12 effort to long-term financeable value. On the other hand, an entity anchored in a specific outcomes mission would be able to identify the set of efforts that might drive the impacts that education programs seek, but may not be able to deliver on alone (that is, without additional support from programs sitting in other areas such as mental health, child welfare, and housing). By taking a cross-sector approach to addressing societal issues, this entity could simultaneously make the connection between education and long-term value, and spark a coordinated path towards outcomes-oriented change. There are many additional questions that would be to be addressed before launching this type of overarching entity, such as whether the entity would have its own budget or would mobilize the budgets of others, and what specific activities it might engage in relative to existing public sector bodies. However, my strategic project suggests that exploring these types of cross-cutting entities, ideally within the public sector, would help catalyze more coordinated and effective outcomes-based financing structures to support education efforts and their long-term outcomes. 73 Embedding Patience in Change Leadership. Reflecting on my experiences over the course of the strategic project, and how I interpreted my progress both during the residency and in this Capstone, it seems that there may be a critical element missing (or de-emphasized) in the change literature, which is the need for patience along the change journey. Few articles on change leadership suggest a specific timeline, which may lead each individual change leader to a different expectation of how long it should take to drive a change. Of course, this surely differs based on context, but in my experience, I believe that ten months was likely too short of a time to expect to identify what needed to change about the PFS model in the K-12 education context, adapt the model, and then develop a comprehensive strategy for future work. Because I set such high expectations, I experienced any movement that was not directly headed towards success as a set-back or constraint to my success. An alternative perspective is that all of the challenges and set-backs that I and my teams faced throughout my strategic project were actually critical steps forward in the change that I was seeking to lead. The moments of perceived failure were critical in establishing urgency to adapt the model, and also gave team members confidence that the challenges they were facing were not due to a fault of their own, and did not imply that a solution was impossible. Instead, these moments of shared “head banging” helped teams vent some frustration and more quickly switch from tactics that were not working in any of our education scenarios. Thus, my strategic project suggests that the change management literature should more explicitly call out the need for patience in change leadership and the role of perceived failure in making progress towards change. By being patient, and expecting 74 failure to occur, change leaders can turn situations that feel like backwards progress into opportunities for learning and, ultimately, shift to more effective paths towards the changes that they seek. VI. Implications for Site Social Finance launched and built the PFS field in the U.S. The current definition and application of PFS, therefore, is largely driven by Social Finance’s efforts and activities over the past seven years. Thus, the broad implications of my strategic project for the field directly translate to implications for Social Finance. However, in many ways, Social Finance is much bigger than the PFS field. While the field has focused on testing and refining the SIB, Social Finance has continued to innovate, grow, and expand its focus. Thus, my project has implications for Social Finance beyond those for the field. Even if the PFS field remains unchanged, Social Finance has significant opportunity to expand its own impact and potentially define a new, broader field of outcomes-based financing beyond PFS and the SIB. Further, although my focus was on Social Finance in education, my project also has implications for Social Finance beyond its engagement in the education space. Social Finance in Education I set out to help Social Finance understand how to adapt its model in K-12 education, and, critically, whether a targeted effort to expand in the sector would benefit the firm. Below, I discuss the implications of my project on both questions. How Social Finance Should Approach K-12 Education. Three key takeaways have emerged from my work about Social Finance’s future work in the K-12 space:  Approach PFS feasibility with a broader outcomes-based financing goal. My project suggests significant opportunity for Social Finance to bring value in the K-12 75 education sector by expanding its toolkit of funding models from the traditional SIB. By framing our work as helping to develop a range of outcomes-based financing structures, such as the four adapted models that we explored in my strategic project, Social Finance can not only have impact for a broader set of education programs and interventions, but the firm can also lead the PFS field in expanding its definition (and relevance) from the traditional SIB.  Segment the K-12 sector based on outcomes. One of the major insights of my residency was that our projects with the most tangible and measurable links to employment outcomes generated the most engagement from stakeholders outside of the education community. Although districts and States have specific departments focused on “K-12 education” broadly, PFS works at the outcomes level, and different outcomes matter to different stakeholders. Thus, as Social Finance moves forward (ideally with a broader outcomes-based financing frame, as suggested above), it will benefit the firm to segment the sector at the outcomes level. My project suggested that programs with direct links to college and career outcomes are high potential for outcomes-based financing models, and with further exploration in the space, Social Finance may identify additional high priority segments11.  Ability to support education partners with a broader set of tools. Another insight from my project was the immense value that Social Finance can to bring to education nonprofits and districts, even outside of the structuring of outcomes-based financing deals. We heard repeatedly from our project partners that our stand-alone analyses, 11 Of note, there have also been indications through Social Finance’s Pre-K work (and the launch of two Pre-K SIBs in the U.S.) that Pre-K is another high potential area; this was not the focus of my Capstone or exploration, but should be included as a high potential “segment” of education in future strategy work. 76 such as cost-benefit and ROI, led to significant progress in their thinking about growth and decision making. The firm is already exploring ways to disaggregate its offerings for government partners through its PII effort; by approaching education through this lens, Social Finance may be able to have even broader impact in the space. In addition to these three implications, my strategic project highlighted the fact that Social Finance still has much to learn about how to best have impact in education. First, we need to explore education work that begins with a potential payor. All of our projects were driven by nonprofit or district service providers. This resulted in an inefficient approach to feasibility because our teams had to consider benefits to a range of potential payor stakeholders. A key next step will be to understand how a project might look different if driven by a Department of Higher Education, Department of Labor, or employer. Grounding a project in their priorities, and engaging them from the outset, may better set a project up for success. This is not news to Social Finance; they are well aware that projects are faster and often more impactful when they start with the payor. Our work has highlighted specific stakeholders who benefit from high quality education, and would be high potential payors on whom to focus our efforts. Second, we need to explore education interventions that sit outside of the core K- 12 system, such as after-school, summer, and wrap-around support programs. These programs are funded through different funding flows than the per-pupil dollars that support core activities within a school day. Further, outcomes of “non-core” programs may be easier to track and attribute because the interventions are “add ons” as opposed to shifts in the educational experience that a child already receives. These factors may create 77 the opportunity for more evidence (and therefore less risk) and a less complex set of potential payors with more flexible or discretionary funding options for PFS. Whether Social Finance Should Expand in K-12 Education. Although my strategic project highlighted a range of answers to how Social Finance can work in education, the question of whether a targeted expansion in the sector is beneficial is more complex. Our work across six different education contexts highlighted the value that Social Finance’s outcomes-based approach brings to education service providers, but also the significant adaptations required to the SIB model and feasibility approach in order for the firm to have impact in the space. In order to determine whether the increased challenge of working in the space is offset by the value of doing so, critical questions must be answered around whether there is a viable business model for Social Finance in education and whether a targeted expansion fits within the firm’s mission.  Business model considerations. Our experience across six projects suggests strongly that districts are unlikely to be able to act as payors in PFS transactions, and even less likely to find funding to evaluate whether a transaction is possible. Thus, a critical next step for Social Finance is to explore what types of stakeholders will pay for the development of these projects. From my experiences across the six project contexts, I believe there is opportunity to seek feasibility funding from two primary sources. First, foundations may be high potential funders for both feasibility assessments and more stand-alone “PII-like” work in the education space. As mentioned in the Introduction, foundations channel significant resources into the sector annually. Funding PFS feasibility and readiness offers foundations an opportunity to identify longer-term, sustainable solutions to support their target outcomes. I suspect that the current 78 scarcity of philanthropic support for PFS feasibility in K-12 education has been driven by the perceived challenges of applying the tool in the space. If Social Finance could expand the field’s mindset around how outcomes-based financing and thinking might support education programs, we may see increased interest from philanthropies to support our work. Relying on philanthropic support is not an ideal business model, as it may not create the buy-in from education partners and other PFS stakeholders that is required for long-term implementation of a project. However, there may be a model in which these stakeholders contribute some resources in partnership with a foundation. Of note, over half of Social Finance’s current advisory work is funded by foundations, so while unideal, this approach is already a key component of our business model. The second potential funding source for Social Finance’s future education work is States (and counties where relevant12). Critically, the work should be based at the central level, not solely within the State Education Agency. As my work highlighted, a core challenge for PFS in education is that stakeholders who benefit from high quality education are not the same as those who currently pay for and deliver education. By launching PFS exploration from a cross-stakeholder perspective (that is, at the State or county level) it may be possible to “connect the pockets,” and put outcomes at the center of the work. Further, State-level budgets often have more flexibility than those of education stakeholders to support PFS exploration.  Mission and strategy considerations. While solidifying a viable business model for Social Finance in education is critical, another important question for Social Finance to consider in determining whether to pursue more education work is whether the 12 For example, in California, counties hold significant power and resources and can be important levers for change. 79 types of projects that are feasible in education support the firm in achieving its mission. Reflecting on my project work, I would suggest that the adapted outcomes- based financing models that we explored do fit within Social Finance’s mission. However, the idea that Social Finance could bring distinct value to education partners through more stand-alone offerings such as cost-benefit and ROI analyses leads the firm towards a grayer area around this mission. In fact, on some of my projects, team members raised the question of: if this doesn’t lead to some form of outcomes-based funding (or “mobilizing of capital”), why are we here? The firm’s new foray into offering a menu of services through the PII effort suggests that the firm is moving towards a broader interpretation of its mission. As it moves forward, it will benefit the firm to clarify this new interpretation and define guardrails around what types of work fit into a broader strategy aligned with that interpretation. As our mission and strategy expand beyond the traditional SIB, there is more rationale for continued effort within education. A second strategy-related consideration that Social Finance must address in determining its next steps in education, is whether it should maintain its issue area agnostic approach. In earlier years, this approach was necessary to seize opportunities that arose to build the organization and the field. As the field expands, new competitors will likely enter with content expertise, and jurisdictions may seek that expertise in their PFS partners. This will likely be the case in education, where funding flows and interventions are complex, a large number of potential competitors with expertise may exist if education consulting firms develop PFS expertise, and education stakeholders may be resistant to a new financing mechanism tied to private investors 80 and profits. This does not necessarily imply that Social Finance should organize itself around issue areas, but it does suggest that building sector-level expertise – and, potentially, sector-specific business plans – may become more important as the firm and field continue to grow. Overall, my strategic project suggests that Social Finance should continue to explore education work, with a particular focus on segments within education that have a more natural fit with the broader PFS philosophy (that is, strong evidence base and closer link to the specific outcomes that matter to potential payors or funders). There is also opportunity to pursue a broader set of offerings that might impact education programs across the space, but before expanding in this direction it is critical to identify funders for the exploration, and clearly define the firm’s increasingly broad mission and strategy. Next Steps for Social Finance in K-12 Education. As Social Finance considers further exploration in K-12 education, my learnings over the course of my residency lead me to four suggestions of areas within education that Social Finance might pursue next:  Segments that directly connect to outcomes with demonstrated societal value. As I mention above, my project suggested that programs with more direct links to career and post-secondary outcomes are riper for PFS funding, as they offer clearer value propositions to stakeholders who care about those outcomes. Moving forward, I would direct Social Finance towards segments that share this characteristic. Education programs that directly work with justice engaged youth – if they can demonstrate reduction in future engagement with the criminal justice system – will likely attract engagement from criminal justice stakeholders. Education programs that directly address child development and health will likely attract engagement from healthcare 81 and child welfare stakeholders. By taking the lens of the stakeholders who derive measurable value from education programs, and exploring the opportunity to expand these niche areas through PFS, Social Finance can build its experience and credibility in the education space and also leverage the expertise and network that it has already built in other issue areas.  Extremely high need populations. Outcomes-based financing transactions are built around a return-on-investment, which is calculated as the expected increase in social impact from an intervention minus its costs. One area of “low hanging fruit” for Social Finance’s exploration is to focus on educational programs that target sub-populations with extremely low baseline outcomes. For example, whereas the national average rate for post-secondary attainment is about 26%, the average post-secondary attainment rate for foster youth is under 10%, and for youth engaged in the juvenile justice system that rate is likely below 5%13 (National Center for Education Statistics, 2017; National Working Group on Foster Care and Education, 2014). Long term outcomes for these sub-populations are also challenging: many face higher incidence of, for example, future engagement in the criminal justice system and unemployment, which are costly to both the individuals and to society. Focusing on these high need populations may lead to a riper opportunity for a PFS project for a few reasons. First, there is more opportunity for a high benefit side of the cost-benefit analysis. Whereas for the average population, the most impact a program might have is a 76 percentage point increase in post-secondary attainment, for these sub-populations the potential impact of a program 13 There is limited data available on actual college completion rates for this population. However, research suggests that less than 20% graduate from high school, and using the (likely high) drop off rate for the national population (roughly 1/3rd of students who graduate from high school go on to graduate from post-secondary education), we can deduce that a very small percentage of juvenile justice engaged youth attain a post-secondary degree. 82 could be as high as 95 percentage points. In addition, the quantified value to society of improving outcomes for these populations is also higher (per outcome) than the value of improving the same outcomes for the average population, as it includes a higher rate of avoidance of those negative long-term outcomes. Separately, it is likely that some segments of these sub-populations can be supported through relatively low cost programs (e.g., for some, access may be the biggest challenge). Thus, these projects may lead to higher benefits at low cost, translating to quite high potential ROIs.  Interventions with existing rigorous evidence. Although the number of educational programs with specific RCT or quasi-experimental evidence is limited, these programs provide low-hanging fruit for Social Finance’s early exploration. In addition, leading research organizations such as MDRC, J-PAL and the Institute of Education Sciences are all currently pursuing RCTs and other rigorous research methods in the K-12 space. For example, MDRC conducted the original RCT on Career Academies in the early 2000s, and is currently exploring a replication and expansion of that research focused on California Partnership Academies (MDRC, 2018). Other programs or interventions with existing rigorous research include Success Academies (in the K-2 space), New York City’s Small Schools of Choice, and Early College High School models14. By continuing our exploration of PFS in education with an explicit focus on interventions that already have this evidence base, and by partnering with leading research organizations to increase our knowledge base of new programs with rigorous evidence, we can enter our next projects with one less critical challenge to address. 14 For more details on this research see: Success Academies < schools-evaluation#overview>; NYC Small Schools of Choice <>; Early College High School <>. 83  Developing an offering or set of partnerships that help build the evidence base. Another way that Social Finance can support future PFS projects in the K-12 education space is by finding ways to support more rigorous research in the space now. As part of the PII initiative, the firm is already exploring offerings for governments focused on the early stages of performance measurement, such as building data systems, conducting programmatic analysis, and developing governance plans. In the K-12 education space, there is opportunity to offer these types of services to government stakeholders (e.g., state or local education agencies, with a focus on particular programs within their districts or schools) or to expand the offering to service providers who partner with districts to deploy those programs. This would support education stakeholders in developing an evidence base that could be used to build confidence for future projects. Of note, there is significant overlap in these suggestions in terms of populations and programs on which to focus. For example, focusing on programs with existing evidence of improving post-secondary attainment for juvenile-justice engaged youth – which can be associated with decreasing rates of future criminal justice engagement – would achieve all of the first three suggestions. Thus, although these suggestions offer Social Finance a number of paths forward, it is possible to narrow in on a short list of the highest potential areas for immediate focus. Social Finance beyond Education My strategic project was intended to explore Social Finance’s work in education, but the work also ultimately has implications for Social Finance beyond the sector. I have already highlighted one broader implication above, around the firm’s mission and 84 strategy; below, I highlight additional implications for Social Finance’s “product” and how it approaches innovation. Implications for Product. When I began my residency, Social Finance was in the midst of thinking about ways to increase its impact and sustainability by offering a broader range of services in line with the PFS philosophy. The PII effort, which I have mentioned several times, is specifically focused on turning Social Finance’s PFS feasibility “product” into a menu of products and services that the firm can offer through its advisory practice. When I arrived, this effort was still a plan on paper, not an active business line. Throughout my residency (and separately from my strategic project), the firm moved the effort forward, launching a series of targeted business development efforts and starting a few projects specifically within the initiative. Although my strategic project was not explicitly a part of this effort, my work across education projects very much reflected this idea of a menu of services. In most projects, our education partners found significant value from specific pieces of our work that held regardless of whether an outcomes-based solution was reached. Thus, my strategic project has created a series of case examples for the PII effort, and has supported the PII thesis that there is significant opportunity for the firm to expand its suite of services to have real impact across issue areas that (like education) may not be best suited for the traditional SIB. By viewing our services as a set of tools that partners can select from, rather than trying to force each situation into a SIB framework, we have the ability to work with a much larger set of partners on a broader range of issue areas. Implications for Innovation. Innovation and adaptation are critical to the long- term success of Social Finance. We confirmed through my strategic project that the 85 traditional SIB is too narrow to drive impact in most education contexts, and, likely, in many other issue area contexts. If Social Finance only worked on SIBs, therefore, it would have a limited set of potential projects. The firm already recognizes this, and is exploring a range of innovations and ideas to expand its scope of work. One key issue, though, is that this innovation is occurring in many different facets simultaneously, and it is challenging to uncover meaningful learnings from any individual innovative foray. Right now, Social Finance is exploring innovations in environmental impact transactions, charter school facilities financing, income sharing agreements, outcomes rate cards, and the PII effort discussed above. Although there are teams dedicated to each of these efforts, it is hard to innovate across all of those areas meaningfully at once. What happened during my residency was (inadvertently) a new approach to innovation at Social Finance. Rather than experiment with one new idea on one project, we actually had a “sandbox” of education projects to learn from together. This was largely based on circumstance; the CTE project funding enabled us to launch four projects at the same time focused on very similar programs, and my residency created a reason to pursue the other education projects that opportunistically arose. However, because the projects occurred at the same time, and in part because my residency forced me to ask teams to look across the projects in a slightly more intentional way than may have otherwise occurred, we were able to gain learnings that span six data points and carry significant weight in thinking about what the next set of similar work might look like. Thus, the final implication of my project for Social Finance, beyond its work in education, is to try to strategically create more sandboxes for learning. There is room to 86 explore how a “sandbox” is defined. In my context, the boundaries of the sandbox related to an issue area. By working across projects in one issue area that was new to the firm, we were able to learn about nuanced challenges within education, and work together on those challenges across projects. Further, we were able to leverage content-specific learnings (such as how to quantify the value of a high school graduation) across teams. Thus, my strategic project suggests significant opportunity for learning and innovation by taking an issue area approach to defining the sandbox, which I believe would apply across a range of issue areas as Social Finance continues to expand its reach. Another question that Social Finance might explore is the potential to define the boundaries of a sandbox around a tool or offering. Right now, the firm is exploring a set of projects focused on developing outcomes rate cards, another new tool within the outcomes-based financing family. It will be critical for the firm to reflect on whether this type of tool-based sandbox enables the same sort of learning and innovation as my issue area-based approach. I imagine that working across projects focused on a single tool will offer some of this opportunity, although the degree to which teams will be able to problem solve together on challenges may vary depending on the degree of similarity or difference between issue areas. A key consideration is how to fund this type of innovation. It may be that this requires internal investment from general operating dollars or finding a funder who is willing to support a set of related projects, rather than just one (this is becoming more common; in fact, the rate card projects referenced above are, like the CTE projects, all funded from the same source). In general though, the firm will learn more by exploring a small number of innovations across multiple projects, rather than exploring a large 87 number of innovations in one-off ways. This “sandbox” approach will enable the firm to learn more meaningfully, and therefore to succeed or fail faster in its innovations. VIII. Conclusion When I began my residency, I had two goals. First, I suspected that the traditional PFS model would present significant barriers in education, and wanted to identify adapted approaches to address them. Second, given my personal passion for education, I wanted to prove that it made sense for Social Finance to focus more in the sector. I saw a huge need to increase the impact of the vast number of dollars being spent in education, particularly by government, and saw PFS as a hopeful solution to drive that shift. Through my strategic project, I was able to achieve my first goal. In partnership with my talented and committed project teams, I brought PFS thinking and analyses into six different education contexts. My work confirmed my hypothesis that change was needed, and we were able to explore a range of adaptations to the SIB to address the key challenges that arose. Our work drove significant impact for our education partners, and further increased my desire to bring an outcomes-based lens to more players in the field. In spite of my successes in adapting the model and bringing value to partners, I am not finishing my residency with a comprehensive recommendation on whether Social Finance should accelerate its work in the space, or a detailed strategy to approach such an acceleration. Reflecting on the value, capacity, and support factors that enabled and hindered me from fully completing my theory of change, I have renewed hope in a future ability to find clarity around both of those questions. Ultimately, six projects and ten months may not be a sufficient set of information to fully determine a business model and strategy, particularly given that the adaptation of the tool was a pre-requisite for the effort. Looking forward, I believe that my work has created a strong foundation for Social 88 Finance to attract new education partners, and I have hypotheses around what we need to explore in order to fully flesh out a strategy. Further, I am hopeful that my work will inform the firm’s broader strategy, and enable it – and the PFS field – to support a broader range of programs and jurisdictions across a broader set of issue areas, ultimately driving better outcomes for more and more individuals in the U.S. and beyond. 89 Bibliography Barrett, L. (2015, July 6). America Forward: State of Play on Elementary and Secondary Education Act. Retrieved from New Profit Blog: and-secondary-education-act Beer, M., & Nohria, N. (2000). Cracking the Code of Change. Harvard Business Review, 78(3), 133-141. Beer, M., Eisenstat, R. A., & Spector, B. (1990). Why Change Programs Don't Produce Change. Harvard Business Review, 68(8), 158-166. Belinsky, M. (2012, January 23). Social Impact Bonds: Lessons from the Field. Retrieved from Stanford Social Innovation Review: Berger, K. (2013, March 14). Over $1 Trillion to 1 Percent of Charities: How Do We Measure the Results? Retrieved from Huffington Post Blog: effectiveness_b_2701812.html Boggild, L., & Hlady, M. (2016, July 8). Funding Revolutions in Pay-for-Success. Retrieved from Stanford Social Innovation Review: Boykin, D. B. (2005, April 18). What Performance-Based Contracting Really Means for Procurement Goals. Retrieved from American City and County: contracting-really-means-procurement-goals Clarkin, J. E., & Cangioni, C. L. (2015). Impact Investing: A Primer and Review. Entrepreneurship Research Journal, 6(2), 135–173. Cohen, R. (2014, October 17). Does “Pay for Success” Actually Pay Off? The ROI of Social Impact Bonds. Retrieved from Nonprofit Quarterly: the-roi-of-social-impact-bonds/ Dear, A., Helbitz, A., Khare, R., Lotan, R., Newman, J., Sims, G. C., & Zaroulis, A. (2016). Social Impact Bonds: The Early Years. Boston: Social Finance, Inc. Downey, M. (2017, July 11). Paying Schools for Success. Retrieved from Stanford Social Innovation Review: Edmondson, J., Crim, B., & Grossman, A. (2015, October 27). Pay-For-Success is Working in Utah. Retrieved from Stanford Social Innovation Review: Ferrazzi, K. (2014). Managing Change, One Day at a Time. Harvard Business Review, 92(7/8), 23-25. Garvin, D. A., & Roberto, M. A. (2005). Change Through Persuasion. Havard Business Review, 83(2), 104-112. Georgetown Center on Education and the Workfoce. (2014). Recovery: Job Growth and Education Requirements Through 2020. Washington D.C.: Georgetown Public Policy Institute. Retrieved from content/uploads/2014/11/Recovery2020.ES_.Web_.pdf 90 Giving USA. (2016). The Annual Report on Philanthropy for the Year 2015. Researched and written by Indiana University Lilly Family School of Philanthropy. Sponsored by Giving USA Foundation, a public service initiative of The Giving Institute. Gustafsson-Wright, E., & Atinc, T. M. (2014, April 2). Social Impact Bonds for Early Childhood Development: Making "Dollars and Sense" in Developing Countries. Retrieved from Brookings Blog: development/2014/04/02/social-impact-bonds-for-early-childhood-development- making-dollars-and-sense-in-developing-countries/ Hanley, K., Nabi, P., Bean, K., Radday, A., & Stanzler, a. C. (2012). Youth Career Development: Social Issue Report. Boston: Root Casue: Social Impact Research. Kim, W. C., & Mauborgne, R. (2003). Tipping Point Leadership. Harvard Business Review, 81(4), 60-69. Kohli, J. (2010, November 18). Financing What Works: Social Impact Bonds Hold Promise. Retrieved from Center for American Progress: g-what-works-social-impact-bonds-hold-promise/ Kotter, J. (1995). Leading Change: Why Transformation Efforts Fail. Harvard Business Review, 73(2), 59-67. Leonard, H. B. (2002). A Short Note on Public Sector Strategy-Building. Harvard Business School, John F. Kennedy School of Government. Cambridge, MA: Herman B. Leonard. McIntyre, E. (2015, December 14). The controversy behind ESSA's Pay for Success initiative. Retrieved from EducationDive: success-initiative/410619/ McKeever, B. S., & Pettijohn, S. L. (2014, October). The Nonprofit Sector in Brief 2014. Retrieved from Urban Institute: Nonprofit-Sector-in-Brief--.PDF MDRC. (2018, April). Next Generation California Partnership Academies: Overview. Retrieved from partnership-academies#overview Mesecar, D., & Soifer, D. (2016, February 24). How performance-based funding can improve education funding. Retrieved from Brookings Blog: performance-based-funding-can-improve-education-funding/ Meyerson, D. E. (2001). Radical Change, the Quiet Way. Havard Business Review, 79(9), 92-100. National Career Academy Coalition. (2017). About Career Academies. Retrieved from National Center for Education Statistics. (2017). Digest of Education Statistics. Retrieved from; National Working Group on Foster Care and Education. (2014, January). Fostering Success in Education: National Factsheet on the Educational Outcomes of 91 Children in Foster Care. Retrieved from content/uploads/2012/05/National-Fact-Sheet-on-the-Educational-Outcomes-of- Children-in-Foster-Care-Jan-2014.pdf Office of Federal Procurement Policy, Office of Management and Budget, Executive Office of the President. (1998). Best Practices for Performance-Based Contracting. Retrieved from The White House: Orszag, P., & Bridgeland, J. A. (2013). Can Government Play Moneyball? Atlantic, 312(1), 62-66. Palandjian, T., & Shumway, J. (2015). Response to The Payoffs of Pay for Success. Stanford Social Innovation Review, 13(4), 37-38. Pettus, A. (2013). Social Impact Bonds. Harvard Magazine, July-August, 11-12. Retrieved from Harvard Magazine. Rangan, V. K., & Chase, L. A. (2015). The Payoff of Pay-for-Success. Stanford Social Innovation Review, 13(4), 28-36. Ray, B. (2016, March 4). What to Make of Social Impact Bonds. Retrieved from Medium: Crosswalk Blog: impact-bonds-f2274210235c Rosenberg, T. (2012, June 12). The Promise of Social Impact Bonds. Retrieved from The New York Times: The Opinion Pages: bonds/ Rosenman, M. (2014, February 19). How social impact bonds put private profit ahead of public good. Retrieved from PBS News Hour: ahead-public-good Saltman, K. J. (2017). The Promise and Realities of Pay for Success / Social Impact Bonds. Education Policy Analysis Archives, 25(59), 1-21. Sirkin, H. L., Keenan, P., & Jackson, A. (2005). The Hard Side of Change Management. Harvard Business Review, 83(10), 108-118. Social Impact Research. (2012, April). Retrieved from Youth Career Development: Social Issue Report: Career-Development/Youth%20Career%20Development- %20Social%20Issue%20Report.pdf Turner, S., Merchant, K., Kania, J., & Martin, E. (2012, July 18). Understanding the Value of Backbone Organizations in Collective Impact: Part 2. Retrieved from s_in_collective_impact_2 U.S. Department of Education. (2017). The Federal Role in Education. Retrieved from Watkins, M. D. (2016). Leading the Team You Inherit. Harvard Business Review, 94(6), 60-67. 92 Appendix A. Evolution of “Intro to PFS” Materials Sample of Intro to PFS” slides used in Kick-off (6th education project) South Bay TRADITIONAL VIEW: WHAT IS PAY FOR SUCCESS? Expansion capital ($) Private funders / impact investors Service provider Repayment ($) Outcomes Payor (often government) Pay for Success is about measurably improving the lives of people most in need by driving government resources toward more effective programs Social Finance, Inc. © 2017 Confidential SOCIAL FINANCE IS EXPLORING A RANGE OF INNOVATIVE, EVIDENCE- BASED APPROACHES BEYOND TRADITIONAL SIBS 2 Range of Outcomes-based funding approaches Social Finance, Inc. © 2017 Confidential WHEN IS OUTCOMES-BASED FUNDING USEFUL? 8 0 Redirect 1 Existing Funding Flows 2 Outcomes Rate Card 3 Prevention Fund 4 Social Impact Bond Outcomes- Based Contract Jurisdiction shifts existing funding streams towards outcomes-based solutions Possible to support upfront costs of new programs with outside funding (and then cover ongoing costs within existing budget) Payor contracts for outcomes, metrics, and pricing May be fully or partially contingent on outcomes Includes robust data, measurement, and engagement Payor pre- specifies outcomes, metrics, and pricing (fully contingent on outcomes) Providers apply jointly with funders Available to multiple providers; potential for broad systems change Philanthropic funders provide working capital with no repayment Instead, agree with public or private payor to reinvest savings into ongoing preventative intervention Private funders provide working capital at full cost of services Payor agrees to repay funders plus interest if outcomes are achieved; pays little if outcomes are not achieved Solutions have: Important challenges faced by communities Social Finance, Inc. © 2017 Confidential Defined Target Population Underserved, large-scale population, with adequate demand for intervention Measureable Impact Goals Clear set of outcomes that intervention seeks to improve for target population Intervention that Works Track record of evidence that intervention reliably demonstrates performance against impact goals Capable Service Provider Provider with capacity to deliver intervention at scale and with high quality Positive Value to Society Achieving intended impact brings clear societal benefit (social, community, economic) in reasonable timeframe Community Engagement Intended impact attracts support from local stakeholders and jurisdictions 9 Sample of “Intro to PFS” slides used in Kick-off (1st education project) Mahoning REMINDER: WHAT IS PAY FOR SUCCESS? Expansion capital ($) Private funders / impact investors Nonprofit intervention provider Repayment ($) Outcomes Payor (often government) Pay for Success is about measurably improving the lives of people most in need by driving government resources toward more effective programs Social Finance, Inc. © 2017 Confidential 2 WHEN IS PAY FOR SUCCESS USEFUL? Important challenges faced by communities Solutions have: Defined Target Population Underserved, large-scale population, with adequate demand for intervention Codified Program Model Well-codified program model with fidelity monitoring Rigorously Evaluated Evidence base studied versus rigorous comparison group Scalable Service Provider High-quality provider with capacity to scale Positive Value Clear link to public-sector benefits (economic and community benefits) within reasonable timeframe Payor and Investor Interest Outcomes attract civic and/or commercial support Social Finance, Inc. © 2017 Confidential 5 Shift to framing SIB model as “traditional view” suggesting alternatives, and shift to “service provider” to include districts (not just nonprofits) New page outlining alternative approaches (note: this was in development as part PII; we expanded and began actively including it in materials and messaging) Expanded relevance: Shift away from strict codification “Track record of evidence” vs. only RCT Capable service provider (given limited ability to “scale” for most districts) Shift from “payor” to “community” for alternative models 93 Appendix B. Summary of Takeaways from February CTE Project Convening February 8, 2018 OUR FEBRUARY CTE CONVENING ENABLED SITES TO ENGAGE WITH EACH OTHER ACROSS A WIDE RANGE OF TOPICS High Quality CTE Focus of Breakout Sessions PFS Readiness • Best practices in pathways • Operationalizing interventions assessed during feasibility • Cost-benefit analysis • Outcomes payors • Key aspects of evaluations • Identifying and integrating data Measurement & Evaluation Plenary Content • Cross-site share-outs: Project overview, status, and learnings • What comes after feasibility? Overview of outcomes-based financing options Social Finance, Inc. © 2018 Confidential 1 KEY TAKEAWAYS FROM THE CONVENING FOCUSED ON HIGH QUALITY CTE, PFS FEASIBILITY LEARNINGS, AND NEXT STEPS Key Takeaways: Implementing High Quality CTE Programs • Codification: Codified models are important, but don’t always meet the needs of the specific contexts; important to “use the right ingredients,” although it is fine to follow “different recipes” • Selecting career / industry focus areas: Focus on industries with labor market demand & progression possibilities or transferable skills Wondering: How do you balance student & labor market demand? • Credits and Credentials: Concern that field is prioritizing having credits / credentials vs. ensuring their relevance towards degrees or jobs • Role of employers: Critical to have employers at the table from the beginning; shared observation that employer engagement has been (slowly) shifting from CSR to HR/talent pipeline Questions re: Next Steps:  How do we think about moving forward? What is the typical timing and how do we leverage the work from feasibility?  Is an evaluator required to pursue outcomes-based financing? If so, how do we identify one?  What happens if we enter an outcomes- based financing situation and don’t reach the pre-determined goals?  What would it look like to have shared payor responsibility across district, foundation(s), and a more traditional outcomes payor? Is a blended approach possible?  What comes next? Key Takeaways: PFS Feasibility • Measurement and Evaluation: Core tension between long-term impact and measurability, particularly tracking post-HS outcomes; sites shared examples of beginning to address (e.g., higher ed partnership, National Clearinghouse) • Payors: Perception that pool of payors (particularly if employers are payors) is more likely than a single payor; open question around how district could act as payor and service provider • Outcomes-based Funding Model: Similar challenges across sites in assessing feasibility for “traditional PFS”; realization that all sites are exploring a range of outcomes-based approaches enabled a broadened mindset around how to use the tool Social Finance, Inc. © 2017 Confidential 2 94 AT THE END OF THE DAY, SITE TEAMS APPRECIATED THE SPACE FOR SHARED LEARNING AND FELT ENERGIZED ABOUT NEXT STEPS “We’re more similar than we’re not. This concept is new, but our work is the same.” “I can’t tell you how glad I am that I came here. We have a whole new perspective based on everything we’ve heard here.” “Our confidence level in pulling this off was 50/50 at one point but it’s a lot higher now.” “It is helpful to share about the common issues that we face.” “Having the chance to talk about different funding models was a big takeaway. It opened doors for us. Just hearing about the other projects is important.” “The analyses from this work will help us approach employers as we can better articulate our value-add.” “This work is messy, and that is ok!” Social Finance, Inc. © 2017 Confidential 3 95 Appendix C. PFS in Education Working Session: Key Takeaways February 23, 2018 RESPONSE: WHAT WAS OUR BIGGEST “VALUE ADD” TO OUR EDUCATION PARTNERS? Participants responded to this question via poll during session; responses are summarized in Word Cloud form (size of text correlates to frequency of response) Social Finance, Inc. © 2018 Confidential 1 RESPONSE: WHAT WAS THE MOST CHALLENGING PART OF FEASIBILITY ANALYSIS? Participants responded via poll during session (each participant had two “votes”) Feasibility Criteria Defined Target Population Measureable Impact Goals Intervention that Works Capable Service Provider Positive Value to Society (“ROI”) Number of “votes” 0 1 1 4 4 7 7 Community Engagement (Potential Payors or Funders) Outcomes-based Funding Structure with Strong Fit Social Finance, Inc. © 2018 Confidential 2 96 KEY TAKEAWAYS AROUND TOP TWO CHALLENGES • Benefit side is particularly challenging:  Muddled research on career pathways  Some strong evidence around career academies, Early College High School, and Linked Learning  However, studies are in particular contexts and no studies demonstrate impact across all post- secondary and career outcomes  Limited ability to attribute outcomes to specific programs  Zero sum game; students entering a pathway would otherwise enter a different pathway or program  Difficult to isolate the impact of a specific pathway; largely dependent on labor market demand  Outcomes accrued over long time horizon and dispersed across stakeholders  Different potential payors may focus on different outcomes goals  Largest fiscal benefits (more tax revenue, less social welfare expense) accrue at Federal level • Most education stakeholders (SEAs, districts) have limited flexibility in budget  In order to engage, must focus on shifting funds from current programs or expenditures  Requires more confidence around the new program being truly better than existing ones • Introducing new payors is exciting but challenging  Non-education stakeholders can bring new resources into a resource constrained environment  Long-term benefits of education align more with payors outside of the K-12 system  However, new payors are not paying for K-12 education; not within current mental model  Further, although we quantify benefits, value has not been compelling for any individual stakeholder • Opportunity: Start with payor, with clear stated outcomes goals  Is it possible to find payors who prioritize outcomes over ROI?  Hard to find payor who realizes the full value captured in our analysis, but mission-aligned payors may agree with the value and be willing to support • Across projects, employers emerged as high potential payors  Employers excited by fiscal benefits associated with hiring, recruiting, and retaining high quality talent 3 Social Finance, Inc. © 2018 Confidential Social ROI Outcomes Payors/ Funders ADDITIONAL REFLECTIONS ON SIB CHALLENGES IN EDUCATION PROJECT CONTEXTS  Size of individual programs is often too small to merit complexity and transaction costs associated with a SIB Might SIBs be better able to support larger, cross-site expansions?  Launching whole-school models requires significant upfront planning and implementation time, exacerbating outcomes timing issue; long time over which to amortize a relatively small investment  Exit strategy is a concern; if an intervention fundamentally changes the core of a school’s programming and does not work, difficult to shift back (e.g., can’t just “stop”)  Significant percentage of budgetary needs were for capital infrastructure; challenging because these capital expenditures do not have a natural “endpoint” for a SIB  Logistically, SIB financing is an add-on to per pupil expenses; how do you consider the impact of the “wrap-around” vs the “core”? Social Finance, Inc. © 2018 Confidential 4 97 LEARNINGS: CONDITIONS FOR SUCCESS IN EDUCATION PROJECTS If the intervention happens in a school setting, district must be on-board and proactive from the start Projects are more efficient and likely to succeed if partners approach Social Finance (and have “skin in the game”), vs. us approaching partners Need a resource at the project site who is specifically dedicated to overseeing the work, and has capacity to do so As much as possible, approach the project from the payor perspective; how can we work for the payors (vs. ask them to engage later) It is helpful to have a partner who knows the program, people, and context very well (and sits outside of the key project stakeholders) Note: Other than the first point, these conditions for success should apply across issue areas, not just in education contexts Social Finance, Inc. © 2018 Confidential 5 HYPOTHESES FOR THE FUTURE: WHAT INSIGHTS DO WE HAVE NOW ABOUT HOW WE MIGHT ENGAGE IN EDUCATION IN THE FUTURE? Value-adding Analyses (PII)  Even on projects that did not lead to financing, partners saw significant (and distinct) value from our work  Two aspects of feasibility analysis that were most valuable across projects:  Cost-benefit analysis (particularly, linking costs and benefits to particular programs)  Outcomes valuation / ROI analysis, and connection to stakeholders who accrue value  Moving forward, Social Finance may benefit from expanding into education through PII work, rather than focusing on deal development side  Note: Also opportunity to use education examples as case studies for PII efforts Segmented Approach  Focus on programs and interventions with direct link to college and career outcomes  Trend of employer engagement across CTE projects suggests high potential for more employer payor / funder models moving forward  Ability to make strong outcomes argument around college and career helps alleviate wrong pockets problem, as long-term stakeholders can more clearly see value  Potential to focus on CTE / high school and pre-k for “lower hanging fruit” (vs. elementary and middle school interventions); different segments require different approaches High Potential Partners  Key considerations for future district partners:  Size matters (larger districts can think more about scaling, and more opportunity to engage investors / community stakeholders)  PFS success requires key committed partner at district level; project should be embedded in district priorities with dedicated staff  Key considerations for future nonprofit partners:  Business model is critical; nonprofits without fee-for-service revenue model have less incentive to adjust services  Nonprofits operating outside of the core (e.g., afterschool, summer, wrap-around) may lead to less complex project design as financing sits outside of per pupil budget Social Finance, Inc. © 2018 Confidential 6 98 Appendix D. Summary of Responses from “Formal” Data Collection Efforts OVERVIEW OF FORMAL DATA COLLECTION METHODS Formal Data Collections Staff Included Duration Topics Covered Analysis Method  4 staff members (2 Directors, 1 Associate Director, 1 Associate)  45 minutes  Social Finance vision and strategy in education  Work in the sector to-date  Key challenges  Simple summary of responses and takeaways August 2017 Focus Group November 2017 Interviews February 2018 Interviews  13 staff members  45 minutes  (2 Directors, 5 Value to Social Finance of being in education space Value that Social Finance brings to education space Degree and drivers of urgency to be in the space Belief in ability of PFS to work in education Value to Social Finance of being in education space Value that Social Finance brings to education space Urgency to do more work in the space in the future Vision for future education work Other changes in perception of education  Thematic content analysis: Associate Directors, 6 Associates)  9 ultimately staffed on education projects  10 staff members  30 minutes (1 Director, 5 Associate Directors, 4 Associates)  All staffed on education projects          Coded for key phrases and themes Summarized each into a thematic “response” Counted number of interviews highlighting that response Note: Many interviews featured multiple themes in a response to a given question, so totals do not equal total number of interviews Identified trends at the interviewee-level in subsequent follow-on analysis 7 Social Finance, Inc. © 2018 Confidential FOCUS GROUP SUMMARY  Lots of potential impact, as education is such an important issue area  Many partners and funders that we care about are focused on education; working in the space might open more doors with them  Significant first mover advantage in the PFS world; no one has done education so we could get that advantage  Many staff members are passionate about education and motivated to pursue the work  Business Model is unclear:  Many conversations end because we can’t find funding to move forward  More engagement from service providers and philanthropies than government  Broad and complex system; work on Federal funding flows highlighted complexity; still trying to figure out where we fit in the system  Payors are complicated; schools / districts are both providers and payors  Value generating outcomes occur far from the direct intervention activities  Overall, functionally, the exact same challenges faced in every other issue area Social Finance, Inc. © 2018 Confidential “We don’t want to be issue area experts.” “The mean focus of the issue area team has been responding to RFPs.” “Beyond “execute well” I don’t think there is a broader strategy.” “The PFS field is still so small that being the first to do a project in a space is a huge advantage in doing more work in that area.” “We don’t want to start doing a bunch of feasibility work that is provider-driven and doesn’t go anywhere.” “All issue areas have challenges and we are always working to address them; education is no different. It should not require a fundamentally different approach.” 8 Focus Reflections from Group Goals / Strategy  No specific issue area goals or strategy, but none in any issue area  Part of the strategy is to be issue area agnostic  Informal goal to-date was to get any business in education, because we had not done work in the space  Social Finance has many conversations around education, but none have turned into actual projects  Now that there are projects, focus is on success on those projects Rationale for Working in Education Key Challenges 99 NOVEMBER INTERVIEWS SUMMARY (1/2) “Education affects all of the issue areas where we do work now. To not be involved in education would indicate that we aren’t serious about a lot of issues. How do you argue about hitting prevention points in other areas if you aren’t in education?” “Working in education gives us credibility and opens up new funders.” “Education is a tremendous cost, and makes up a huge set of programs, at the federal, state, and local levels. We are spending so much money, and not getting great outcomes. This is a huge opportunity for PFS.” “We are good at understanding how programs positively impact meaningful outcomes and translating those into value” “Our focus on ROI, measurement and evidence is all helpful.” “It is nothing new for people in education to care about outcomes; we come in with a discrete method towards an outcomes based product.” 9 Mission value  Business value  Limited value Response Underlying impact on other issue areas Large share of government resources Large share of foundation resources “Early mover” opportunity in new area Limited value vs. other issue areas # of interviewees Summary 5 of 13 highlighted only mission-driven rationale 3 highlighted only business-driven rationale 3 highlighted both in their response 2 were unsure of value 8 6 2 3 3 Response # of interviewees PII-related Limited value Summary 9 of 13 highlighted aspects of our work that may offer stand-alone value to education partners (in line with PII effort) 4 felt that Social Finance has limited or unclear value in the space Determine key 2 outcomes to track Bring an outcomes-lens to decision making Ability to quantify value of outcomes 8 Negotiate with complex stakeholder group 4 4 2 Some, but no 2 comparative advantage Unclear to-date; need more information What is the value to Social Finance of pursuing work in education? What value can Social Finance bring to the education sector? Social Finance, Inc. © 2018 Confidential NOVEMBER INTERVIEWS SUMMARY (2/2) “I sense an urgency for us to be opportunistic to prospective partners think that we can be of value, but we haven’t been overly deliberate.” “I sense an urgency around broader applications (e.g., infrastructure) but not around education specifically – it more around how we can broadly apply the work that we do” “I sort of feel like we have happened onto our ed projects opportunistically; DOE put out these RFPs and we responded.” “Before we had any education projects I felt an urgency; I think it made us look bad to have no experience in the space.” “I would be extremely surprised if my project leads to a transaction; there are too many reasons that it is not a good fit.” “I think we are bringing a lot of value as consultants, but definitely don’t think this is right for PFS” 10 Limited 8 8 of 13 highlighted opportunistic or strategic drivers (note: all 3 staff who highlighted strategic rationales also highlighted opportunistic ones) 5 of 13 felt no urgency or less than for other sectors Opportunistic  Strategic Response Opportunistic; based on winning grants Important for our reputation Important for our sustainability Don’t sense particular urgency or strategy Lower urgency vs. other issue areas # of interviewees 1 Summary How would you describe any urgency for Social Finance do work in education? If you had to guess now, what will be the likely outcome of your education project? 2 2 3 PII-related Limied value Response Unlikely to result in a PFS deal Expect to play more of a consultant role Expect to learn about PFS in education N/A (not on current education project) # of interviewees Summary All 9 interviewees currently staffed on education projects expected no PFS deal to result from the work, and/or for Social Finance to play a consultant role only 2 5 7 4 Social Finance, Inc. © 2018 Confidential 100 FEBRUARY INTERVIEWS SUMMARY (1/3) Response Important to fulfilling our mission Part of our broader learning agenda Unclear because of business model Unclear because of fit with mission/strategy # of interviewees Summary Only 2 of 9 highlighted only mission rationale 4 highlighted only business rationale 4 highlighted concerns with business model or mission fit Mission value  Business value  Unclear value 3 5 2 4 Response Putting a mirror on data in new ways Ability to quantify value of outcomes Having an outcomes focus Broad suite of consulting services Exploring creative funding options # of interviewees PII-related All 9 interviewees highlighted aspects of our offering that may offer stand- alone value, or general consulting services, to education partners (in line with PII effort) 1 also highlighted the exploration of funding options (more traditional Social Finance focus) Summary 3 3 4 1 5 What is the value to Social Finance of pursuing work in education? What value can Social Finance bring to the education sector? “Our education projects have been a great test of how new tools can be applicable to broader range of issues; they have forced us to really break some of our frameworks that we are used to using” “I don’t think we have a clear enough sense of whether there is funding for future education work to say whether there is actually value for Social Finance” “Getting districts to be even a little more outcomes based is better” “I undervalued this in the past, but there is real value in just shining a mirror on your own data about where you are today.” “Our biggest value add is orienting people around evidence and outcomes” 11 Social Finance, Inc. © 2018 Confidential FEBRUARY INTERVIEWS SUMMARY (2/3) “Now that I have done education projects, I wonder if focusing on other issue areas with more natural fit makes more sense.” “I think the past year has created a real need for us to figure out who we are at Social Finance; there is more urgency for answering that question than creating sector-level strategies” “If anything, I think it has become less urgent as we have seen how many challenges arise on education projects.’ “There is a lot of momentum now about disaggregating the tools; our education projects have been great test cases for PII” “We are not ready for a unified strategy; we would need at a minimum another six projects to land on a strategy that wouldn’t just get proven wrong” “I think that future education work will look more like PII; a key question though is where is the line between us and Bridgespan? Our work should have something to do with financing.” 12 How would you describe any urgency for Social Finance do work in education? What are your reflections on a future vision for Social Finance in education? Response I sense less urgency now than before No change in urgency Education has been a great sandbox for PII I sense more urgency around defining who we are then sector strategy # of interviewees 5 Summary No interviewees sensed an increase in urgency or excitement after the six projects 5 felt less urgency 4 highlighted need to define broader strategy before determining urgency for education work 2 3 4 Response # of interviewees Summary All 9 interviewees highlighted aspects of our offering that may offer stand- alone value, or general consulting services, to education partners (in line with PII effort) Should continue to be opportunistic Better fit within the PII portfolio Different models may work better than SIBs Need to take a narrower approach within 1 education 3 3 7 We still need to learn more Unsure that Social Finance is the right organization 4 8 Social Finance, Inc. © 2018 Confidential 101 FEBRUARY INTERVIEWS SUMMARY (3/3) Would you be excited to work on another education project? What, if anything, has changed about your perception of PFS in education? (Responses too varied for meaningful thematic content analysis) Do you think PFS makes sense in education? Response # of interviewees Response Not the traditional model Need more information # of interviewees 8 2 Yes No 3 No sure 2 5 “The cohort model gave us a new ability to draw insights across education projects.” “We have greatly sharpened our understanding of the education sector and the difficulties of PFS in education.” “I think there is more hesitance about education because we saw so many projects at the same time and there were so many challenges across all of them.” “I think these projects have really highlighted the need for a broader framing of PFS.’ “For the first time, we were intentional upfront about not just striving for a SIB; I think this gave us much more room to understand the context and make the recommendation fit.” “Honestly, I think we are still exploring, and still have a lot more exploring to do.” “The way we used to think about PFS was never going to work in education, and that is why we never did education; now we have had the opportunity to think about education with an open mindset and we were able to get more creative and expansive.” “Before, PII was just a concept. These education projects were an opportunity to test out some of the PII thinking and we have learned a lot about PII through them.” Social Finance, Inc. © 2018 Confidential 13 102