We would probably be better off as a society if the decision makers in the nation’s large private foundations took up surfing. Why? Because surfing is about letting go, and that’s what foundations must do to achieve higher impact. Surfing is incredibly humbling, an encounter with the enormous power, beauty, and unpredictability of the ocean. No surfer would attempt to change the shape of the waves or the schedule of the tides, because these forces are far beyond any one person’s control.
But two common practices of major foundations—the design of specific solutions to social problems and the narrow focus on one pathway to a goal—are the equivalent of ordering the ocean to change shape. Just as men cannot control oceans, individual foundations cannot control social systems. Such an approach underestimates the vast power and complexity of the systems in which foundations are attempting to intervene.
The strategic philanthropy movement has been a positive influence in recent years by encouraging foundations to clarify their goals and regularly evaluate their progress. But it has also fueled practices that undermine the nonprofit sector’s impact, rather than amplify it. Too often, funders insist on controlling the ways in which social problems are solved. This is a move in the wrong direction.
To make steady forward progress solving problems in dynamic environments of complexity and uncertainty, foundations must shift from centrally planned, narrowly focused grantmaking strategies to more decentralized, diversified strategies that are better able to catch the waves of effective leadership, distributed wisdom, and innovation. There are two ways foundations need to let go. The first is to enable effective nonprofits to take the lead in designing solutions to social problems. The second is to diversify investments across multiple solutions or pathways to the goal. Let’s take a closer look at the problems with current practice in philanthropy.
Problem #1: Foundation-Designed Solutions
When solutions are centrally planned by people who are distanced from the real work in the field, the solutions are often poorly implemented. This is a classic principal-agent problem. The organizations tasked with implementation feel little ownership or passion for projects they didn’t dream up themselves. For example, in 2004 the William and Flora Hewlett Foundation made a grant to create the Legislative Education Project, as part of its California Education grant portfolio.1 The project was a response to term limits and the loss of institutional memory in the state legislature. It was intended to provide a nonpartisan professional development forum for legislative staff to learn about the history and current status of California education policy. The foundation developed the idea and then invited a respected university research center to implement it. Unfortunately, the researchers weren’t able to keep the legislative staff engaged and satisfied with the programming, nor did they respond aggressively enough to complaints from participants about an imbalance between liberal and conservative viewpoints. Eventually, the Republican legislative staff refused to participate, and those who did participate gave only lukewarm reviews of the sessions. The foundation considered the project a failure and did not renew the grant.
Later on, some consultants who were marginally involved in the first effort approached the foundation with a different vision. They knew many legislative staff well and had thought carefully about how to modify the project. In 2006, Hewlett made a grant for this new version, which included the following changes: The consultants created a steering committee of legislative staff to guide the project and decide on session topics and format; they followed the interests of the legislative staff and focused the sessions on visiting school districts and schools, rather than on presentations from researchers; and they acted as conversation facilitators rather than presenters. This time, the project was a success. Over the past five years there has been strong bipartisan participation, with legislative staff reporting that the experience has improved their basic knowledge and helped thaw the partisan divide as well as identify areas of common interest for education policymaking.
The point is not that foundations need to do a better job picking nonprofit implementers, but that for the best results the implementers need to pick and design the solutions themselves. The key difference between these two versions of the Legislative Education Project was who owned the solution.
There is another way foundation-designed solutions can undermine effective implementation. When each foundation develops its own unique strategy for solving social problems, it becomes difficult for a grantee organization to have a coherent vision and strategy of its own. (Public charities with 501(c)(3) tax status must raise funds from multiple sources, as they cannot legally receive all of their funding from one foundation.) Implementation efforts suffer greatly when the nonprofit groups doing the work are pulled in 10 different directions, carrying out 10 different foundation strategies, to get adequate funds.
A similar phenomenon has developed with funding for public school districts nationwide. State and federal politicians, who control education funding, want to direct how the funding is used. So they create “categorical programs,” which allocate funding to school districts for specific purposes, such as textbooks, librarians, after-school programs, and teacher training. All these activities are important, but having hundreds of small, narrowly defined funding programs that support different parts of a school’s activities creates enormous paperwork and inefficiencies for school leaders and leaves them with very little flexibility to innovate and improve their student services.
The same holds true for foundation funding of nonprofits. The more that foundations dictate to grantees how they should solve social problems, the more they constrain the grantees’ leadership, expertise, and ability to innovate—and the more bureaucratic work they create for them.
Problem #2: Tunnel Vision
To avoid spreading funding too thinly, many foundations choose to invest in only one solution or pathway to their goal. Instead of letting 1,000 flowers bloom, they think they can afford just one variant. But focusing narrowly on one solution is a fragile strategy, particularly in complex, unpredictable environments.
For example, in 2001 the Hewlett Foundation launched a grant initiative focused on increasing the achievement of California’s K-12 students. The foundation chose state policy as the best pathway to the goal because it could help ensure that the state government’s many billions in education funding were spent more effectively. This strategy looked promising for a few years. In 2005, Gov. Arnold Schwarzenegger created a Committee on Education Excellence to develop reform recommendations. The committee, with other elected officials, requested research to determine how to make the education finance and governance systems more effective. In 2007, the research was completed. The governor declared 2008 “the year of education,” and he and a host of elected officials stood together with the researchers and pledged to work together to improve California education. The media coverage was extensive. It seemed as though the state was on the brink of a major overhaul of its school finance and governance systems.
The Hewlett Foundation chose a powerful leverage point in state policy change, but it made a mistake by putting all its eggs in one basket. When the state budget crisis began, the foundation was left without other pathways to achieve impact with this portfolio. There have been improvements in state education policy since 2008 and important progress has been made in building the state’s education data systems, but the school finance reform that once seemed possible has not been achieved.
The Hewlett Foundation is not alone in narrowly focusing on one solution or pathway to achieve its goals in a grant portfolio. A few other examples include the Bill & Melinda Gates Foundation’s high school grants initiative (2001-2005), which focused on creating small high schools as the pathway to higher student achievement;2 and the Irvine Foundation’s coral initiative (1999-2007), which invested in afterschool programs as the pathway to higher student achievement.3 None of these strategies generated the results that were hoped for, even after investments of years and hundreds of millions of dollars.
Neither extreme—letting 1,000 flowers bloom or putting all eggs in one basket—is likely to produce high impact. As private sector investors learned long ago, the sweet spot is in the middle: investing in a diversified portfolio that includes a range of possible solutions.
Besides undermining implementation and producing disappointing results, there are additional downsides to foundations trying to control exactly how social problems are solved.
Exerting too much control causes foundations to ignore highly effective programs and organizations. If a foundation’s strategy focuses on one specific way of solving a problem, then all other solutions do not fit that strategy and cannot qualify for funding, even if they produce great results. In this way, foundations are choosing control over impact. For example, there is extensive research showing that participation in a high-quality preschool program makes a huge difference in a child’s readiness to learn in kindergarten and in academic achievement later on. There is similarly strong literature on summer learning loss and the impact of extended learning time on the achievement of low-income students. But few of the nation’s largest education funders are making grants to promote these two well-proven solutions. Similarly, many nonprofit organizations with proven results, such as America’s Choice, Aspire Public Schools, Citizen Schools, KIPP Schools, the New Teacher Center, the New Teacher Project, the Success for All Foundation, and Teach for America, get passed over for funding by large private foundations. In fact, the 19 nonprofits that ranked highest in the U.S. Department of Education’s Investing in Innovation (i3) competition, which was based on evidence of impact, on average had grants in 2010 from only three of the nation’s top 50 education foundations before winning the i3.
Exerting too much control causes foundations to ignore innovation. During the time that the Hewlett Foundation was investing in the California education strategy with a focus exclusively on state policy, several innovative new approaches were developed that had the potential to dramatically increase student achievement. The Long Beach Unified School District’s MAP2D math program produced dramatic achievement gains for elementary students throughout the district, and now four other California districts are implementing the program and generating similar results. Long Beach and Fresno school districts formed a partnership to assist each other in achieving specific student performance goals, providing a much-needed alternative to the external assistance model used unsuccessfully by the state for years to turn around low-performing schools. More than 60 of the nation’s largest school districts created a data-sharing network called the Key Performance Indicators Project, managed by the Council of the Great City Schools, to benchmark administrative and academic costs and performance, identify high and low performers, and share best practices. Rocketship Charter Schools in San Jose created a hybrid schooling model that uses fewer teachers and more time with computers than traditional schools, generating impressive achievement gains for low-income students. And Roadtrip Nation, which helps high school students understand the real-life relevance of education, expanded its program into California. Unfortunately, none of these innovations aligned with the foundation’s California education strategy.
Perhaps the single highest impact grant that the Hewlett Foundation’s Education Program made in the past 10 years was to help launch the Massachusetts Institute of Technology’s OpenCourseWare initiative. The objective was to make all of the university’s courses, lecture notes, and course materials freely available on the Internet. MIT’s pioneering effort helped spark what has become a global open education resources movement, enabling perhaps the most significant worldwide expansion of access to higher learning in decades. Today, more than 250 universities around the world have followed MIT’s lead, making the materials for more than 13,000 college courses in 20 languages openly available online. This investment proposed by MIT President Charles Vest—and the $120 million grant portfolio that followed—might not have been possible if the foundation already had decided on a specific strategy for how to use technology to improve education.
Lessons from the Cutting Edge
Many cutting-edge organizations today have achieved high performance by using distributed decision making and rapid prototyping to test solutions. In the business sector, high-performing organizations are empowering employees to make decisions, design new solutions, and continuously reassess and improve systems and products. The success of the Toyota manufacturing process, which relies on decision making and problem solving by production line workers, is a classic example. More recently, Google has provided its staff with flexible time to work on new projects of their own design.
Some companies have gone beyond empowering their employees to engaging their customers in co-creating products through open innovation and crowdsourcing. Apple allows users to write applications for its iPhone, and Facebook does the same for its social networking website. Netflix recently held a public contest to improve its movie recommendation algorithm. The Linux computer operating system and the Firefox Internet browser are both open-source products that thousands of programmers helped create, and they are considered by many to be better functioning than comparable products from traditional companies.
In the nonprofit sector, Wikipedia uses a small staff and a massive global corps of volunteers to create an online encyclopedia with far more information than traditional encyclopedias that are produced by small teams of experts. Alcoholics Anonymous, a successful international movement with more than 2 million members, uses an extremely decentralized organizational model in which each local AA group is a self-governing entity. MoveOn.org, a nonprofit public policy advocacy group, asks its 5 million members to vote on its agenda and priorities.
In the government sector, market-based economies tend to be better than centrally planned economies at efficiently providing goods and services that people want. Leading thinkers in international development are suggesting a move away from centralized planning for development by aid agencies to a decentralized approach. 4 The Long Beach Unified School District, one of the most consistently high-performing, high-poverty school districts in the nation, has achieved its success by encouraging and nurturing innovations in local schools, evaluating carefully, and replicating models that work broadly throughout the district. In health care, some effective efforts at stopping the spread of dangerous antibioticresistant MRSA infections in hospitals have engaged nursing staff and other lower-level workers to help design solutions.
Then there is systems thinking theory, which suggests that effective changes in complex systems cannot be dictated by actors in any one part of the system—that lasting changes require many diverse actors and points of view to help produce solutions. Effective system changes also require flexibility and openness to watching how the system responds to an intervention and readjusting the approach. Donella Meadows, systems thinker and author of the best-selling book The Limits to Growth, explains: “Self-organizing, nonlinear, feedback systems are inherently unpredictable. They are not controllable. They are understandable only in the most general way. The goal of foreseeing the future exactly and preparing for it perfectly is unrealizable.” 5 Meadows concludes that dancing with systems, rather than trying to control them, is the best approach.
Last, proponents of design thinking advocate that the most successful solutions come from a deep understanding of the needs of end users. Rather than focusing too early on any one solution, the user-centered design process tests many different solutions quickly, eventually landing upon a solution after many rounds of user feedback and iteration. This process has produced highly innovative products as various as Apple’s iPod and Intuit’s Quicken financial software. Arnold Wasserman, industrial designer and design thinking pioneer, says: “Foundations are about 25 years behind the private sector. Cutting-edge businesses have long since recognized that powerful solutions come from continuous, rigorous research into their employees’ and customers’ needs and wants, from co-design with stakeholders, and from progressive iteration and prototyping toward ever better solutions.”
Why Is Philanthropy Stuck on Control?
Although reliance on top-down control is not exclusive to philanthropy, many attributes of the sector make it susceptible to this mode of operation. The feedback loops in business and government that create pressure for organizations to improve are missing in the philanthropy sector. Foundations are not set up to be accountable to shareholders or constituents. The private sector has standardized public information flows about profits and losses, but foundations don’t have good ways of comparing outcomes across different kinds of investments. There is little transparency in the philanthropy sector, so when funders do collect good data, it is hard for others to learn from it or to recognize trends. In addition, the power imbalance in the grantor-grantee relationship makes it hard for grantees to challenge funders’ plans and breeds a belief among funders that they know best. All these factors insulate foundations from honest feedback on their investment strategies, making it easier for them to maintain the belief that they can design the best solutions from on high.
We aren’t suggesting that foundations give up all control. In other sectors, the leaders of high-performing organizations typically set clear goals, create accountability mechanisms, and provide constant information flows to drive performance. They also provide the flexibility and support to allow those who are working in the trenches to experiment, innovate, and continuously improve. The leaders of these organizations have figured out how to be tight on goals and loose on means. If foundations want better results, they also should adopt a tight-loose approach.
Stronger Grantmaking Approaches
Several forward-looking foundations have chosen strategies that relinquish control over solutions to social problems. They are tight on goals and loose on means. Below are three promising approaches.
General support for effective organizations and leaders | Some foundations are focused on providing general support to nonprofits and individuals with proven track records. General support funding promotes effective implementation by supporting grantees’ own strategies and allows them to invest in their organizational infrastructure and capacity. This approach is naturally diversified, because each grantee may have a different way of achieving a foundation’s ultimate goal.
The Edna McConnell Clark Foundation (EMCF) is a good example of this approach. The foundation’s goal is to help economically disadvantaged young people become independent, productive adults, as measured by outcomes in education, employment, and reduction of risky behaviors. The foundation’s approach is to identify high-performing youth service organizations with effective programs and growth potential, and to support the entire organization rather than a few projects. Using this model, EMCF supported the scale-up of the Nurse-Family Partnership, a home visitation program that improves the health and development of mothers and children in low-income families. EMCF went one step further in rationalizing this funding approach. In 2008, the foundation raised $81 million in growth capital for three of its grantees (in addition to its own investment of $39 million). The funders agreed to use the grantees’ business plans as the basis for their grants and agreed to use the same terms, conditions, reporting requirements, and performance metrics for the grants. Nancy Roob, EMCF’s president and CEO, described the foundation’s approach: “We are committed to doing everything we can to get coordinated co-investment right, and to do so in a way that frees grantees to focus more sharply on execution, helps funders realize larger and more rapid social returns on their investments, and benefits more of America’s youth.” 6
Other foundations focusing primarily on general support for organizations and leaders include Ashoka, the Draper Richards Foundation, the Mulago Foundation, New Profit Inc., Sea Change Management, the Skoll Foundation, and the Sobrato Family Foundation. In addition, the Boston Foundation, the California Wellness Foundation, and the Hewlett Foundation have made substantial investments in general support, although it is not their sole focus.
A related investment strategy is to fund leadership development and professional networks. A few foundations using this approach include the Brainerd Foundation and the Robert & Patricia Switzer Foundation. These foundations provide flexible funding to help leaders further develop their skills, capacities, and professional networks to create high-impact solutions.
Community-designed strategies | Another promising investment approach is to rely on community-designed strategies, which do a better job than foundation-designed strategies of harnessing distributed wisdom for solving tough, systemic problems. These foundations fund strategies that are developed collectively by nonprofits and other stakeholders in the field.
For example, in 2007 the American Society for the Prevention of Cruelty to Animals launched ASPCA Partnerships, a grantmaking initiative with the goal of increasing the live release rate of animals from shelters to 75 percent. ASPCA picked 10 communities and asked stakeholders in the animal shelter system to collectively design a set of strategies to achieve the goal. In Austin, Texas, the partnership’s pilot community, this initiative boosted the live release rate from 45 percent to 69 percent in four years; in Spokane, Wash., the rate increased from 50 percent to 64 percent; and the other communities have made substantial improvements as well.
Other examples of foundation-supported, community-designed strategies include the Community Clinics Initiative funded by the California Endowment to attain health equity for underserved communities in California; the Positive Deviance Initiative at Tufts University, funded by the Rockefeller Foundation, which has used community-designed strategies to address health, malnutrition, and other social problems in developing countries and in the United States; and the New Mainstream strategy funded by the Columbia, Heller, and Kellogg foundations and others to make California a national leader in developing a sustainable food system.
Fostering innovation | Other foundations are focusing on innovation to achieve high impact. This approach holds the promise of making giant leaps forward in solving social problems, but it also requires funders to be comfortable with failure.
The Robert Wood Johnson Foundation’s Pioneer Portfolio, for example, was launched in 2003 to support organizations using innovative techniques to solve health problems. The foundation funds projects in stages, so it can assess progress and better understand which projects merit longer-term funding. Some examples of the foundation’s pioneer ideas include building the world’s largest repository of genetic, environmental, and health data; the Games for Health Project, which develops digital interactive games to teach health concepts and motivate healthy behavior; and stock markets that can predict the next influenza hot spot, to avert the next global pandemic. Although some projects may fail, Paul Tarini, director of the program, says: “That’s okay, actually. We needed a place where the foundation could explore. We don’t go into projects thinking each project has to meet with instant success; we go into them understanding their risk profile.”
Other examples of this approach include the Knight Foundation’s Media Innovation Initiative; the MacArthur Foundation’s Digital Media & Learning initiative; and the Rockefeller Foundation’s Advancing Innovation Processes to Solve Social Problems initiative. Judith Rodin, president of the Rockefeller Foundation, recently said, “At Rockefeller, we’ve learned that innovation techniques like crowdsourcing, design thinking, and user-driven innovation can tap new sources of knowledge … engage thinkers and doers from around the world in solving problems together … and then scale new ideas and best practices from one village to five, then to 500, then around the world.” 7
Another related approach is using incentive prizes and competitions to spur performance and groundbreaking innovations, such as the X Prize Foundation’s competitions for technological innovations, the Broad Foundation’s Prize for Urban Education, and the U.S. Department of Education’s Race to the Top competition for statewide education reforms.
Of course, as funders shift away from more prescriptive approaches and toward general support, field-developed strategies, and innovation, it becomes tougher to predict what kind of solutions will be generated by their investments. But as Alberto Ibargüen, president and CEO of the Knight Foundation, recently said: “It is incredibly liberating to admit you don’t know the answer. Then you don’t have to go out and pretend and say, ‘I am the foundation, I have an idea, and I have the money.’ Instead you can afford to say, ‘I have some money, here’s the problem we’re worried about, do you guys have any ideas?’” 8
The fundamental question is whether foundations are ready to relinquish some control to increase their impact. We think this change is long overdue.
Kristi Kimball has served as a program officer for the William and Flora Hewlett Foundation’s Education Program since 2004, where she has been responsible for more than $75 million in investments aimed at improving student achievement in California, increasing arts education, and reforming political overnance in California.
Malka Kopell has worked for the last 30 years to develop collaborative relationships between government and the people it serves. She has served as a program officer for the William and Flora Hewlett Foundation in the Conflict Resolution and Special Projects programs. Kopell founded Community focus, a California-based civic engagement organization, and was the founding managing director of the Center on Philanthropy and Civil Society at Stanford University.