| Notes |
SUMMER 2012 VOLUME 24 NUMBER 2
Federal Reserve Bank of San Francisco
Community Development
and Education
Plus:
CDFI Industry Analysis
The Affordable Multifamily Mortgage Industry

CI Notebook Aby Laura Choi
s summer draws to a close, kids across America are preparing for the inevitable: the start of a new school year. Whether they greet this season with dread or excitement, the fact remains that their educational experience will shape the course of their lives. Having
the means to access and absorb high quality K-12 educational resources lays the groundwork for postsecondary success and ultimately higher paying jobs. The converse effectively closes these doors.
While school and teacher quality are paramount for educational achievement, there is growing recognition that academic success depends heavily on meeting the needs of the “whole child.” These include proper nutrition, stable housing, adequate health services, a safe neighborhood, and positive adult role models. Such issues often present challenges in low- and moderate-income communities and addressing them is part of the daily work of community development. As such, while the community development field may not have a direct role to play in the classroom, there are reasons why the field should be considerably more attuned to the relationship between its work in low- and moderate- income areas and the educational outcomes for children growing up there.
This issue of Community Investments focuses on the intersection between education and community development in an attempt to identify shared goals and seed a conversation between the two sectors. The articles in this issue examine broad trends in educational equity and new models for better integrating community development and schools. Jeff Edmondson of the Strive Network and Nancy Zimpher of the State University of New York discuss the importance of setting standards for collective impact and getting a better social return on investment in education. Diana Hall describes how a thriving network of community schools in Multnomah County, Oregon is strategically aligning youth, family, and community services with schools to improve educational outcomes. Sean Reardon of Stanford University provides evidence of the widening achievement gap between the rich and the poor, which has important implications for inequality in America.
Our Eye on Community Development section includes a summary of new findings from a detailed analysis of community development financial institutions (CDFIs) on issues of capitalization, liquidity and portfolio, and risk management from 2005 to 2010. In addition, the California Community Reinvestment Corporation, a CDFI with a 23 year history of offering affordable multifamily mortgages in California, reflects on the lessons learned in adapting to the changing realities of the industry.
We hope this issue of CI encourages you to think critically (and optimistically!) about the opportunities for the community development field to partner with schools and improve educational outcomes for youth. We’d love to hear your thoughts on the subject and always welcome your feedback.
Enjoy what’s left of the summer!
Laura Choi
Community Development Department Federal Reserve Bank of San Francisco 101 Market Street, Mail Stop 215 San Francisco, CA 94105 www.frbsf.org
(415) 974-2765 / fax: (415) 393-1920
Joy Hoffmann
Group Vice President Public Information and Community Development joy.k.hoffmann@sf.frb.org
Scott Turner
Vice President, Community Development and Economic Education scott.turner@sf.frb.org
Laurel Gourd
Conference and Administrative Coordinator laurel.gourd@sf.frb.org
Esther Fishman
Administrative Specialist esther.fishman@sf.frb.org
RESEARCH STAFF
David Erickson
Manager, Center for Community Development Investments david.erickson@sf.frb.org
Ian Galloway
Senior Investment Associate ian.galloway@sf.frb.org
Naomi Cytron
Senior Research Associate naomi.cytron@sf.frb.org
Laura Choi
Senior Research Associate laura.choi@sf.frb.org
FIELD STAFF
Jan Bontrager
Regional Manager Arizona, Nevada, Utah jan.bontrager@sf.frb.org
Melody Winter Nava
Regional Manager Southern California melody.nava@sf.frb.org
Craig Nolte
Regional Manager
Alaska, Hawaii, Idaho, Oregon, Washington craig.nolte@sf.frb.org
Lena Robinson
Regional Manager Northern California lena.robinson@sf.frb.org
Darryl Rutherford
Regional Manager
San Joaquin Valley darryl.rutherford@sf.frb.org
This publication is produced by the Community Development Department of the Federal Reserve Bank of San Francisco. The magazine serves as
a forum to discuss issues relevant to community development in the Federal Reserve’s 12th District, and to highlight innovative programs and ideas that have the potential to improve the communities in which we work.
In this Issue
Special Focus: Community Development and Education
Community Development and Education: A Shared Future ......................................................................... 4 By Laura Choi, Federal Reserve Bank of San Francisco
An examination of the intersection between education and community development
and how cross-sectoral efforts can improve achievement among low-income students.
The New Civic Infrastructure: The ‘How To’ of Collective Impact and Getting a
Better Social Return on Investment.............................................................................................................. 10 By Jeff Edmondson, Strive Network and Nancy L. Zimpher, State University of New York
Cross-sector strategies require a common framework and set of standards for achieving
maximum impact, to avoid a “spray and pray” approach to improving educational outcomes.
Schools Uniting Neighborhoods: Community Schools Anchoring Local Change .......................................... 14 By Diana Hall, Multnomah County, Oregon
Learn how Multnomah County, Oregon is reinventing the school as a place that strengthens
the entire community and addresses the full spectrum of family needs.
The Widening Academic Achievement Gap between the Rich and the Poor ................................................ 19 By Sean Reardon, Stanford University
As the income gap between high- and low-income families has widened, has the achievement
gap between children in high- and low-income families also widened? The answer, in brief, is yes.
Looking Back and Moving Forward: Changes in the Affordable Multifamily Mortgage Industry .................. 25 By Mary Kaiser, California Community Reinvestment Corporation and George Vine, Vine Associates
A look back at some of the lessons learned from CCRC’s 23 year history and new lessons for the
affordable multifamily mortgage industry today.
CDFI Industry Analysis: Summary Report .................................................................................................... 29 By Michael Swack, Jack Northrup and Eric Hangen, The Carsey Institute and the CDFI Fund
The Carsey Institute and the CDFI Fund conducted a detailed analysis of a large sample of CDFIs
on issues of capitalization, liquidity and portfolio, and risk management by CDFIs from 2005 to 2010.
Research Briefs ............................................................................................................................................. 34 Dr. CRA........................................................................................................................................................ 36 Data Snapshot: Education............................................................................................................................. 37
Eye on Community Development
Quarterly Features
Federal Reserve Bank of San Francisco

Community Development and Educatio
A Shared Future
By Laura Choi
4
Community Investments, Summer 2012 – Volume 24, Number 2
n

:
IWntroduction
hat is the common link between higher wages, lower un- employment, reduced incarceration and crime, longer life expectancy and better health, and increased civic engagement? The theme of this issue of CI gives the
answer away: increased educational attainment is tied to each of these positive outcomes.1 Active recognition of this linkage is central to making headway on community development goals, as low-income children tend to have worse educational outcomes than their higher-income peers, a challenge that shadows low-income children throughout their lives. Of course, a complex set of individual and neighborhood factors influence educational outcomes, including parental education, school quality, so- cioeconomic status, peer effects, health, and neighborhood conditions. But what is interesting about this set of factors is that some of them lie squarely within the domain of community development. Yet, despite the crossovers between education and community development outcomes, the two sectors have historically operated independently of one another. Generally speaking, educators focus on in-school factors while commu- nity developers focus on neighborhood factors—a somewhat false di- chotomy, given the critical role that schools play in neighborhoods.
This distinction between in-school and out-of-school factors has led to a growing divide within the education sphere. Over a decade ago, educa- tion reformers gathered under the slogan “No Excuses,” as an indication of their refusal to accept poverty as an excuse for low achievement. Wanting to take immediate action where they could, they prioritized school-related changes, such as teacher quality and accountability, charter schools, and smaller class sizes. In contrast is the movement known as the “Broader, Bolder Approach,” which emphasizes the importance of non-K-12 school factors, such as early childhood education, health, social development,
Figure 1. Education Pays
Unemployment rate
Median weekly earnings
16 14 12 10
8 6 4 2 0
$1,800 $1,600 $1,400 $1,200 $1,000 $800 $600 $400 $200 $0
No High Some Assoc. Diploma School college Degree
Prof. Degree
Ph.D
BA
Highest Level of Education Completed
MA
Community Investments, Summer 2012 – Volume 24, Number 2
5
Source: Bureau of Labor Statistics, Current Population Survey
Special Focus: Community Development and Education
n
Unemployment Rate (2011)
Median Weekly Earnings in 2011
and poverty, in improving academic achievement among low-income students. Despite their shared end goals, the two camps are often in conflict over how to best achieve them. In a recent speech, Secretary Arne Duncan of the U.S. Department of Education called for an end to what he calls “the wrong education battles,” which includes the debate over the impact of in-school influences, like teachers and principals, on student achievement, versus the impact of out-of-school influences, like poverty and poor health.2 “Well-intentioned advocates on both sides present policy choices as an either-or choice—not as a ‘both-and’ compromise, however imperfect, that needs to be ironed out... In the wrong education battles, tough- minded collaboration gets dismissed as weakness, not as a way to work out a breakthrough win for children,” said Duncan.
The issue of public education in America is notoriously thorny, complicated by matters of politics, public funding, labor unions, and accountability—it’s enough to scare off any well-intentioned community developer. And while community developers are not educators, and would be wise to leave the reform of pedagogy and instruction to the experts, the community development field has a central role to play in working to improve the educational out- comes of low-income youth. While there are certainly ex- amples of the “tough-minded collaboration” that Duncan encourages, such as the efforts by CDFIs to finance charter school facilities serving low- and moderate-income (LMI) students,3 the field has struggled to more systematically
apply its comparative advantage in working to reduce poverty and improve household and neighborhood sta- bility toward educational ends. This article examines the intersection of education and community develop- ment, with a particular focus on recent efforts to improve achievement among low-income students through better collaboration across sectors.
Inequities in Education
The widely cited Coleman Report published in 1966 demonstrated the significant influence of socioeconomic factors (such as the economic status of a student’s peers, family, and neighborhood) on student achievement.4 Almost fifty years later, a wide body of research confirms that those findings still hold true. Data from the National Center for Education Statistics show that in 2011, schools with higher proportions of students eligible for free or reduced lunch (FRL), a proxy for student poverty, had lower average test scores in both reading and math at the 4th and 8th grade levels (see Fig. 2).5 Similarly, socioeco- nomic disadvantage shows an inverse relationship to high school graduation rates; as student FRL eligibility increas- es at the district level, graduation rates decrease.6 New research by Sean Reardon of Stanford University shows that the achievement gap between affluent and low-in- come students grew by about 40 percent since the 1960s, and is now double the black-white achievement gap (see the article “The Widening Academic Achievement Gap between the Rich and the Poor” in this issue).
Figure 2. High Poverty Schools Have Lower Average Test Scores
Average Reading Scores by School Poverty, 2011
Average Math Scores by School Poverty, 2011
Average Reading Scores by School Poverty, 2011
Average Math Scores by School Poverty, 2011
4th Grade 8th Grade 4th Grade 8th Grade
4th Grade 8th Grade
4th Grade 8th Grade
300
250
200 150 100
50 0
279
238
268
226
258
217
247
203
350 300 250 200 150 100
50 0
0–25 percent
26–50 percent
51–75 percent
76–100 percent
Percentage of students in school eligible for free or reduced price lunch
300
255
287
245
276
237
264
226
0–25 percent
26–50 percent
51–75 percent
76–100 percent
Percentage of students in school eligible for free or reduced price lunch
6
Community Investments, Summer 2012 – Volume 24, Number 2
Source: National Center for Education Statistics, U.S. Department of Education.
Average Reading Scale Score
Average Math Scale Score
This achievement gap is closely tied to household level factors. Some of the strongest predictors of educa- tional outcomes for youth include parental education and household income, yet even very specific household traits, such as the number of books in the house or pa- rental vocabulary levels, can impact a child’s educational trajectory.7 But children from poor households tend to live in poor neighborhoods, and thus face not only their own household disadvantage, but also a number of neigh- borhood-level characteristics that are correlated with educational outcomes. Beyond the obvious issue of dis- crepancies in school quality,8 children from low-income neighborhoods often contend with more local crime and violence, greater housing instability, fewer community re- sources such as libraries and after school programs, and weaker social networks, particularly with respect to adult role models. On the whole, the economic composition of their neighborhoods matters for students; research has shown that having high-income neighbors has a positive effect on school readiness and achievement outcomes for youth, even after accounting for individual and family characteristics.9
It is troubling, then, that residential patterns over the past forty years indicate that neighborhoods have become increasingly segregated by income, suggesting that low- income youth have fewer opportunities to interact with middle- and higher-income peers and adults. A study from the US 2010 project at Brown University found that in 1970, 15 percent of families were in neighborhoods classified as either affluent or poor; by 2007, this share doubled to 31 percent, reflecting the growth in neighbor- hood income concentration at both ends of the income spectrum (with fewer families living in middle-income neighborhoods).10 As the authors note, “Income segrega- tion is particularly salient for children because it leads to disparities in social context and access to public goods that are particularly relevant for children, such as edu- cational opportunities and school quality.”11 This point is reinforced by recent research from the RAND Corpo- ration, which takes advantage of Montgomery County’s large inclusionary zoning policy and its scattered site public housing program to study the effects of increased economic integration on educational performance. The study revealed that over a period of five to seven years, children in public housing who attended the schools in affluent areas of the district far outperformed in math and reading those children in public housing who attended the elementary schools in higher poverty areas.12 Inclu- sionary zoning policies, which mandate that a given share of new construction be affordable for low-income house- holds and thus help foster mixed-income neighborhoods, are a boon to the select low-income children that get the opportunity to attend economically integrated schools,
but how can the community development field address the persistent challenges facing schools with high concen- trations of students in poverty?
Improving Partnerships between Communi- ties and Schools
Numerous strategies have emerged in recent years to comprehensively address the unique educational, health, and social development needs of low-income children. Perhaps the best known is the highly publicized Harlem Children’s Zone (HCZ), which takes a holistic approach to educating low-income students by integrating a high- performing charter school with after-school, parental edu- cation, social-service, health and community-building programs.13 In 2010, the Department of Education (ED) launched the Promise Neighborhoods program, modeled after HCZ, which provides funding for the planning and implementation of a “complete continuum of cradle- to-career solutions of both educational programs and family and community supports, with great schools at the center.”14 Similarly, the Department of Housing and Urban Development (HUD) launched the Choice Neigh- borhoods program in 2010, which aims to transform dis- tressed neighborhoods and public and assisted projects into viable and sustainable mixed-income neighborhoods by linking housing improvements with appropriate ser- vices, schools, public assets, transportation, and access to jobs.15 A strong emphasis is placed on local community planning for access to high-quality educational opportuni- ties, including early childhood education. A key compo- nent of these federal programs is the required alignment across participating agencies, which include ED, HUD, and the Department of Justice, demonstrating the im- portance of taking an integrated approach to improving neighborhoods and schools.
In addition to these efforts at the federal level, locally driven initiatives to support broader partnerships between public schools and their surrounding communities have also emerged. For example, six years ago in Cincinnati, more than 300 leaders from the education, nonprofit, community, civic, and philanthropic sectors came to- gether to form the Strive Partnership, under the common goal of improving the educational success and career readiness of children from the region. The Strive Partner- ship aligns the efforts of these multiple entities through the infrastructure of a common set of core metrics tied to the Partnership’s shared goals, such as kindergarten readiness, high school graduation, and postsecondary retention and completion (see the article “New Civic Infrastructure: The ‘How To’ of Collective Impact” in this issue of CI for more information). The improvement in outcomes catalyzed and supported by Strive’s approach has been held up as a successful example of collective impact, defined by John
Community Investments, Summer 2012 – Volume 24, Number 2
7
Special Focus: Community Development and Education
Kania and Mark Kramer of the Foundation Strategy Group as, “The commitment of a group of important actors from different sectors to a common agenda for solving a spe- cific social problem.”16 Kania and Kramer are careful to point out that, “collective impact is not merely a matter of encouraging more collaboration or public-private partner- ships. It requires a systemic approach to social impact that focuses on the relationships between organizations and the progress toward shared objectives.”17
The community schools movement also draws upon the principles of collective action in its approach to im- proving educational outcomes. Using public schools as a hub, community schools build relationships among educators, families, community-based organizations, business, health and social service agencies, and youth development organizations to implement activities that promote high educational achievement and use the community as a resource for learning.18 The community schools approach builds upon the idea of “joint use” of district-owned school facilities by non-district entities,19 but it’s not simply a matter of co-location of services; com- munity schools focus on fostering strong partnerships and strategically integrating diverse services to achieve spe- cific, measurable results.20 A critical component of the community schools approach is a full-time “community school coordinator,” who is responsible for overseeing and integrating services in a coordinated fashion, while also participating on the management team for the school. In the absence of such coordination, each individual service used by a student occurs in isolation, and the likelihood
of identifying a critical service gap that may have multiple downstream effects is diminished. By strategically inte- grating across schools and community services, the com- munity schools approach aims to meet the full spectrum of a child’s educational and developmental needs, with the primary purpose of improving educational outcomes (to learn more about a community school approach in Oregon, see the article “ Schools Uniting Neighborhoods: Community Schools Anchoring Local Change” in this issue of CI).
Conclusion
This is not to suggest that any one of these specific pro- grams or approaches will be the silver bullet for improving educational outcomes for youth. But all of the approaches noted above share a common feature – cross-sector align- ment. This resonates with the growing conviction within the community development industry that multi-sector approaches can more effectively transform communities than the siloed single sector approaches of the past. In en- gaging more intentionally and systematically in initiatives that aim to support the educational achievement of LMI students, the field can deepen the impact of its efforts in affordable housing, workforce development, accessible fi- nancial services, place-based revitalization, and commu- nity development finance, all of which help foster an en- vironment where children can learn, thrive, and succeed. Community development experts should seek out op- portunities to build relationships with local education in- tegration initiatives, such as the Strive Cradle to Career
Figure 3. What Happens in a Community School?
Key Areas for Programs and Services
Expected Outcomes
Quality education
High-caliber curriculum and instruction enable all children to meet challenging academic standards and use all of the community’s assets as resources for learning.
Youth development
Young people develop their assets and talents, form positive relationships with peers and adults, and serve as resources to their communities.
Family support
Family resource centers, early childhood development programs, and coordinated health and social services build on individual strengths and enhance family life.
Family and community engagement
Family members and other residents actively participate in designing, supporting, monitoring and advocating quality activities in the school and community.
Community development
All participants focus on strengthening the social networks, economic viability and physical infrastructure of the surrounding community.
8
Community Investments, Summer 2012 – Volume 24, Number 2
Source: Coalition for Community Schools

Network or the Coalition for Community Schools. They can also engage with metropolitan planning partnerships that work with local schools districts, such as the efforts of UC Berkeley’s Center for Cities and Schools.
Just as the challenges and resources facing each com- munity are unique, the appropriate responses and op- portunities for community development engagement with schools must be locally driven. There’s no single
Linking Community Development with Schools
Adapted from “Connecting Public Schools to Community Development” by Connie Chung21
Community development organizations can link up with schools in a variety of ways, depending on their institutional experiences, focus, and capacity.
Public School Facility Development
Community development practitioners can use their real estate and development expertise to support a school district’s efforts to build more neighborhood- oriented school facilities. Some community groups have worked to bring new public schools to their neighborhoods by developing and leasing property to school districts. Community development practi- tioners can also use their development expertise to assist in the rehabilitation and enhancement of exist- ing school facilities.
Affordable Housing Development
Developers of affordable housing can boost the long-term viability of their projects by investing in the quality of nearby schools. Similarly, community development groups can make neighborhood im- provements near a public school to attract families and qualified teachers to a neighborhood. Improv- ing schools and the surrounding area can be a particularly useful strategy to support the success of mixed-income housing projects. A high-quality school in the neighborhood can entice home buyers to purchase market-rate units in a mixed-income development. Additionally, community development practitioners can work with schools to develop work- force housing for teachers, enabling school staff to live in the communities they serve.
Economic Development
Schools are often the largest institutions and employ- ers in a neighborhood, making them an invaluable partner in economic development efforts. Community development organizations can harness this econom- ic influence by linking schools with the local business community and labor force. Some communities are also partnering with schools to provide job-training and trade school classes for community members.
Joint Uses
Community groups can also promote the shared use of facilities between schools and other community en- tities. The joint use of a library or a park, for instance, offers an effective solution in urban areas where land for new community facilities is not readily available. In rural areas, shared use projects can make economic sense for communities that must concentrate their re- sources.
Transportation
Community development organizations can work with public school districts to alleviate neighborhood traffic concerns. Many community development or- ganizations are advocating the placement of schools within walking distance of residential areas and transit stops. These transportation strategies not only reduce school traffic in neighborhoods, but also help to address childhood obesity by encouraging children to walk to school.
answer for how to best link community development with schools, but systematic integration and strong partner- ships should be seen as fundamental to any approach that is taken (see the section “Linking Community Develop- ment with Schools” below for ideas on partnership). As Secretary Duncan emphasizes, we can no longer afford to take an “either-or” approach to education—we must consider how we can improve the “both-and” solutions.
Community Investments, Summer 2012 – Volume 24, Number 2
9
Special Focus: Community Development and Education
10
Community Investments, Summer 2012 – Volume 24, Number 2
The New Civic Infrastructure:
The “How To” of Collective Impact and Getting a Better Social Return on Investment
By Jeff Edmondson, Strive Network and Nancy L. Zimpher, State University of New York
Secretary Arne Duncan recently said, “Many people believe we have to first address poverty in order to improve education. I believe we have to first improve education in order to address poverty.”1 If you agree with the secretary, it is easy to see that education is the single most important engine of indi- vidual opportunity and economic growth in our country. The question then becomes: In this challenging economy where new resources are scarce, how do we make critical improvements so that we get a better return on our current investment?
To answer this question, leaders from the education, business, nonprofit, civic, and philanthropic sectors in the urban core of the Greater Cincinnati region joined
together in 2006 to form The Strive Partnership. The Part- nership focused on an ambitious vision—supporting the success of every child, every step of the way, from cradle to career—and a corresponding set of ambitious goals: working together to ensure every child is prepared for school, is supported inside and outside of school, suc- ceeds academically, enrolls in some form of college, and graduates and enters a career.
But most importantly, the Partnership identified and set measurable targets for a core set of eight overarching outcomes that span the cradle to career continuum. Prog- ress toward meeting these targets are tracked across the three cities that make up the urban core of the region for early childhood, the public and parochial schools, and the

local colleges and universities. In order to make prog- ress toward the designated targets, relevant practitioners and funders formed networks related to each outcome to review local data on their performance and build co- hesive action plans around what actually works in terms of helping students succeed. The result: of the 34 mea- sures of student achievement on which the Partnership is focused, 81 percent are trending in a positive direction, up from 68 percent three years ago.2
This work overall, and the networks of practitioners and funders specifically, provides the community-de- velopment sector with a new way to engage with the education field and improve outcomes for children. By working arm-in-arm with education systems—early child- hood, K-12, and higher education—and using data as a constructive tool to improve, as opposed to a tool to pick winners and losers, we can begin to leverage precious resources to get the improved results we all so desire. This article summarizes some of the lessons learned from The Strive Partnership’s experience in Cincinnati/North- ern Kentucky, as well as the experience of other pioneers in this work, and identifies a framework for building the “New Civic Infrastructure” needed to support the success of every child from cradle to career and move the dial on critical social outcomes in general.
Defining How to Have “Collective Impact”
In their popular article in the Stanford Social Innova- tion Review, John Kania and Mark Kramer define collec- tive impact as, “The commitment of a group of impor- tant actors from different sectors to a common agenda for solving a specific social problem.”3 This simple defini- tion has caught the imagination of communities across the country looking to address complex social issues in a struggling economy. In the end, this concept gives us a way to think differently about how to get a better social return on investment.
But as is often the case, a great idea can spread so quickly and be adapted in so many ways that its original or true meaning can become muddled or lost. Recently, one community reached out to us and claimed to have nine collective-impact initiatives underway related to education. When asked about what was common or col- lective across the efforts, there was no clear answer. That there were so many separate but similar initiatives oper- ating simultaneously is antithetical to the entire point of collective impact.
In order to prevent the concept of collective impact from getting diluted, it is critical to establish some basic standards for what it takes to make this very challenging work happen on the ground. Fortunately, long before the “Collective Impact” article hit the press, a consor- tium of financial institutions and foundations known as
. . . we need to shift our focus to how we align public and private resources in new ways so we can effectively deploy all resources at our disposal, regardless of the source, to improve outcomes for children.
Living Cities funded leaders of The Strive Partnership to gather lessons from their work and see how they could inform similar work in four other communities. Since this initial investment, the work of the Strive Partnership has spread beyond Cincinnati and a separate effort called the Strive Network was launched in 2011 to build a national network of cradle to career communities. To date, over 150 communities have reached out to learn about this work, and our staff has worked with over 20 communities to help them move from aspirations of collective impact to real action on the ground.
Our most important finding from all this work could not be less flashy. It turns out that the key to improv- ing student outcomes at the population level is not a program, but a process. It is clear that no single program, no matter how effective, can be scaled to solve all our education challenges. Instead, we need to return to an age-old process that has itself been watered down over the years: employing disciplined team work to build civic infrastructure.
In the many definitions that can be found, there are two key themes regarding civic infrastructure that require us to think differently about this work as we move forward. First, civic infrastructure has historically been primarily focused on how myriad public sector resources are aligned for “building a shared sense of belonging and purpose, facilitating the setting of shared goals and coor- dinating action.”4 However, in this economy, we cannot rely on the public sector alone, regardless of how we co- ordinate our efforts. Instead, we need to shift our focus to how we align public and private resources in new ways so we can effectively deploy all resources at our disposal, re- gardless of the source, to improve outcomes for children.
Second, the historical definition of civic infrastructure can potentially be confused with the softest versions of collaboration—a loose affiliation or connection of pro- grams and services focused on similar ends, but which continue to operate in silos. As one site we worked with expressed at the outset of their efforts, “I fear this will end up just becoming another ‘kumbaya circle’ where every- one talks about working together but keeps on doing the exact same thing.” We must take a more rigorous and focused approach to coordinating these disparate efforts if we want to avoid reverting to the status quo.
Community Investments, Summer 2012 – Volume 24, Number 2
11
Special Focus: Community Development and Education
12
Community Investments, Summer 2012 – Volume 24, Number 2
The New Civic Infrastructure: Putting Data to Work
The new civic infrastructure responds to both of these challenges by ensuring we bring together cross-sector leaders at several levels to focus their collective energy not on talking, but on actually developing and continu- ously improving concrete action plans for how to move common outcomes forward. And the key ingredient for making this focused action planning possible is pretty simple—it’s data.
As Jim Collins highlights in Good to Great for the Social Sector, the disciplined use of data to drive where we focus our energy and what we do to have impact is our single greatest challenge to improving social outcomes at scale.5 Specifically, as it relates to education, the new civic infrastructure responds to this challenge by enabling community leaders across sectors and at all levels to use data in a more purposeful way to: (1) identify those prac- tices that actually get results for children, (2) invest the community’s precious resources differently to increase impact, and (3) hold themselves accountable for moving specific outcomes across the cradle to career continuum.
Whether an individual likes the federal No Child Left Behind legislation or not, it provides a concrete mecha- nism to have data on the educational outcomes of every single child. We no longer have an excuse for not using data to, at a minimum, help us focus on our greatest areas of need collectively and identify those practices that ac- tually get results for children individually. And if we do not like the data we have at our disposal—and concerns about the standardized tests are justifiable—it is now in- cumbent upon us to improve these measures rather than simply complain about them.
Establishing Standards for Collective Im- pact: The Framework for Building Cradle to Career Civic Infrastructure
Strive has developed the Framework for Building Cradle to Career Civic Infrastructure by drawing upon lessons not just from the pioneering work in Cincinnati and Northern Kentucky with The Strive Partnership, but from talking and working with more than 150 communi- ties across the country that are considering undertaking this challenging work. Our most important lesson learned is that there is no single model for how to do this. One community can’t simply do exactly what another did, as the local assets always vary.
Instead, the Framework acts as a guide to building civic infrastructure by helping communities identify their critical gaps as well as local assets, and knit together their investments in children in new and different ways. It is important to note that no community starts building civic
Framework for Building Cradle to
Career Civic Infrastructure
Examples of key standards of practice related to each of the four pillars of the Framework for Building Cradle to Career Civic Infrastructure include:
Shared Community Vision:
There must be consistent engagement of top-lev- el executives across at least five sectors – educa- tion, business, civic, non-profit, and philanthropic – around a common vision.
An “accountability structure” is developed and defined that enlists community partners at all levels who commit to specific operating principles for working together to improve student outcomes.
Evidence Based Decision Making:
A limited number of student outcomes is adopted and reported transparently on a regular basis to show population-level trends.
Comprehensive data-management systems are in place to monitor how individual students benefit from an array of support services and how this work collectively feeds up to population-level im- provements.
Collaborative Action:
Networks of existing practitioners come together around priority outcomes, agree on a common set of measures to understand impact, and utilize a continuous-improvement process to use data to get better over time.
Cross-sector leaders define clearly how they will support networks of practitioners to develop and implement collaborative-action plans, such as fa- cilitation, data analysis, and advocacy.
Investment & Sustainability
Core staffing is in place to shepherd the work forward, including a director to consistently “herd the cats,” keeping participants focused on the common vision and outcomes, and a data analyst to ensure information is made available in such a way to inform decision making.
Funders are actively engaged to repurpose exist- ing investments over time behind collaborative- action plans developed by networks.

infrastructure from scratch: by following Strive’s tested process, they should very intentionally walk through a rig- orous process to build on existing strengths to fill in gaps.
The Framework consists of four pillars that highlight specific areas a community needs to consider when build- ing civic infrastructure (see sidebar on previous page). Two of these pillars deal directly with how communities use data at different levels: at the community level to identify the most critical issues and the individual level to identify what practices are really having an impact on children. The other two pillars of the Framework ensure key leadership is in place to advocate for what works and other indispen- sible factors for sustaining the work, such as community voice, funder alignment, and critical staffing are in place to ensure improvements continue over the long term.
The evolving Strive National Network has developed a Progress Assessment Tool that offers significant detail around each of these pillars so that a community can better understand how this process of infrastructure build- ing might unfold from start to finish.6 This tool provides a critical first attempt at establishing detailed standards of practice with regard to how we can best achieve collective impact. The specifics behind the Framework are constantly being updated as sites learn more about how to sustain the civic infrastructure. Indeed, it is this practical, real-world experience that must inform these standards if we are to ensure that collective impact is more than a passing fancy.
Implications for the Field: Getting a Better Social Return on Investment
The potential implications of creating uniform stan- dards of practice for building civic infrastructure could have far-reaching effects on how we invest our resources to address social issues. The current method of tackling these problems is primarily through a Request for Propos- als (RFP) process. Using the RFP, funders identify a prac- tice they wish to test and scale, and practitioners hasten to develop proposals that align with a funder’s given interest.
The problem with this approach is that it perpetuates a “spray and pray” mentality for addressing social problems: we spray new ideas and related resources all over the place and pray that good things will come of it. Rarely do the efforts that result align effectively with current work, and communities end up with one more “point of light” that may or may not target the most pressing issue and scale the most effective practice.
By building the civic infrastructure, public and private investors can identify communities that are already taking a more strategic approach to collectively improving an outcome they are interested in seeing move. They can engage with the community leadership to understand the
current plan and identify ways to complement the existing work of a network of practitioners, instead of dropping a new idea into the mix of work already underway.
Communities that build this kind of civic infrastructure could be ripe for the emerging “Pay for Success” concept being tested across federal agencies.7 In this concept, the federal government will “guarantee” an investment by a private donor if a proposed intervention actually leads to a specifically defined outcome—not the number of people served, but the measurable improvements felt by the people served. In the end, the government is able to target its dollars more effectively, and private funders can reinvest dollars they recover back into the emerging prac- tices that are getting results.
In short, those communities that have built the civic infrastructure have: (1) the staffing to make sure an action plan is implemented over time, (2) the data in hand to constantly monitor progress toward the outcome, and (3) a process for leveraging and scaling what really gets results. Investments are more secure and the potential for widespread impact is increased.
The final result of this work, and the yardstick by which this new civic infrastructure will be measured, is social return on investment. Cradle to career civic infrastruc- ture puts in place systems that assess whether the dollars being invested toward a given outcome are going further than they otherwise would, helping us answer the age-old question, “Are we getting more bang for our buck?” The investment is minimal—it does not have to be more than $500,000 in overhead—but the impact can be utterly transformational.
Conclusion
In the “new normal” where resource limitations are a fact of life, it is more necessary than ever to ensure we are investing our time, talent, and treasure as efficiently and effectively as possible.8 The concept of collective impact gives us the conceptual underpinnings for how to make this change. But in order for us to prevent a powerful idea from becoming a watered-down version of what it was meant to be, we need a common set of standards for what it means to make this work happen. The new civic infrastructure, informed by practical experience on the ground, is a way to not only make this concept a reality and develop common standards, but completely rethink how we get a better social return on investment when tackling some of our most challenging issues.
Jeff Edmondson is Managing Director of the Strive Network and Nancy L. Zimpher is Chancellor of the State University of New York.
Community Investments, Summer 2012 – Volume 24, Number 2
13
Special Focus: Community Development and Education
14
Community Investments, Summer 2012 – Volume 24, Number 2
Schools Uniting Neighborhoods:
Community Schools Anchoring Local Change
By Diana Hall, Multnomah County
Across the United States, communities are think- ing differently about the challenges they face to achieving community prosperity and health. In- creasingly, youth educational success is being recognized as a cornerstone for the attainment of a wide array of key outcomes including poverty reduction and improvements in physical and mental health, public safety and community vitality.
In Multnomah County, Oregon, which includes the City of Portland and is home to roughly 750,000 people, the community has made youth educational achievement a priority, and has developed an innovative and highly successful model for cross-sector collaboration. The part- nership, known as SUN Community Schools, brings to- gether schools and partners from across the community to collectively impact educational success and family self- sufficiency.
Recognizing the need for support at all ages and at- tention to transitions in and out of the K-12 system, as well as between grade levels, SUN Community Schools are located in elementary, K-8, middle and high schools. The focus is on the whole child, integrating academics, social services, supports and opportunities in order to meet student and family needs. The specific services and programs offered are tailored to the individual assets and needs of a school, and community resources are orga- nized strategically to support student success. This article describes SUN’s community school approach and high- lights emerging opportunities for the community develop- ment field to work in closer partnership with schools.
Schools as Centers of the Community
While many public schools offer before- and after- school activities, Schools Uniting Neighborhoods (SUN)

goes much further by reinventing the school as a place that addresses the full spectrum of family needs. On a typical day, the school opens early, providing students with a safe place to eat breakfast, do homework or participate in rec- reational and skill-building activities. During the school day, a community school site manager is working with school staff, families and community partners to identify specific student and family needs and broker services, and develop innovative partnerships that bring critical child and family supports directly to campus. Families and com- munity members come to the school for advisory commit- tee and leadership activities, parenting classes, health sup- ports, food, clothing, and access to anti-poverty, mental health, and other social services.
The day does not end with the school bell. Instead, the extended-day program picks up where the traditional classroom leaves off, providing students with an array of academic, enrichment and recreational activities that complement and build on the school day. Students receive a full dinner and engage in physical activity to improve their physical health. In the evening the school serves as a community center offering a variety of educational and recreational activities for adults and youth as well as pro- viding space for community projects and meetings. This is SUN carrying out the vision its name so clearly describes, truly using schools to unite neighborhoods.
Rooted in Collaboration
In the late 1990s, Multnomah County community members and leaders recognized a need for a new ap- proach. The environment posed multiple challenges in- cluding shrinking budgets, a significant racial achieve- ment gap, growing poverty, a severe shortage of affordable housing, and an increase in the number of children being left unsupervised during out-of-school hours. Demograph- ic changes were dramatically increasing the cultural and linguistic diversity in the region, requiring schools and social service organizations to develop new skills in order to educate and support these populations effectively.
An individual’s level of educational attainment is the primary predictor of poverty in adulthood. The effect of family poverty on school success was also clear, as bar- riers such as homelessness, mobility, hunger, illness, and trauma made it impossible for many students to come to school ready to learn. It became clear that you couldn’t talk about alleviating or eliminating poverty without talking about education.
With leadership from elected officials in the City of Portland and Multnomah County, the decision was made to partner together to support schools. The initial goal was two-fold: (1) to support education and school success and (2) to improve the way resources for students and their families were delivered by developing a school-
A SUN Community School is not a program, but rather a place and support hub where schools and communities work together to have a collective impact on the success of children and families . . .
based delivery model. An ad hoc committee of a broad array of stakeholders was convened to determine the best strategy to accomplish this goal. The committee included leadership from an existing Community Building Initiative and After School Cabinet. After a year of research and deliberation, the full-service community school model was chosen and the first eight SUN Community Schools were implemented in the fall of 1999, with the city and county providing core funding.
Since that time, the community has chosen to expand SUN Community Schools from 8 to 67 schools with a vision for every school to be a SUN Community School. Supportive policy has been adopted in the county, city and school districts and a more expansive network of care, named the SUN Service System, has been developed to organize and prioritize the county’s investments and part- nerships to support school age children and their families.
Community Schools as a Place-Based Strategy
A SUN Community School is not a program, but rather a place and support hub where schools and communities work together to have a collective impact on the success of children and families and provide a comprehensive array of services. The strategy reflects the unique needs of each school and its neighborhood. Schools are ideal locations for these focused efforts, both because they are the most direct and convenient route for reaching stu- dents and families, but also because, in the majority of neighborhoods most impacted by poverty and low student achievement, schools are the only public facilities.
SUN Community Schools also serve as a platform for implementing other place-based education reform initia- tives, such as Cradle to Career. In recent years, Portland and Multnomah County leaders have been building a Cradle to Career (C2C) framework adapted from the Strive model in Cincinnati, Ohio (see the article “The New Civic Infrastructure” in this issue for more information). Portland is one of seven national demonstration sites for the Strive Network’s C2C Initiative. Leaders see SUN as a core strat- egy for the local C2C initiative and SUN is aligned with the evolving C2C structure. Many of SUN’s community and school leaders serve on C2C governance, planning and action committees, such as the C2C.
Community Investments, Summer 2012 – Volume 24, Number 2
15
Special Focus: Community Development and Education
16
Community Investments, Summer 2012 – Volume 24, Number 2
The inclusion of non-profit partners in
the role of lead agency capitalizes on the unique capacity of these community-based organizations.
The SUN Model
As full-service neighborhood hubs, where school and community partners work together to ensure kids and families have what they need to succeed, SUN Communi- ty Schools serve as the vehicle to link community institu- tions, such as libraries, parks, community centers, neigh- borhood health clinics and area churches and businesses.
At the school site, SUN Community Schools mobi- lize and strategically organize community resources to provide:
• Strong core instructional program;
• Educational support and skill development for youth and adults;
• Enrichment and recreation activities;
• Family involvement and support;
• Social, health and mental health resources;
• Family and community events.
In the SUN model, a non-educational lead agency partners with an individual school and together, with help from school and community leaders, they co-manage the community school collaboration at the site. The inclusion of non-profit partners in the role of lead agency capitalizes on the unique capacity of these community-based organi- zations. That capacity includes expertise in anti-poverty services, youth and family engagement and community development fields; relationships and standing within communities and with community leaders; and the ability to fund- and “friend”-raise in ways that governments and educational agencies cannot. Lead agencies receive core funding that supports the hiring of a SUN Community School site manager as well as limited flexible dollars to fill resource gaps in key underfunded services.
The use of site managers is an essential component of the SUN model. Site managers coordinate and broker services at the school and support the development of the partnerships and collaboration between the school and its youth, families and community. Effectively, they act as the “glue” to attach all the community resources – from public services to neighborhood volunteers – to the school in a strategic way. It is critical to have a dedicated person with the capacity, in both time and skills, to carry out these
functions. SUN has learned that absent such a position, schools and communities are unable to develop or sustain such strategic collective efforts and thus, unable to make a significant impact.
In addition to site management, or coordination as it is called in many other community schools initiatives across the country, there are four other defining components to the SUN model at the site level. The first is that the array of services and programs provided to youth and adults in- cludes offerings from academics and skill development to social, health and mental health services to enrichment and recreation. Second, services are planned, developed and implemented within the context of youth, family and community engagement. Engagement is a way of doing business and building relationships to form the SUN col- laboration, rather than a service or activity offered to the community. Authentic engagement helps ensure that what happens in the SUN sites is culturally appropriate, rele- vant and targeted at the issues most affecting student and family success. District and school support, particularly that of the school principal, is the third essential ingredi- ent in ensuring the connection of supports with educa- tional success and in influencing school reform.
Lastly, SUN’s model calls out an important element for developing true collaboration: shared leadership and accountability. At the school site, the principal and the community school site manager share leadership across the school day and the out-of-school time. Principals often refer to the site manager as an assistant principal. A broad group of stakeholders participate in advisory and leader- ship roles using an annual planning process that is aligned with the school improvement plan. Progress and results are documented and shared with the community.
Community-Level Collaboration
SUN is a multi-jurisdictional partnership and its spon- soring partners each see the community schools strategy as advancing their core mission. Community schools are a vehicle for everyone to get their work done - whether that work is education, crime prevention, anti-poverty, com- munity and economic revitalization, workforce develop- ment or other community-focused efforts. SUN sponsors have understood from the beginning that none of the organizations can accomplish their missions by working alone. This understanding has become all the clearer to SUN’s partners, as economic realities have worsened at the same time that expectations of the organizations pro- viding services, particularly educational institutions, have continued to grow.
SUN sponsors share responsibility and investment in the community schools model. Shared governance and accountability happen through the SUN Service System Coordinating Council, which has representation from

the city, county, six school districts, the State of Oregon, the Coalition of Communities of Color, the Commis- sion on Children, Family and Community, the Cradle to Career backbone organization, business, non-profit pro- viders and youth. The county, city (including a local levy for children’s services) and school districts contribute $7 million annually to fund the core functions of SUN at the 67 sites. That contribution then leverages and attracts ap- proximately $17 million in other resources to those local communities. In the broader SUN Service System, over $30 million in additional service funding is aligned and delivered through SUN Community Schools and regional school-linked centers.
Historically, the connections between SUN Communi- ty Schools and local community development have been limited to planning and development efforts supported by the City of Portland’s Bureau of Planning and Sustainabili- ty and local community development organizations in dis- crete neighborhoods. More recently, however, the City’s planning process and resulting Portland Plan, which lays out the roadmap for the next 30 years, heavily involved SUN partners and called out SUN in all three integrated strategy areas: (1) Thriving Educated Youth, (2) Economic Prosperity and Affordability, and (3) Healthy Connected City. Emphasizing the ability of the community schools model to create more efficient and effective change through alignment and partnership, the Plan includes the transformation of every school into a SUN Community School as a specific action.
The Impact of Collective Efforts
Educational success and self-sufficiency are inherently issues of equity, and the SUN effort is recognized as a strat- egy for achieving equity. One aspect of SUN’s success is the degree to which the children, youth and families most affected by disparities (people of color and those living in poverty) are served and included in its efforts. SUN Com- munity Schools consistently serve the most vulnerable kids and families and are structured to ensure that individual student and family needs are identified and met through the coordination of services. In 2010-11, of the 19,127 children and youth served in enrolled services, 80 percent qualified for Free and Reduced Lunch, 70 percent were children of color and 21 percent were English Language Learners. These rates compare to the surrounding districts’ rates of 54 percent, 46 percent and 14 percent respectively.
By coming together, the community is supporting these vulnerable students and families in a significant way. Annual evaluations conducted by Multnomah County using school district, teacher and student data demon- strate the consistent effect of SUN. In the 2010-11 school year, regularly participating students showed strong results in academics, attendance and behavioral areas including:
• •
• •
•
74 percent of students met state benchmarks or growth target in Reading;
Students’ average benchmark gains were equal to or higher than expected in the majority of grades;
Average daily school attendance was 94.5 percent;
74 percent of students improved in at least one behav- ioral or academic area (such as behaving well in class, motivation to learn, or homework completion); and
86 percent of students reported having at least one adult who cares about them and to whom they can go for help.
Community Investments, Summer 2012 – Volume 24, Number 2
17
In addition, 96 percent of families who receive anti- poverty case management, life and job skills services, rent assistance and other basic needs support remained in per- manent housing after support ended. The community also reports improvement in other related indicators including parent involvement, community safety and vibrancy.
Special Focus: Community Development and Education
18
Community Investments, Summer 2012 – Volume 24, Number 2
Evolving in an Environment of Continuous Change and Learning
The SUN model is built on a strong history of com- munity involvement and school partnerships in the region. Multnomah County and the City of Portland each had a history of investing in services delivered by community- based non-profit organizations in local schools. Imple- mentation of SUN drew on successful existing programs and initiatives and was done without new money. What the SUN initiative learned from this experience was that it is possible to create systems change and develop a new model by drawing on existing resources. In fact, declin- ing budgets forced institutions to reconsider how they could work more efficiently and effectively, and capitalize on partners to achieve their goals. It also became clear that while new money might become available, usually through time-limited grants, the community had large amounts of funding in existing systems already dedicated to serving youth and families. Due to their size and sus- tainability, it is those resources that offer the greatest pos- sibility for fostering community-wide change, if they can be evaluated and redeployed in innovative ways.
Systems change requires patience, persistence, and the ability to exhibit flexibility and teamwork—unsurpris- ingly, the same skills we seek to foster in young people. Flexibility and adaptability are also essential aspects of the community school model, which make it well suited for supporting other initiatives, whether place-based or issue- focused.
Lastly, SUN’s experience highlights that it takes ca- pacity to build capacity. Its success in building a highly- functioning collaboration and system of care has required the dedication of both financial and human resources. Funding is dedicated for a small staff that carries out in- termediary functions for the multi-jurisdictional partner- ship, including convening, planning, policy development, contract management, program development, evaluation, technical assistance and professional development. In addition, school districts and other sponsoring partners commit the time of leaders, designated liaisons and other staff to support shared governance and alignment within their home organizations.
Implications for the Community Development Field
The community schools strategy offers a tremendous opportunity for the community development sector to
impact educational achievement without moving outside its expertise or getting derailed by the daunting world of education reform. Community schools can assist commu- nity development in achieving the inclusive and multi- faceted interventions necessary to address the complexi- ties that exist in communities impacted by the intertwined issues of poverty, place and racism.
There are a variety of ways that community develop- ment entities can partner with community schools and take advantage of the infrastructure, relationships and lev- eraged resources they offer. One key way is to promote and support the inclusion of community schools as a strategy in other place-based initiatives, such as Promise Neigh- borhoods, Choice Neighborhoods, and Enterprise Zones, among others. Many traditional community development initiatives can be offered as part of community school efforts or partnered with community schools to have an amplifying effect. Stable, affordable housing and individ- ual savings accounts are two examples of such initiatives.
Community development professionals can also support education and community schools in a more sys- temic way by contributing their expertise and social capital to collaborative projects, particularly their relationships within the private sector. Technical assistance and finan- cial advice on ways to take community school models to scale, adapt continuous quality improvement systems and engage the private sector most effectively would be of great assistance to those immersed in the worlds of educa- tion, social service or local government.
But where to begin? One simple step that all com- munity development organizations can do is to get to know the schools in the communities they are seeking to improve. Invite the schools to the conversation, ac- knowledging that their success in educating youth has a significant impact on community-wide long-term success. Like all people working for positive change, educators are acutely aware that they need the support of their commu- nities to accomplish their mission. And, indeed, many of the barriers to learning lie outside education’s role, such as illness, hunger, and poverty. These are the responsibility of the community – local governments, neighbors, busi- nesses and community development entities. It’s time for us to open conversations and doors to each other.
Diana Hall is program supervisor for Multnomah County Department of County Human Services, School and Com- munity Partnerships.

PHOTO 6
The Widening Academic Achievement Gap between the Rich and the Poor
By Sean F. Reardon, Stanford University1
Almost fifty years ago, in 1966, the Coleman Report famously highlighted the relationship between family socioeconomic status and
2
student achievement. Family socioeconomic
characteristics continue to be among the strongest predic- tors of student achievement, but while there is a consider- able body of research that seeks to tease apart this relation- ship, the causes and mechanisms of this relationship have been the subject of considerable disagreement and debate.
Much of the scholarly research on the socioeconom- ic achievement gradient has focused largely on trying to understand the mechanisms through which factors like income, parental educational attainment, family struc- ture, neighborhood conditions, school quality, as well as parental preferences, investments, and choices lead to dif-
ferences in children’s academic and educational success. Still, we know little about the trends in socioeconomic achievement gaps over a lengthy period of time.
The question posed in this article is whether and how the relationship between family socioeconomic characteristics and academic achievement has changed during the last fifty years, with a particular focus on rising income inequality. As the income gap between high- and low-income families has widened, has the achievement gap between children in high- and low- income families also widened? The answer, in brief, is yes. The achievement gap between children from high- and low-income families is roughly 40 percent larger among children born in 2001 than among those born twenty-five years earlier.
Community Investments, Summer 2012 – Volume 24, Number 2
19
Special Focus: Community Development and Education
Trends in Socioeconomic Status– Achievement Gradients
To begin with, consider the difference in achievement between children from high- and low- income families. One way to measure this difference is to compare the average math and reading skills3 of children from families with incomes at the 90th percentile of the family income distribution (about $160,000 in 2008) to those in families with incomes at the 10th percentile of the family income distribution (about $17,500 in 2008),4 hereafter referred to as the “90/10 income achievement gap.”
Figures 1 and 2 present the estimated 90/10 income achievement gap for cohorts of students born from the mid-1940s through 2001.5 These estimates are derived from thirteen nationally representative studies available that include family income as well as reading and/or math scores for school-age children.
Although the tests used are not exactly comparable across all the studies included, both figures show a clear trend of increasing income achievement gaps across cohorts born over a nearly sixty-year period. The estimated income achievement gaps among children born in 2001 are roughly 75 percent larger than the estimated gaps among children born in the early 1940s. The gap appears to have grown among cohorts born in the 1940s and early 1950s, stabilized for cohorts born from the 1950s through the mid-1970s, and then grown steadily since the mid- 1970s. Although the trend in achievement gaps prior to 1970 is somewhat unclear, the trend from the mid-1970s to 2001 appears relatively clear—statistical models in-
dicate that the income achievement gap has grown by roughly 40 to 50 percent within twenty-five years, a very sizable increase.
One important question is whether the trend in the income achievement gap is driven by the changing racial and ethnic composition of the U.S. population. In sepa- rate analyses,6 I find that the income achievement gap grew within the white, black, and Hispanic student popu- lations separately, as well as within the population as a whole. For whites and Hispanics, the income achieve- ment gap appears relatively stable through the mid-1970s and begins to grow rapidly thereafter; for blacks, the gap appears to grow steadily from the 1940s through 2001.
How Large Are These Gaps?
Figures 1 and 2 report income gaps in standard-devi- ation units. Although this is a metric familiar to research- ers and one that is useful for comparing the size of gaps across studies using different tests, it may not be immedi- ately obvious how large these gaps are in real terms. One way to get a sense of the size of the gaps is to compare them to the amount that an average student learns during the course of a year. Data from the National Assessment of Educational Progress (NAEP) indicate that the average student gains 1.2 to 1.5 standard deviations in math and reading between fourth and eighth grade and between 0.6 and 0.7 standard deviations in math and reading between eighth and twelfth grade.7 Thus, a gap of 1 standard devia- tion is substantively very large, corresponding to roughly 3 to 6 years of learning in middle or high school.
Figure 1. Trend in 90/10 Income Gap in Reading, 1940–2001 Cohorts Trend in 90/10 Income Gap in Reading, 1940-2001 Cohorts
Source: Author’s compilation based on data from Project Talent (Flanagan et al. n.d.); NLS, HS&B, NELS, ELS, ECLS-K, ECLS-B (U.S. Department of Education, Center for Education Statistics 1999, 2000, 2001, 2004, 2009, 2010); Prospects (U.S. Department of Education 1995); NLSY79, NLSY97 (U.S. Bureau of Labor Statistics 1980, 1999); SECCYD (National Institute of Child Health and Human Development 2010); and Add Health (Harris 2009). Note: See note 3 and online appendix for further details.
1.50 1.25 1.00 0.75 0.50 0.25 0.00
Study
TALENT
NLS
HS&B
NLSY79
NELS
Add Health Prospects
NLSY97
ELS
SECCYD
ECLS-K
ECLS-B
Quartic Fitted Trend
1940 1950
1960 1970
1980 Cohort Birth Year
1990 2000
20
Community Investments, Summer 2012 – Volume 24, Number 2
Average Difference in Standardized Test Scores Between 90th & 10th Income Percentile Families
Special Focus: Community Development and Education
1.50 1.25 1.00 0.75 0.50 0.25 0.00
1940
Study
TALENT
NLS
HS&B
NLSY79
NELS
Prospects
NLSY97
ELS
SECCYD
ECLS-K
HLS
ECLS-B
Quartic Fitted Trend
Source: Author’s compilation based
on data from Project Talent (Flanagan et al. n.d.); NLS, HS&B, NELS, ELS, ECLS-K, HLS, ECLS-B (U.S. Department of Education, Center for Education Statistics 1999, 2000, 2001, 2004, 2009, 2010); Prospects (U.S. Department of Education 1995); NLSY79, NLSY97 (U.S. Bureau of Labor Statistics 1980, 1999); and SECCYD (National Institute of Child Health and Human Development 2010).
Note: See note 3 and online appendix for further details.
Figure 2. Trends in 90/10 Income Gap in Math, 1940–2001 Cohorts Trend in 90/10 Income Gap in Math, 1940-2001 Cohorts
1950
1960
Cohort Birth Year
1990
2000
Another way of getting a sense of how large these gaps are (and how meaningful their trend is) is to compare the income achievement gaps to contemporaneous black- white achievement gaps. The black-white achievement gap narrowed substantially among cohorts born from the mid-1950s through the mid-1970s—by roughly one-half a standard deviation—according to NAEP data.8
Figures 3 and 4 display both the 90/10 income gaps (as shown in Figures 1 and 2) and the black-white achieve- ment gaps as estimated from the same samples.9 In each figure the solid line indicates the estimated trend in the 90/10 income achievement gap. For comparison, the es- timated black-white achievement gap from each study is displayed in the figure (the hollow circles), along with a
1.50 1.25 1.00 0.75 0.50 0.25 0.00
Study
TALENT
NLS
HS&B
NLSY79
NELS
Add Health
Prospects
NLSY97
ELS
SECCYD
ECLS-K
ECLS-B
Income Gap
Black-White Gap
Black-White Gap (NAEP)
Source: Author’s compilation based on data from Project Talent (Flanagan et al. n.d.); NLS, NAEP, HS&B, NELS, ELS, ECLS-K, ECLS-B (U.S. Department of Education, Center for Education Statistics n.d., 1999, 2000, 2001, 2004, 2005, 2009, 2010); Prospects (U.S. Department of Education 1995); NLSY79, NLSY97 (U.S. Bureau
of Labor Statistics 1980, 1999); SECCYD (National Institute of Child Health and Human Development 2010); and Add Health (Harris 2009).
Note: Solid symbols represent 90/10 income achievement gaps; hollow symbols denote black-white achievement gaps. See online appendix section 5.A5 for further details.
Average Difference in Standardized Test Scores (90/10 Income Gap or Black-White Gap)
Average Difference in Standardized Test Scores Between 90th & 10th Income Percentile Families
1940
1950
1960
1970
1970
1980
Figure 3. Trends in Income and Black-White Gaps in Reading, 1943–2001 Cohorts Trends in Income and Black-White Gaps in Reading, 1943-2001 Cohorts
1980 Cohort Birth Year
1990
2000
Community Investments, Summer 2012 – Volume 24, Number 2
21
Figure 4. Trends in Income and Black-White Gaps in Math, 1943–2001 Cohorts Trends in Income and Black-White Gaps in Math, 1943-2001 Cohorts
1.50 1.25 1.00 0.75 0.50 0.25 0.00
1940 1950
1960 1970 1980 1990 2000 Cohort Birth Year
Study
TALENT
NLS
HS&B
NLSY79
NELS
Prospects
NLSY97
ELS
SECCYD
ECLS-K
HLS
ECLS-B
Income Gap
Black-White Gap
Black-White Gap (NAEP)
22
Community Investments, Summer 2012 – Volume 24, Number 2
dark dashed line describing the trend in the black-white achievement gap during the same time period. For com- parison, a third trend line is included in the figure—the estimated trend in black-white gaps as estimated from NAEP data.
The striking feature of Figures 3 and 4, however, is not so much the well-known trends in the black-white gaps but the difference between the trends in the income gaps and the black-white gaps. For cohorts born in the 1940s to the 1960s, the black-white achievement gap was substantially larger than the 90/10 income achievement gap, particu- larly in reading. For cohorts born in the 1970s and later, however, the opposite is true. Among children born in the last two decades (those cohorts currently in school), the 90/10 income gap at kindergarten entry was two to three times larger than the black-white gap at the same time.
Why Has the Income Achievement Gap Grown?
The evidence thus far indicates that the relationship between a family’s position in the income distribution and their children’s academic achievement has grown sub- stantially stronger during the last half-century. I suggest four possible broad explanations for this trend.
1. Rising Income Inequality
After decades of decline, income inequality in the United States has grown substantially in the last four decades and as of 2007 was at a level similar to the levels in 1925 to 1940, when U.S. income inequality was at its twentieth-century peak.10
Source: Author’s compilation based on data from Project Talent (Flanagan et al. n.d.); NLS, NAEP, HS&B, NELS, ELS, ECLS-K, HLS, ECLS-B (U.S. Department of Education, Center for Education Statistics n.d., 1999, 2000, 2001, 2004, 2005, 2009, 2010); Prospects (U.S. Department of Education 1995); NLSY79, NLSY97 (U.S. Bureau of Labor Statistics
1980, 1999); and SECCYD (National Institute of Child Health and Human Development 2010).
Note: Solid symbols represent
90/10 income achievement gaps; hollow symbols denote black-white achievement gaps. See online appendix section 5.A5 for further details.
Figure 5 shows income inequality trends over time, with changes in the 90/10 family income ratio (the ratio of the family income of the child at the 90th percentile of the family income distribution to that of the child at the 10th percentile), the 90/50 family income ratio, and the 50/10 family income ratio among school-age children from 1967 to 2010.11 What is particularly striking is that the 90/10 family income ratio grew rapidly from 1967 to the early 1990s, more than doubling in twenty-five years, declined modestly during the 1990s and rose sharply over the past decade.
But how might income inequality relate to achieve- ment? In a separate analysis, I investigate whether the children of the rich score higher than the children of the poor because the income difference between the rich and poor is so much larger than it used to be, or because the relationship between achievement and dollars of income has grown stronger.12 In other words, does a dollar buy more achievement than it did before, or do the rich just have more dollars than they did before? These analyses, although not conclusive, suggest that the growth of the income achievement gap is not explained solely by rising income inequality. Rather, the association of achievement with family income has grown stronger over time, particu- larly among families in the upper half of the income dis- tribution. Thus, it is not only rising income inequality per se that has caused the income achievement gap; rather, a dollar of income (or factors correlated with income) appears to buy more academic achievement than it did several decades ago.
Average Difference in Standardized Test Scores (90/10 Income Gap or Black-White Gap)
Figure 5. Income Inequality Trends, 1967–2010, All Families with School-Age Children Income Inequality Trends, 1967-2010, All Families with School-Age Children
Weighted by Number of School-Age Children
Weighted by Number of School-Age Children
2.2 2.0 1.8 1.6 1.4 1.2 1.0 0.8
90/10 Income Ratio 50/10 Income Ratio
90/50 Income Ratio
1970 1980
Year1990
2000 2010
2. Differential Investments in Children’s Cognitive Development
The evidence showing that the returns to income have grown, at least among higher-income families, suggests that families may be changing how they invest in their children’s cognitive development. If so, this may explain some of the rising income achievement gap. Sociologists and historians have argued that parents, particularly those in the middle class, have become increasingly focused on children’s cognitive development during the last fifty years. Researcher Julia Wrigley, for example, examined the types of parenting advice published in popular mag- azines between 1900 and 1985 and found that articles published in the early part of the century were largely written by medical doctors and focused overwhelmingly on medical and nutritional advice.13 A focus on the in- tellectual development of children became much more prominent beginning in 1960s. Although some of this shift was driven by the era’s interest in social inequality and the need for compensatory preschool education for poor chil- dren, Wrigley argues that children’s cognitive develop- ment quickly became a concern of middle-class parents as well, as these parents increasingly saw education as essential for later economic success.
Another factor that may contribute to parents’ increas- ing focus on their children’s cognitive development is the rise of test-based accountability systems in education. Al- though some forms of standardized testing, including IQ tests and the SAT, have been prevalent for much of the twentieth century, standardized achievement testing has become much more common with the rise of the account-
Source: Author’s calculations, based on U.S. Bureau of the Census (King et al. 2010).
Note: Each line shows the trends in the ratio of household incomes at two percentiles of the income distribution. All trends are divided by their value in 1967 in order to put the trends on a common scale.
ability movement. The combination of the increasing im- portance of educational success in determining earnings and the increasing importance of test scores in defining ed- ucational success may have caused parents to focus more on their children’s cognitive development, with higher income parents more able to invest resources in their chil- dren’s education than their lower income counterparts.
3. Changes in the Relationships among Family Income, Family Socioeconomic Characteristics, and Children’s Achievement
Another possible explanation for the rising income achievement gap is that high-income families not only have more income than low-income families, but also have access to a range of other family and social resourc- es. On average, families with higher incomes tend to be those in which the parent(s) are highly educated. This has long been true, though the link between parental educa- tional attainment and family income has grown stronger in recent decades, as the wage returns to educational at- tainment have increased since 1979.14 Because highly educated parents are more able and more likely than less- educated parents to provide resources and opportunities for their children to develop cognitive and academic skills in both the preschool years and the school-age years, chil- dren of parents with college degrees may have higher aca- demic achievement, on average, than children of parents with lower levels of education, all else being equal.
This argument suggests two possible explanations for the rising income achievement gap. First, the trend may result from an increase in the correlation between paren-
Community Investments, Summer 2012 – Volume 24, Number 2
23
Special Focus: Community Development and Education
Income Ratio (Relative to 1967)
24
Community Investments, Summer 2012 – Volume 24, Number 2
tal educational attainment and family income—which would mean that high- and low-income families are in- creasingly differentiated by education levels, leading to larger differences in children’s achievement. Second, the trend may derive from an increase in the achievement returns to parental education, net of income. This would mean that children of highly educated parents benefit more from their parents’ educational attainment than they did in the past.
Another possible reason for increasing correlation between parental education and income is the increas- ing polarization of families. Sara McLanahan argues that trends since 1960 in family structure and composition have led to an increasingly polarized distribution of family contexts for children—mothers with low levels of educa- tion are increasingly likely to be young, unemployed, and single or divorced; mothers with high levels of education are, conversely, increasingly likely to be older, employed, and married.15 As a result, the correlation of parental edu- cation and income among families with children is likely to increase with time.
4. Increased Segregation by Income
A final possible explanation for the rising income achievement gap is the pattern of increasing income seg- regation during the last forty years. Several recent studies have found that residential segregation by income in- creased from 1970 to 2009, partly as a result of rising income inequality and likely partly as a result of low- income housing policy.16 In particular, rising income inequality has led to the increasing segregation of high- income families from middle- and low-income families;
high-income families increasingly live spatially far from the middle class. Because residential patterns are closely linked to school-attendance patterns, the rise of residen- tial income segregation has likely led to a concurrent rise in school segregation by income, though there is little em- pirical evidence on this.17 Because the growth in income segregation has been largely a result of increasing seg- regation of the affluent, this might explain the pattern of the rising association between income and achievement among higher-income families. However, there is little evidence to answer the question of whether rising income segregation has played a role in the increasing income achievement gap.
Conclusion
The forces at work behind the rising income achieve- ment gap are likely complex and interconnected. Cer- tainly more research to understand the causes of these trends is necessary. Equally important, however, is re- search to understand the consequences of these patterns. At the same time that family income has become more predictive of children’s academic achievement, so have educational attainment and cognitive skills become more predictive of adults’ earnings. The combination of these trends creates a feedback mechanism that may decrease intergenerational mobility. As the children of the rich do better in school, and those who do better in school are more likely to become rich, we risk producing an even more unequal and economically polarized society.
Sean Reardon is Professor of Education and (by courtesy) Sociology at Stanford University.

Looking Back and Moving Forward:
Changes in the Affordable Multifamily Mortgage Industry
By Mary Kaiser, California Community Reinvestment Corporation and George Vine, Vine Associates LLC
Much like looking at old photos of yourself, re-reading thought pieces you wrote years ago, especially those in which you made predictions about the future, can be a hum- bling experience. Twelve years ago, we published an article in Community Investments on the topic of afford- able multifamily mortgage risk.1 We were both recent ar- rivals at California Community Reinvestment Corporation (CCRC), a multi-bank multifamily lending consortium, and the economy was thriving—job growth was strong and was driving housing demand, a very different picture from today. We observed strong credit performance of the Low Income Housing Tax Credit (LIHTC) mortgages that CCRC specialized in and concluded that the lessons to be learned were: (1) although LIHTC mortgages will pay like clockwork, do not expect to see strong cash flows, (2) nonprofit sponsors require careful analysis, and (3) the
structures of these complex loans need to be well thought out prior to close because after closing, the lender’s tools are blunt. Now, more than a decade later, there are new lessons to be learned as the affordable housing industry grapples to deal with the significant policy and economic changes that impact our work. In this article, we reflect on the industry’s historical performance and identify how recent lessons can improve our collective ability to meet the affordable housing needs of low- and moderate-in- come communities, despite the many challenges of the current environment.
What Has Changed?
With respect to credit quality—not much. A 2011 study by the Reznick Group revealed that among 16,399 LIHTC properties surveyed, 98 experienced foreclosure through the end of 2010—an aggregate foreclosure rate
Community Investments, Summer 2012 – Volume 24, Number 2
25
Eye on Community Development
26
Community Investments, Summer 2012 – Volume 24, Number 2
of 0.62 percent, measured by property count.2 Approxi- mately 50 percent of the stated foreclosures were reported to have occurred from 2008–2010.3 In CCRC’s 23 years, it has foreclosed on two loans—an aggregate foreclosure rate of 0.49 percent. Its total realized loan losses over the 23 years is less than $1 million, about 0.15 percent of total loan originations (not including tax exempt bonds for which it has realized no losses). Annualized, this is just a few basis points of credit losses per year. It is very difficult to find directly comparable information, but the indus- try tends to outperform other real estate debt classes. For example, the California Bankers Association reports that as of the 3rd quarter of 2011, 1.6 percent of all multifam- ily loans and 2.4 percent of all single family loans held by California-based banks were on non-accrual – indicating that repayment in full was not expected.4
What explains this incredible performance among LIHTC mortgages?5
• Rents are typically at least 10 percent, and often 20 percent to 40 percent, below market. The result is con- tinual low vacancy rates and low marketing expenses.
• The loans are usually funded upon completion of construction or substantial rehabilitation. The result is several years of low repairs and maintenance expenses.
• Themortgageamountperunitislow,usuallynotmuch more than 10 percent of total development cost. The implication is that other capital providers have large investments to protect and the cost of keeping the mort- gage current is small relative to their investments.
• In particular, LIHTC investors (typically Fortune 500 companies) have major incentives to keep projects from defaulting on their mortgages during their first 15 years due to the potential for tax credit recapture.
• Finally, a foreclosure will eliminate or greatly reduce the rent restrictions, allowing a conversion of the project to market, or at least to a much lower level of affordabil- ity, often creating millions of dollars of additional real estate value. Whereas the average loan to value ratio of CCRC’s mortgage portfolio is 67 percent, its loan to value ratio after a foreclosure, assuming a conversion to market, is 48 percent. This implies a 40 percent increase in property value following a conversion.
While credit quality remains strong, the broader land- scape of the affordable housing industry has changed dra- matically. State and local government budget shortfalls hit California particularly hard last year, when the gover- nor eliminated local redevelopment agencies (historical- ly a source of subsidies for most CCRC projects). CCRC escaped immediate damage because it funds loans only after projects are completed, rented, and all other financ-
ing is in place, but CDFIs providing earlier stage financing are up a creek without a paddle. Even though we survived the immediate effects unscathed, we wonder how future deals will be done without this important source of subsidy.
Banks, many of which suffered a near-death experi- ence in the Great Recession and are facing regulatory uncertainty, are less likely to accommodate CDFIs. We have seen an increase in the number of our credit line banks that now look through individual loans in order to underwrite CCRC. This requires us to spend more time providing information and answering questions, and we must allow more time for funding requests and credit line renewals than ever before. Some banks feel the need to impose financial covenants which allow CCRC even less latitude for reacting to the current turmoil.
Other changes are in the wind and we know they will affect the industry, we just don’t know how. Examples are the uncertain futures of the GSEs and the possibility of Federal corporate tax reform and its effect on LIHTCs.
Lessons Learned from the Great Recession
In mid-September 2008, we took a call from a rep- resentative of one of the nation’s largest financial institu- tions, which had agreed to buy $26 million of CCRC’s tax-exempt bonds the following week. CCRC depends on the sale of mortgage loans and bonds to replenish its origi- nation capacity and bond sales are particularly difficult because of the small pool of prospective buyers.
“The sale’s off,” the representative said. “But we have a signed agreement,” we argued. “Haven’t you heard? Lehman filed for bankruptcy and the market fell 500 points this morning. The world has changed...”
As CCRC and its member banks continued to be rocked by the effects of the Great Recession several truths became apparent:
• The industry had changed, and our business model needed to adapt to survive. For 23 years, we offered permanent mortgages to affordable housing developers of LIHTC-financed construction in California, funded by our member banks. In essence, we delivered a single product to a narrow set of customers in a niche industry, in a single geography, funded by one source of funds. Clearly it would be organizational suicide to expect this business model to work for the next 20 years.
• New financial regulation and policy directly impact- ed our ability to do business. With a mortgage line in excess of $350 million on a net asset base of $14 million, CCRC has reached the end of its members’ abilities to provide additional credit in an era of Dodd Frank and Basel III.

• A strong secondary market for community develop- ment loans was not going to materialize any time soon. CCRC’s traditional secondary market purchas- ers reduced their appetites, and our historically largest purchaser closed the door entirely.
In light of these realizations, one of the critical lessons we learned was the importance of engaging in a company- wide strategic planning effort. CCRC surveyed its member banks and engaged its staff in brainstorming sessions to bring forth all of the changes occurring in the environ- ment and to develop speculative judgments on ways the organization could exploit these changes. These “specula- tive judgments” would then be subject to further testing. It developed a financial model of its operations to project the effects of differing business model scenarios on its fi- nancial statements up to four years into the future.
Through this process, CCRC determined that it needs to diversify its customer base, its product offerings and its source of funds. The financial model demonstrated that CCRC had evolved to a business model that neither allowed for growth, nor for the additional investments in knowledge and personnel required to accomplish its di- versification.
A key to CCRC’s past success has been the favorable mortgage credit line provided by member banks since its inception. The line finances 100 percent of CCRC’s mortgage amounts at a rate of the organization’s portfolio yield minus its servicing fee for an indefinite term with no prepayment penalties. Recognizing how favorable that credit line is, CCRC cut its servicing fee to 25 basis points to maximize the yield to the member banks. In so doing CCRC was only able to cover operating costs, but unable to grow its balance sheet. It was in effect provid- ing the credit enhancement offered by its balance sheet to the member banks for free. And until recently, member banks were willing to offer this concessionary financing without looking too hard at CCRC’s financial statements and without any financial covenants.
Now that the world has changed, we believe the path forward requires four steps:
1. Continue developing alternative outlets for affordable housing mortgages. In the last year, CCRC became an approved FNMA affordable multifamily lender, a HUD MAP lender, and completed its first participation trans- action (essentially a sale) with the pension plan of the United Methodist Church (UMC). CCRC continues to attempt whole loan sales which increasingly seem to require GSE-style underwriting. It would require an en- tirely separate article to explain why GSE underwriting of California LIHTC mortgages is akin to mixing oil and water, but we must continue to try.
In light of these realizations, one of the critical lessons we learned was the importance of engaging in a company-wide strategic planning effort.
2. Maintain flexibility to adapt to new regulatory require- ments. CCRC is considering ways of reducing the size of the mortgage credit line and pricing and structuring it more conventionally with shorter and more definite terms to keep commercial banks at the table as their numbers decline and as Dodd Frank and Basel III kick in.
3. Build organizational capacity to operate in a more complex environment. CCRC is investing in the ad- ditional staff and training needed to meet the HUD, FNMA and UMC underwriting and asset management requirements. We must also develop the CFO skills re- quired by a more complex organization and manage the interest rate risk that will come from “conventional- izing” the mortgage credit line.
4. Identify ways to raise additional equity. The required additional investment and most of the alternative mort- gage outlets generate a need for additional equity funds. The staff and training investments are needed years before they result in additional earnings, and that gap must be bridged. And most of the alternative mort- gage outlets demand some credit enhancement (which requires equity funds). Since CCRC cannot raise these funds internally in a reasonable time frame, it is about to undertake a campaign to raise equity-like funds from the corporate social responsibility sections of the member banks and other corporations, and from foun- dations and government agencies.
In regions across California, rents have been increas- ing while rental vacancy rates have been declining,6 sug- gesting that the housing bust did not solve California’s shortage of affordable housing. Over the past 23 years, CCRC has demonstrated its ability to finance affordable rental housing safely and efficiently. The times may have changed, but by taking these steps now, CCRC intends to continue to be part of the solution. It’s our hope that twenty years from now, we can look back on these times and see that the industry rose to the challenge, bringing creativity, adaptability, and passion to meet the affordable housing needs of low- and moderate-income communities across the nation.
Community Investments, Summer 2012 – Volume 24, Number 2
27
Eye on Community Development
23 Years of CCRC
Since 1989, CCRC has provided permanent financing for housing for seniors, families, and individu- als with special needs, for renters with incomes ranging from 20 percent to 60 percent of area median income. This financing has led to the substantial rehabilitation and new construction of more than 25,000 units of affordable housing in rural and urban communities across the state of California.
In addition to enabling these social benefits, CCRC has provided value to its 44 member banks. Their participation in CCRC’s mortgage line and bond program has simultaneously given them CRA credits and good investments – the often unreachable “double bottom line.” The graph below shows the portfolio yield on CCRC’s mortgage loan portfolio compared to the yield on the U.S. 10 year Treasury note that is frequently used as a pricing benchmark for commercial mortgages.
Over the past 10 years, the spread between them has averaged 3.35 percent. For the past five years, until the beginning of this year, CCRC charged a servicing fee of 25 basis points. During that time CCRC’s member banks earned 300 bps over 10 year U.S. Treasury notes (CCRC’s current servicing spread for 2012 only is 40 bps). And since its inception, CCRC’s member banks have not lost a dime on their invest- ment in the mortgage line or the tax exempt bond programs. The $1 million in from-inception loan losses mentioned previously were fully absorbed from CCRC’s resources.
Another benefit to member (as well as some non-member) banks is that CCRC’s forward mortgage com- mitments are a source of repayment for the banks’ construction loan business. Finally, member banks earn CRA services credit by allowing employees to serve on CCRC’s board of directors and loan commit- tee, and by providing the credit review teams that review CCRC’s portfolio annually.
9% 8% 7% 6% 5% 4%
10 yr US Treasury
CCRC Gross Yield
History of CCRC Loan Portfolio and 10 Year US Treasury Note Yields
3% 2% 1% 0%
28
Community Investments, Summer 2012 – Volume 24, Number 2
Mary Kaiser is President of CCRC.George Vine, CFA is the founder of Vine Associates LLC. CCRC is a member of the Association of Reinvestment Consortia for Housing (ARCH). To learn more, visit http://www.frbsf.org/ community/craresources/archlandingpage.html.
Aug‐95 May‐96 Feb‐97 Nov‐97 Aug‐98 May‐99 Feb‐00 Nov‐00 Aug‐01 May‐02 Feb‐03 Nov‐03 Aug‐04 May‐05 Feb‐06 Nov‐06 Aug‐07 May‐08 Feb‐09 Nov‐09 Aug‐10 May‐11 Feb‐12
CDFI Industry Analysis: Summary Report
By Michael Swack,University of New Hampshire, Jack Northrup, New England
Market Research, and Eric Hangen, I2 Community Development Consulting, Inc.
Introduction
Community development financial institutions (CDFIs) fill a market gap by supplying financial products and ser- vices tailored to the needs of underserved communities and are targeted to promote community development. The economic challenges stemming from the recent re- cession have significantly impacted the CDFI industry and have required organizations to adjust their practices and rethink their strategies going forward.
In order to understand the changing landscape of the CDFI industry, the Carsey Institute, under contract to NeighborWorks® America and the U.S. Department of Treasury’s Community Development Financial Institu- tions (CDFI) Fund, conducted a detailed analysis of a large sample of CDFIs on issues of capitalization, liquidity and portfolio, and risk management by CDFIs from 2005 to 2010. This study involved a large sample of CDFIs of all types, including loan funds, credit unions, banks, holding companies, and venture funds within the finance/insur-
ance/real estate industry sector. It is important to note that the analysis is not necessarily representative of all CDFI loan funds;but it is representative of CDFI banks, CDFI credit unions and CDFI bank holding companies, as infor- mation obtained is from all institutions with CDFI certifica- tion. This article is an excerpt from the full report, which is available from the CDFI Fund and the Carsey Institute.1
Primary Findings
Finding 1: CDFIs have been “stepping into the breach” to address lending-related needs during the recession—and have paid a financial price for doing so.
CDFIs are willing to take risks and serve customers with financial products that traditional capital markets are unlikely to provide. CDFIs have expanded their assets and their loan portfolios since the market peak in 2005, as the economic crisis has made it harder to access traditional credit markets.
Community Investments, Summer 2012 – Volume 24, Number 2
29
Eye on Community Development
• Among loan funds in this sample, median assets doubled and loan portfolios increased 76 percent. The median CDFI loan fund deployment ratio grew 3.1 percent annually from 2006 to 2009.
• The median CDFI credit union portfolio grew 47 percent from 2005 to 2010, compared with 29 percent growth for non-CDFI credit unions. Assets grew by 38 percent, compared with 47 percent for non-CDFI credit unions.
• CDFI banks saw median assets grow at an annual- ized rate of 7.9 percent from 2006 to 2010, while the assets of corresponding traditional banks grew at an annualized rate of 0.63 percent. CDFI banks saw their median loan portfolios grow 33 percent over the same period, versus 27 percent for the comparison group.3
At the same time, CDFIs appear to have paid a finan- cial price for their actions during the recession.
• CDFI credit unions experienced declining earnings and rising delinquency rates from 2005 through 2010, and they had higher delinquency rates than the credit union industry as a whole.
• Median net income for CDFI banks, which equaled median net income for corresponding traditional banks in 2006, dropped sharply in the recession. As of 2010, median net income for CDFI banks was 63 percent of that of traditional banks. Net loss to average total loans and leases grew from 0.13 percent in 2005 to 0.88 percent in 2009 before falling back to 0.82 percent in 2010.
Finding 2: CDFI portfolio performance has been mixed, but only for a minority of organizations is it an issue that significantly affects overall financial performance.
The very limited data available on delinquencies and charge-offs for CDFI loan funds are mostly positive. In 2009, CDFI loan funds that were dedicated exclusively to
home financing reported a median portfolio at risk (i.e., 90+ day delinquency) of two percent, up from 0.9 percent in 2008. Similarly, CDFI loan funds engaged solely in business lending had a median charge-off rate in 2009 of 1.3 percent. Loan funds engaged solely in real estate development lending had a median portfolio at risk of 1.6 percent in 2009, up from 1.4 percent in 2008. Charge-offs were two percent in 2009, up from 0.6 percent in 2008.
For a minority of CDFI loan funds, however, loan losses have affected financial sustainability. Among real estate and home financing loan funds in this study, 27 percent of portfolios showed risk greater than seven percent, which is the CDFI Fund’s Minimum Prudent Standard (MPS). Among business loan funds, 26 percent of their portfolios exceeded the MPS of 10 percent at risk. Among all the loan funds in this study, 11 percent reported portfolios at risk greater than the CDFI Fund’s “overall” MPS of 15 percent.
Data on credit unions indicate that CDFI credit unions have been experiencing greater risk in their loan portfo- lios than traditional credit unions. As of 2010, CDFI credit unions had more than double the rate of delinquent loans as a percentage of total assets relaive to the overall credit union industry: 2.9 percent compared with 1.0 percent. Charge-offs to average loans were only slightly higher, at 0.93 percent for CDFIs versus 0.89 percent for non-CDFIs.
Finding 3: Significant scale effects exist in all sectors of the CDFI industry.
The analyses strongly support a finding that CDFIs with larger assets are much more likely to achieve high self-sufficiency ratios than institutions with smaller assets. Among CDFI Loan Funds, larger funds outperform smaller ones along a range of factors that may result in greater self- sufficiency. At the same time, larger loan funds are able to achieve greater self-sufficiency despite operating at lower margins (smaller pricing mark-ups) than smaller funds, as can be seen in Table 1, showing three-year averages.
Table 1. Financial Metrics by CDFI Loan Fund Asset Size
Asset size4
% of applicants
Self- sufficiency ratio
Leverage ratio5
Combined interest / operating expense ratio
Margin6
Mean deployment ratio7
Mean charge-off ratio8
<$500k 10.3 $1M-$5M 23.1 $10M-$50M 25.2 >$100M 13.5
0.107 0.385 0.623 0.848
-0.574 8.16 1.599 1.24 2.538 0.421 8.138 0.079
-1.640 0.23 -0.348 0.68 -0.137 0.82 -0.033 0.86
0.00% 0.52% 0.38% 0.06%
$500k-$1M
8.2
0.232
2.522
14.19
-0.651
0.54
0.00%
$5M-$10M
13.1
0.540
2.258
0.382
-0.210
0.71
0.40%
$50M-$100M
6.8
0.903
3.304
0.264
-0.094
0.92
0.18%
30
Community Investments, Summer 2012 – Volume 24, Number 2

Table 2. Financial metrics by CDFI Credit Union asset size
2010 numbers
<$10 M
$10M- $25M
$10M- $25M
$50M- $75M
$75M- $100M
$100M- $200M
$200M- $400M
>$400M
Loan interest Cost of funds
Non-interest income
Net income
8.46% 7.50% 1.55% 1.64%
3.84% 3.11%
-1.25% -0.37%
7.70% 1.67%
3.12%
0.49%
6.73% 7.02% 7.14% 6.16% 1.74% 1.48% 1.91% 1.76%
3.08% 2.82% 3.46% 2.25%
0.23% 0.85% 0.87% 0.14%
5.77% 2.36%
1.63%
1.07%
Gross yield
9.05%
8.21%
8.28%
7.30%
7.64%
7.82%
6.81%
6.31%
Net yield with provision
5.12%
5.12%
5.07%
4.33%
5.25%
4.84%
3.72%
2.82%
Non-interest expense
10.21%
8.59%
7.70%
7.17%
7.22%
7.42%
5.84%
3.38%
Similarly, among CDFI credit unions, larger credit unions have stronger net income performance while charging lower interest rates and fees on their loans, in large part by keeping non-interest expenses low (see Table 2). Economies of scale are also found in the CDFI banking sector, although these scale effects are more pronounced in traditional banks.
Finding 4: Operating expenses play the driving role in determining whether CDFIs achieve self-sufficiency.
As a cost driver for CDFI loan funds, operating expense is by far the largest component of an organi- zation’s expenses, dwarfing both cost of capital and loan loss expense, thus representing a key determinant of organizational sustainability. For 21 of the 34 loan funds studied, operating expenses make up more than 70 percent of total expenses. For only three of the loan funds studied do operating expenses make up less than 50 percent of total expenses, and two of these three funds report that an affiliate performs some operating functions for them at no charge.
Indeed, as alluded to in Finding 3, a major reason why larger CDFI loan funds may be more likely to have high self-sufficiency ratios is that they have drastically lower levels of operating expense per dollar of assets managed. Given the results obtained from the “deep dive” analysis, it is safe to assume that operating expense is the main com- ponent of the combined interest and operating expense ratio that was calculated for all loan funds. This ratio is significantly lower for large loan funds. There is some evi- dence that organizations with smaller operating expense ratios may have less intensive development services or may receive development services or other services from an affiliated organization, thus reducing their expenses.
Even among CDFI credit unions and banks, there is a similar dynamic, in which operating expense is consis-
tently a much more powerful driver of profitability than loan performance or cost of capital. For example, among the largest CDFI banks ($1 billion to $3 billion in assets), non-interest expense runs at 3.14 percent of assets. This compares with interest expense at 2.12 percent and loan and lease losses at 0.98 percent. This dynamic is as strong or stronger among the smallest CDFI banks (under $100 million in assets), where non-interest expense is on average 3.65 percent of assets and interest expense is only 2.2 percent, and loan and lease loss provisions count for 1.12 percent of assets.
The factors driving CDFI operating expenses are clearly complex, but the bottom line is that more efficient delivery mechanisms may be critical for CDFIs’ survival. These mechanisms could include greater use of technol- ogy, more collaboration between organizations, and ex- panding overall assets so that fixed expenses are spread over a much larger asset base. Perhaps a larger challenge for the field is that portfolio performance is directly tied to providing the very same services that are driving up the operating costs. The challenge therefore resides not simply in improving efficiency, but may be a core component of the basic business model.
Finding 5: CDFIs, particularly CDFI loan funds, face numerous barriers preventing them from using and leveraging capital more effectively.
CDFI loan funds are generally not well leveraged, possibly reflecting the cost of debt available to them.
Particularly among loan funds, a large number of CDFIs have very little leverage (i.e., they fund themselves mainly through net assets, not debt). The median CDFI loan fund in 2009 was leveraged at just $1.10 in liabilities for every $1 in net assets. About eight percent of loan funds had no liabilities whatsoever. Banks and credit unions are typi- cally leveraged at a rate of 10:1 or more.
Community Investments, Summer 2012 – Volume 24, Number 2
31
Eye on Community Development
One reason why CDFI loan funds use little leverage may be that their “equity” (net assets) is free, whereas their cost of debt can be surprisingly high. The 31 loan funds se- lected for deeper analysis that reported having debt (notes payable and lines of credit) on their audited balance sheets, had a median cost of debt (interest expense/debt) of 2.7 percent. This compares to banks, which may have an overall cost of funds of less than one percent.
CDFI loan funds struggle to perform the asset transfor- mation function and thus may need more help to meet market needs for longer-term financing.
Another issue affecting loan fund leverage levels is that generally, loan funds do not appear to have access to long-term debt. Of the 34 loan funds studied in the deep dive, only four had a term of 10 or more years remaining on most of their debt. By comparison, 17 loan funds had less than five percent of their debt with 10 years or more remaining on it, and three loan funds had no debt at all.
On-balance-sheet CDFI loan products appear largely oriented toward shorter-term products, particularly for business loan funds, real estate loan funds, and multi-line loan funds. Longer-term products appear largely to be either sold to secondary market players or are funded by net assets. For home financing CDFIs in particular, the col- lapse of Neighborhood Housing Services of America has made the secondary market route more difficult.
What appear to be absent from the CDFI loan fund business model are strategies by which the organization funds longer-term assets using shorter-term debt. Unlike banks and credit unions, many CDFIs have no role in asset transformation. Only 17 percent of CDFI loan fund survey respondents said they borrow short and lend long. The study results suggest, albeit not conclusively, that some mechanisms may be needed to help CDFI loan funds originate longer-term loan products, whether by enabling these CDFIs to borrow long-term debt, or by helping them hedge the asset-liability management risk stemming from borrowing short and lending long.
Potential exists to more effectively use large amounts of undeployed capital in the industry.
Of the 282 CDFI loan funds studied, the 112 organiza- tions that were leveraged at less than $1 of debt per $1 of net assets had over $350 million in aggregate cash. About $53 million of this cash was held by loan funds with less than $10 million in assets, and $297 million held by loan funds with more than $10 million in assets. Given that there are about twice as many CDFI loan funds (572) than the 282 in this study, there might be over $700 million in cash at under-leveraged loan funds across the entire sector. The availability of this cash raises the question of whether inter-CDFI transactions could somehow be fa- cilitated to improve liquidity for those CDFIs that need it, while providing a better return for the investing CDFIs than they receive at the bank.
Inadequate data and non-standardized auditing practices may present a barrier to CDFI capitalization.
In developing this report, the research team encoun- tered significant data limitations at every turn. These limi- tations are substantial enough to be a significant barrier to CDFI capitalization, especially for CDFI loan funds, but also, to some degree, for other types of CDFIs. The limita- tions include:
1. Very little product-specific portfolio performance in- formation is available for loan funds.
2. Loan level data are not available for the CDFI industry, short of compiling and harmonizing datasets from in- dividual organizations.
3. Standards and formats for audited financials vary.
4. Uniformity in underlying business models is lacking, so a given financial ratio cannot be compared across organizations.
Policy Recommendations
Policy Recommendation 1: Create Networks, Build Infrastructure, Attract Resources and Build Scale
For community development, scale means: (1) Provid- ing services to a large number of low-income people; (2) Providing services to a significant percentage of those in need; (3) Being able to leverage size to improve results; (4) Having enough capital to develop new products and ser- vices; (5) Getting beyond year-to-year funding concerns; (6) Capturing enough market share to influence for-prof- it providers; and (7) Being significant enough to have a voice with legislators and regulators.
Developing models for scale in the community finance sector can create an antidote to inefficiency, strengthen small organizations, and develop the blueprint that will promote thriving models of community development finance in urban and rural areas while maintaining the mission objectives of CDFIs.
Policy Recommendation 2: Promote the Availability of Longer Term Capital
The availability of long-term debt and equity capital for CDFIs, particularly loan funds, is one of the major struc- tural issues facing the industry. The lack of long-term debt financing forces CDFIs to “hoard cash,” pushing down le- verage and giving the appearance that many underlever- aged CDFIs are not lending as much as they could, thus neglecting demand among its targeted consumers. It is not a reluctance to borrow that pushes leverage down, it is the lack of long-term debt and equity or near-equity funding that is undermining the capital structure of many CDFIs.9
In addition, the lack of long-term capital distorts the CDFIs’ product suite by default. Demand for longer-term consumer debt products is either not being met at all, or is being met by providing mismatches of assets and li-
32
Community Investments, Summer 2012 – Volume 24, Number 2

abilities. Many CDFIs simply do not lend long, and the demand for long-term debt is either ignored or fit into the available product mix, which typically is a shorter-term debt product.
The CDFI Bond Guarantee Program, which will be able to offer long-term, fixed-rate debt financing, at terms just slightly above comparable Treasury securities may help address the issue of access to long-term, fixed-rate debt. Another possible source of this type of capital will be collaborations among CDFIs.
Policy Recommendation 3: Promote Streamlined Access to Industry Data
Consistent with policies that promote scale creation, is a policy that promotes the availability of transparent industry data from which managers can make informed decisions. Data are available for banks and credit unions, but not for loan funds or venture funds. Why not require applicants to the CDFI Fund or recipients of CDFI funding to provide uniform, consistent and accurate financial and performance data on their portfolio and operations? Bank and credit union quarterly reports can be provided using Financial Performance Reports (FPR) and Uniform Bank Performance Report (UBPR) data and call reports. Yet in- formation for 60 percent of the industry (CDFI loan funds) is not available. Any understanding of the industry, and therefore any sensible planning, is severely handicapped by this lack of data.
In place of some of the current documentation re- quired by the CDFI Fund, the Fund could consider cre- ating a standardized quarterly report, similar to the call reports submitted by banks and credit unions, and require all CDFIs to submit them (or at least all CDFIs over a certain asset amount.) The Fund could make these reports public (like the Federal Deposit Insurance Corporation and the National Credit Union Administration do), which would be a great service to the industry. A quarterly call report that includes the impact data now required in the Fund’s Institutional Level Report (ILR), would collect data more efficiently and would create standardized data from a universal data pool year after year. That report would ac- curately represent the industry and would provide mean- ingful data for research purposes. In addition, the CDFI Fund might consider assembling a group of CDFIs to meet with the Financial Accounting Standards Board (FASB) to establish a common set of industry reporting standards.
Policy Recommendation 4: Promote and Document Innovation
Every CDFI is slightly different, no matter what the in- stitutional type. High performers have similar character- istics and operations. Many CDFIs are mission-bending, throwing out the capital net year after year, often linking programs and products to services. But it is often difficult to determine whether new programs are the result of in-
novation, or of copying other programs, or the result of “writing to the grant.”
There are major, if unintended consequences for having no knowledge bank or other online resource for systematically cataloging or analyzing best practices. These information gaps stifle innovation and cause rep- lication of ineffective approaches to capital deployment. Adequate data collection and performance metrics may diminish this consequence, but an institutional approach to promoting innovation, documenting the innovation and disseminating the results is critical in reducing overall in- efficiencies within the field.
Policy Recommendation 5: Promote Education and Training
CDFIs need ongoing education and training on fa- miliar issues: market definition, asset design, cash flow management, standardization of documentation, portfolio analysis, interest rate spreads, etc. Some need basic help with loan policies and procedures while many others need capitalization assistance and definition of that assistance.
Conclusion
The analysis suggests that the CDFI “story” is largely accurate. That story is that CDFIs are institutions that have learned to effectively manage the “risk” that discourages conventional financial institutions from serving low- and moderate-income individuals and communities. The data analysis suggests that CDFIs have succeeded in lending to and investing in individuals and communities not served by conventional financial institutions, while maintaining loan performance standards generally equivalent to those of the conventional financial sector. However, it is also true that the costs of serving these individuals and com- munities is somewhat higher because good performance is, in part, due to the additional technical and training ser- vices provided by most CDFIs. But some additional costs incurred by CDFIs could be mitigated if CDFIs, as a group, undertook certain changes in their operating procedures. Support for building CDFI “infrastructure,” as described in this report could enhance the efficiency, productivity and impact of CDFIs. This report also suggests the need for additional research to address some of the ongoing issues faced by CDFIs including, but not limited to access to long-term capital, creating capacity for transformational activities, understanding of market failure/inefficiencies, and analysis of workforce development and retention issues for CDFIs.
Michael Swack is Faculty Director, Center on Social Inno- vation and Finance, the Carsey Institute, University of New Hampshire. Jack Northrup is President of New England Market Research, Inc. Eric Hangen, AICP, is President of I2 Community Development Consulting.
Community Investments, Summer 2012 – Volume 24, Number 2
33
Eye on Community Development
RESEARCH BRIEFS
PTublic Housing Transformation and Crime
he process of demolishing distressed public housing and relocating families using housing vouchers can have a wide range of effects at both the individual
level (e.g., education outcomes and employment oppor- tunities) and the neighborhood and regional level (e.g., property values and poverty concentration). A recent report from the Urban Institute explores the relationship between this kind of public housing transformation and crime rates at the neighborhood and city level in Chicago and Atlanta. Using data from HUD, Census, and local police departments, the researchers measured the effect of resident relocation on crime in destination and demoli- tion neighborhoods.
Their analysis indicates that public housing transfor- mation and resident relocation reduced crime citywide in both Chicago and Atlanta. In general, crime decreased in neighborhoods where public housing was demolished and in many neighborhoods where former public housing residents relocated. But in a small number of neighbor- hoods that received a relatively larger number of relocat- ed families, crime decreased less than it would have if no former public housing residents relocated there. Overall, neighborhoods with a modest or high density of relocated residents experienced crime rates that were higher than those of areas without relocated residents.
The authors conclude that public housing transfor- mation requires large-scale comprehensive relocation strategies in order to mitigate the potential challenges of transformation and relocation. They recommend that housing authorities provide intensive support for relo- cated residents in a wide range of communities and that local policy discourage the reconcentration of poverty in other vulnerable neighborhoods.
Popkin, S., Rich, M., Hendey, L., Hayes, C. and Parilla, J. (2012), Public Housing Transformation and Crime, Making the Case for Responsible Relocation. Urban Institute. http://urban.org/UploadedPDF/412523-public- housing-transformation.pdf
Metropolitan Fragmentation and Health Disparities
tinct municipal districts, special service districts, T
he term “metropolitan fragmentation” refers to the division of a metropolitan region into separate, dis-
and school districts. Proponents of metropolitan frag- mentation argue that by dividing big metropolitan scale governments into smaller units, citizens will have greater access to effect change in their communities. Critics of fragmentation contend that the phenomenon creates ex- clusive special service districts that exacerbate fiscal ineq- uities by siphoning resources away from low- and moder- ate-income communities.
A recent study investigates one facet of this issue: the possible relationship between metropolitan fragmentation and racial health disparities, as measured by mortality rates for blacks and whites. The authors use data from the U.S. Census of Government on the country’s largest 171 metropolitan statistical areas to count the total number of governments within each metropolitan area for the year 1997, and measure mortality rates using county data from the Centers for Disease Control and Prevention. The study finds a relationship between increased metropoli- tan fragmentation and greater disparities in mortality rates between blacks and whites. Specifically, increasing frag- mentation is associated with a higher mortality rate for blacks but not for whites. The authors indicate a need for further research to explore the interrelated forces behind metropolitan fragmentation, racial segregation, racism, and poverty. For future research, the authors propose in- vestigating the governmental and institutional channels through which metropolitan fragmentation contributes to the differences between black and white mortality rates. For practitioners, they suggest that increased collaboration between the fields of urban planning and public health could help to mitigate health disparities.
Hutson, M., Kaplan, G. A., Ranjit, N. and Mujahid, M. S. (2012), Metropolitan Fragmentation and Health Disparities: Is There a Link?. Milbank Quarterly, 90: 187–207.
34
Community Investments, Summer 2012 – Volume 24, Number 2

Survey Reveals Changes in Family Finances
In June 2012, The Federal Reserve Board of Governors released key findings from its Survey of Consumer Fi- nances, a report released every three years that tracks changes in the financial conditions of U.S. families. The survey reveals that median family income before taxes fell almost eight percent from 2007 to 2010. The decline in median income was widespread across demographic groups, with only a few groups reporting stable or rising incomes. Families living in the South and West regions experienced some of the greatest declines in median incomes.
Net worth declined by a greater percentage than income, with median net worth falling by almost 40 percent. Median net worth declined for families through- out the country, but most dramatically in the West, where median net worth fell by about 55 percent. This pattern reflects the collapse of housing markets in several regions in the West.
In addition to geography, the magnitude of net worth loss was also impacted by a family’s relative level of net worth. For example, the median net worth for a family in the lowest quartile fell 100 percent, from $1,300 to zero, while median net worth for a family in the second quartile fell by about 43 percent and by 11 percent for a family in the top decile.
The collapse of the housing market also explains differences in net worth declines for homeowners and renters. Between 2007 and 2010, the median net worth for homeowners fell by about 30 percent. Comparatively, the median net worth for renters fell by 5.6 percent.
From 2007 to 2010, financial assets rose as a share of families’ total assets, which was driven by the decline in house prices. At the same time, the homeownership rate, which had increased between the 2001 and 2004 surveys, continued to fall – roughly to the same level as in 2001.
Saving rates also changed between 2007 and 2010, with the share of families reporting saving over the previ- ous year falling by about four percentage points. Families’ reasons for saving also changed. Fewer families said they were saving for retirement, education, or buying their own home. More families reported that their reason for saving was for liquidity to ensure they had enough cash to cover unexpected expenses. At the same time, the percentage of families using credit cards for borrowing dropped and the median balance on consumer credit card accounts fell 16 percent. The percent of families borrowing for educa- tion-related expenses increase from 15 to 19 percent, and the median balance of education-related debt increased about 3.5 percent (mean balance rose 14 percent). Finally, the percent of debtors with any payment 60 days or more past due increased from about 7 percent to almost 11 percent in 2010.
Bricker, J., Kennickell, A.B., Moore, K.B., and Sabelhaus, J. “Changes in U.S. Family Finances from 2007 to 2010: Evidence from the Survey of Consumer Finances.” Federal Reserve Bulletin (June 2012): Vol 98, No 2. http:// www.federalreserve.gov/pubs/bulletin/2012/pdf/scf12.pdf
Community Investments, Summer 2012 – Volume 24, Number 2
35
Quarterly Features
DOCTOR CRA
36
Community Investments, Summer 2012 – Volume 24, Number 2
Dear Dr. CRA,
Like many other banks, we have a large inventory of residential other real estate owned (OREO) properties. I’ve heard that rental demand is increasing in my market – vacancies are down and rents keep going up. Could we rent out our OREO properties as part of our disposition strategy, and if so, would we get CRA credit?
Dear Double Stuffed,
The Federal Reserve released a policy statement earlier thisyearconfirmingthatitpermitstherentalofresiden- tial OREO properties as part of an orderly disposition strategy. Banking organizations may rent OREOs (within statutory and regulatory holding-period limits) without having to demonstrate continuous active marketing of the property. As long as certain policies and procedures are followed, banks would not violate supervisory ex- pectations that they show “good-faith efforts” to dispose of OREO by renting the property within an appropriate time frame. However, in order to receive favorable CRA consideration, the OREO rental properties must meet the definition of community development under the CRA regulations (which includes activities that provide affordable housing for low- and moderate-income indi- viduals as well those that revitalize or stabilize areas, see 12 CFR 228.12(g)(1) and (4)).
You can learn more about OREO rentals through the Fed’s April 5, 2012 policy statement, available online at http://www.federalreserve.gov/newsevents/press/bcreg/ bcreg20120405a1.pdf. The statement provides guid- ance to banking organizations and examiners and also describes specific supervisory expectations for banking
Sincerely,
Double Stuffed with OREOs
organizations with a larger number of rental OREO properties (generally more than 50 properties). As always, if you have specific questions, don’t hesitate to contact your examiner. Finally, it should be noted that banking organizations’ residential property rental activ- ities are expected to comply with all applicable federal, state, and local laws and regulations, including:
• Landlord-tenant laws;
• Landlord licensing or registration requirements;
• Property maintenance standards;
• Eviction protections (such as under the Protecting Tenants at Foreclosure Act);
• Protections under the Servicemembers Civil Relief Act; and
• Anti-discrimination laws, including the applicable provisions of the Fair Housing Act and the Ameri- cans with Disabilities Act.
This is important as a pattern or practice of violations of certain laws can have an adverse impact on the bank’s CRA rating.

DATA SNAPSHOT
Education Intergenerational Mobility by College Education
Intergenerational Mobility by College Education
Dropout Rates by Income Level, 2009
A college degree can buffer downward mobility
Dropout Rates by Income Level, 2009
A college degree can buffer downward mobility
20%
16% 15%
10% 5% 0%
Lowest Middle Low
Middle High
Highest
10%
Income Quartile
5%
2%
60%
40%
20%
Without a college degree
45%
With a college degree
54%
16%
23%
0%
Source: Haskins, Ron, Julia B. Isaacs & Isabel V. Sawhill. 2008. Getting Ahead or Losing Ground: Economic Mobility in America. Washington, D.C.: Pew
Started and ended in the bottom quintile
Started and ended in the top quintile
Charitable Trusts, Economic Mobility Project.
Source: National Center for Education Statistics.
Source: Haskins, Ron, Julia B. Isaacs & Isabel V. Sawhill. 2008. Getting Ahead or Losing Ground: Economic Mobility in America. Washington, D.C.: Pew Charitable Trusts, Economic Mobility Project.
State Funding for Public Education
Source: National Center for Education Statistics.
Percentage change from 2008 - 2009
State Funding for Public Education
Percentage change from 2008 - 2009
-14.6%
-8.9%
-7.2% -5.9%
-4.3% -2.6%
-1.7%
Hawaii Washington Idaho
Arizona
Nevada
California Utah
Alaska Oregon United States
2.4% 3.1%
3.6%
Source: U.S. Census Bureau 2008 and 2009 Annual Surveys of Local Government Finances - School Systems.
Source: U.S. Census Bureau 2008 and 2009 Annual Surveys of Local Government Finances - School Systems.
Community Investments, Summer 2012 – Volume 24, Number 2
37
Quarterly Features
38
Community Investments, Summer 2012 – Volume 24, Number 2
Endnotes
Community Development and Education: A Shared Future
1. Card, D. (1999). The causal effect of education on earnings, In: Ashenfelter, O. & Card, D. Editor(s), Handbook of Labor Economics, Elsevier, 1999, Volume 3, Part A, Pages 1801-1863; Bureau of Labor Statistics. (2010). Education pays: More education leads to higher earnings, lower unemploy- ment. Occupational Outlook Quarterly, Summer 2010; Lochner, L. & Moretti, E. (2004). The Effect Of Education On Crime: Evidence From Prison Inmates, Arrests, And Self-Reports. American Economic Review, 94(1): 155-189; Robert Wood Johnson Foundation (2011). Education Matters for Health. Issue Brief Series: Exploring the Social Determinants of Health, Education and Health – April 2011; Dee, T. (2004) Are there civic returns to education?, Journal of Public Economics, 88 (9–10): 1697-1720.
2. Duncan, A. (2012). Fighting the Wrong Education Battles. Remarks of Secretary Duncan at the Askwith Forum, Harvard Graduate School of Educa- tion, February 7, 2012. http://www.ed.gov/news/speeches/fighting-wrong- education-battles
3. Donovan, A. (2008). Charter School Facilities Finance: How CDFIs Created the Market, and How to Stimulate Future Growth. Federal Reserve Bank of San Francisco Community Development Working Paper Series, 2008-02. http://www.frbsf.org/publications/community/wpapers/2008/wp08-02.pdf
4. Coleman, J. S., et al. (1966). Equality of Educational Opportunity. Washing- ton, D.C.: U.S. Government Printing Office.
5. Aud, S., et al. (2012). The Condition of Education 2012. National Center for Education Statistics, U.S. Department of Education.
6. Swanson, C. (2004). Who Graduates? Who Doesn’t? A Statistical Portrait of Public High School Graduation, Class of 2001. The Urban Institute, Educa- tion Policy Center. http://www.urban.org/UploadedPDF/410934_WhoGradu- ates.pdf
7. Hart, B. & Risely, T. (2003). The Early Catastrophe: The 30 Million Word Gap by Age 3. American Educator, 27(1): 4-9; Evans, M.D.R., Kelley, J., Sikora, J., & Treiman, D. (2010). Family scholarly culture and educational success: Books and schooling in 27 nations. Research in Social Stratification and Mobility, 28(2), 171–197.
8. Cytron, N. (2011). School Quality and Affordable Housing in the Bay Area. Community Development Research Brief, Federal Reserve Bank of San Francisco. http://www.frbsf.org/publications/community/research-briefs/ school-quality-and-affordable-housing-in-bay-area.html
9. Leventhal, T. & Brooks-Gunn, J. (2000). The neighborhoods they live in: The effects of neighborhood residence on child and adolescent outcomes. Psychological Bulletin, 126(2): 309-337.
10. Reardon, S. & Bischoff, K. (2011). Growth in the residential segregation of families by income, 1970-2009. Retrieved from the US 2010 Project, http:// www.s4.brown.edu/us2010/Data/Report/report111111.pdf
11. Ibid.
12. Ibid.
13. Harlem Children’s Zone, “About Us: The HCZ Project,” available at www.hcz. org/about-us/the-hcz-project.
14. U.S. Department of Education. (2010). Promise Neighborhoods Program Description. Retrieved from http://www2.ed.gov/programs/promiseneighbor- hoods/index.html
15. U.S. Department of Housing and Urban Development. Choice Neighbor- hoods. Retrieved from http://portal.hud.gov/hudportal/HUD?src=/program_ offices/public_indian_housing/programs/ph/cn
16. Kania, J. & Kramer, M. (2011). Collective Impact. Stanford Social Innova- tion Review. Winter 2011. http://www.ssireview.org/articles/entry/collec- tive_impact
17. Ibid.
18. Ibid.
19. 21st Century School Fund and Center for Cities & Schools. (2010). Joint Use of Public Schools: A Framework for a New Social Contract. Washington, DC: 21st Century School Fund.
20. Coalition for Community Schools. Community Schools: Partnerships for Excellence. Retrieved from http://www.communityschools.org/assets/1/Page/ partnershipsforexcellence.pdf
21. Chung, C. (2005). Connecting Public Schools to Community Development. Communities & Banking, Federal Reserve Bank of Boston. Winter, 2005.
The New Civic Infrastructure: The ‘How To’ of Collective Impact
1. Duncan, A. (2012). Remarks at the National Conference on Volunteerism and Community Service. Available at http://www.livestream.com/pointsoflight/ video?clipId=pla_85d22493-b567-4e2e-a3e0-17447fb2f197
2. The Strive Partnership Community Report Card: http://www.strivetogether. org/wp-content/uploads/2010/08/2011-Strive-Partnership-Report2.pdf
3. Kania, J. & Kramer, M. (2011). Collective Impact. Stanford Social Innovation Review. Winter 2011. Available at: http://www.ssireview.org/articles/entry/col- lective_impact
4. The Regional Institute: University of Buffalo, The State University of New York. (2008). What is Civic Infrastructure? Knowledge Now. March 2008. Available at: http://rkn.buffalo.edu/data/KnowledgeNow/CivicInfrastructure.pdf
5. Collins, J. (2005). Good to Great and the Social Sectors: A Monograph to Accompany Good to Great. New York: Harper Collins.
6. Information on the Strive Network Progress Assessment Tool can be found at: http://strivenetwork.org/strive-approach/progress-assessment-tool
7. The White House. (2102). Pay for Success: Investing in What Works. Available at: http://payforsuccess.org/sites/default/files/pay_for_success_ report_2012.pdf
8. Duncan, A. (2010). The New Normal: Doing More with Less. Remarks at the American Enterprise Institute. Nov 2010. Available at: http://www.ed.gov/ news/speeches/new-normal-doing-more-less-secretary-arne-duncans- remarks-american-enterprise-institut
The Widening Academic Achievement Gap between the Rich and the Poor
1. This article is a condensed and updated version of: Reardon, Sean F. “The Widening Academic Achievement Gap Between the Rich and the Poor:
New Evidence and Possible Explanations.” In Whither Opportunity? Rising Inequality, Schools, and Children’s Life Chances, edited by Greg J. Duncan and Richard Murnane. © 2011 Russell Sage Foundation, 112 East 64th Street, New York, NY 10065. Reprinted with Permission. https://www.russellsage. org/publications/whither-opportunity Online appendix is available at: http:// www.russellsage.org/duncan_murnane_online_appendix.pdf
2. Coleman, J. S., Campbell, E., Hobson, C., McPartland, J., Mood, A., Wein- feld, F. & York, R. (1966). Equality of Educational Opportunity. Washington, D.C.: U.S. Department of Health, Education, and Welfare, Office of Education.
3. I use data from 19 nationally representative studies, including studies conducted by the National Center for Education Statistics (NCES), the Long-Term Trend and Main National Assessment of Educational Progress (NAEP) studies, U.S. components of international studies, and other studies with information on both family background and standardized-test scores. Although these studies vary in a number of ways, each of them provides data on the math or reading skills, or both, of nationally representative samples of students, together with some data on students’ family socioeconomic charac- teristics, such as family income, parental education, and parental occupation. Although the specific tests of reading and math skills used differ among the studies, they are similar enough to allow broad conclusions about the rough magnitude of achievement gaps.

4. My calculations, based on 2009 Current Population Survey data. See online appendix section 5.A3 (see Note 1).
5. Figures 1 and 2 display estimated 90/10 income achievement gaps from all available nationally representative studies that include reading- or math- achievement test scores for school-age children and family income. For most of the longitudinal studies (HS&B, NELS, Prospects, ELS, and ECLS-K), only estimates from the initial wave of the study are included. ECLS-B estimates come from wave 4, when children were five years old and tested on school readiness; SECCYD come from wave 5, when children were in third grade and were first administered a broad academic achievement test. The quartic fitted regression line is weighted by the inverse of the sampling variance of each estimate. Included studies are Project Talent, NLS, HS&B, NLSY79, NELS, Add Health (reading only), Prospects, NLSY97, ELS, SECCYD, ECLS-K, HLS, and ECLS-B. Family income is student-reported in Project Talent, NLS, and HS&B. See online appendix for details on computation of 90/10 gaps (see Note 1).
6. See online appendix 5.A4 (see Note 1).
7. My calculations, based on Main NAEP math and reading scores. See National Center for Education Statistics website, available at: http://nces.ed.gov/nation- sreportcard/naepdata/dataset.aspx (accessed March 7, 2011).
8. Grissmer, D. W., Flanagan, A. & Williamson, S. (1998). “Why Did the Black- White Score Gap Narrow in the 1970s and 1980s?” In The Black-White Test Score Gap, edited by C. Jencks and M. Phillips. Washington, D.C.: Brook- ings Institution Press; Hedges, L. & Nowell, A. (1998). “Black-White Test Score Convergence Since 1965.” In The Black-White Test Score Gap, edited
by C. Jencks and M. Phillips. Washington, D.C.: Brookings Institution Press; Magnuson, K. & Waldfogel, J. eds. (2008). Steady Gains and Stalled Progress: Inequality and the Black-White Test Score Gap. New York: Russell Sage Foundation; Neal, D. A. (2006). “Why Has Black-White Skill Convergence Stopped?” In Handbook of the Economics of Education, edited by Hanushek, E. A. & Welch, F. Vol. 1. New York: Elsevier.
http://www.s4.brown.edu/us2010/Data/Report/report111111.pdf. Watson, T. (2009). “Inequality and the Measurement of Residential Segregation by Income.” Review of Income and Wealth 55(3): 820–44.
17. Because of the relatively small within-school samples in many of the studies that include measures of family income, it is difficult to assess the trends in school income segregation using the data available.
Looking Back and Moving Forward: Changes in the Affordable Multifamily Mortgage Industry
1. Kaiser, M. & Vine, G. (2000). Affordable Multifamily Mortgage Risk: One Lending Consortium’s 10 Year History. Community Investments, May 2000. http://www.frbsf.org/publications/community/investments/cra00-1/May- CINewsAffordable.pdf
2. Reznick Group. (2011). The Low-Income Housing Tax Credit Program at Year 25: A Current Look at Its Performance. Retrieved from http://www. reznickgroup.com/sites/reznickgroup.com/files/papers/reznickgroup_lihtc_ survey_2011.pdf We hasten to add that both the Reznick and CCRC data may be favorably biased, in the case of the Reznick data because of survi- vorship bias as discussed in the article and in the case of CCRC because CCRC sold over $500 million of its mortgages and doesn’t formally track
its sold loans. We did check with CCRC’s three major secondary market mortgage purchasers and they confirmed that they had not foreclosed on any CCRC-originated loans.
3. Ibid.
4. Thornberg, C. & Levine, J. (2012). March 2012 Banking and Economic Update, California Bankers Association.
5. Vine, G. The Breakeven Point. Vine Associates LLC, Vol. 1, No. 1.
6. According to the Marcus & Millichap Apartment Research Market Reports, 3rd quarter 2012.
CDFI Industry Analysis: Summary Report
1. This article is an excerpt from the report “CDFI Industry Analysis: Summary Report,” funded by the CDFI Fund, under Contract TPD-CDF- 10-C-0003, Task Order 0002 and 0003. The curriculum and opinions expressed in these documents are those of the authors, who are solely responsible for the content, and do not reflect the opinions of the CDFI Fund or any other person, entity, or organization. The full report can
be accessed at http://www.cdfifund.gov/docs/CBI/2012/Carsey%20 Report%20PR%20042512.pdf or http://www.carseyinstitute.unh.edu/ publications/Report-Swack-CDFI-Industry-Analysis.pdf
2. Although 282 CDFI Loan Funds were sampled, the outstanding question is: are the CDFI Loan Funds examined (as a result of their applying for 2010 funding to the CDFI Fund) different than those that did not apply? If one assumes that they are no different, then the results presented are representa- tive of all CDFI Loan Funds, within the confidence levels and error margins discussed below. If, in fact, they are different, then the results may be repre- sentative of all CDFI Loan Funds. For CDFI Banks, CDFI Holding Companies and CDFI Credit Unions, a census was performed; in other words the data represents all of these CDFI institutions.
3. Median loans and lease value.
4. In this table, each year’s number is averaged, so there is one number per organization.The median number is taken. The N Value for number is taken. The N Value for all CDFI loan funds is 282.
5. Leverage ratio= total notes payable/net assets.
6. Margin ratio = loan yield ratio minus charge-off ratio – combined interest and operating expense ratio.
7. This number is the average of each year’s median deployment ratio.
8. This number is the average of each year’s median charge-off ratio.
9. For a full discussion of this issue see: Tansey, C., Swack, M., Tansey, M., & Stein, V. (2010). Capital Markets, CDFIs and Organizational Credit Risk. The Carsey Institute. Available at http://www.carseyinstitute.unh.edu/docs/ Swack_CapitalMarkets.pdf.
9. Figures 3 and 4 show estimated 90/10 income gaps (solid symbols) and
estimated black-white gaps (hollow symbols) based on the thriteen studies ent
with family income data. The estimated trends in the income and black-white gaps are fitted lines (quartic for income gaps, quadratic for black-white
gaps), weighted by the inverse of the sampling variance of each estimate. The estimated black-white gap trend from NAEP is a fitted line (quartic for reading, cubic for math) through all available NAEP-LTT and Main NAEP black-white gap estimates. The NAEP trend is adjusted for the age of the NAEP samples and the difference between Main and LTT NAEP (the line is the predicted trend for thirteen-year-old students in NAEP-LTT). See appendix section 5.A5 for details (see Note 1).
10. Burkhauser, R. V., Feng, A., Jenkins, S.P. & Larrimore, J. (2009). “Recent Trends in Top Income Shares in the USA: Reconciling Estimates from March CPS and IRS Tax Return Data.” NBER Working Paper No. 15320. Cambridge, Mass.: National Bureau of Economic Research; Piketty, T. & Saez, E. (2003). “Income Inequality in the United States, 1913–1998.” Quarterly Journal of Economics 118(1): 1–39.
11. My calculations, based on Current Population Survey, 1968–2009. See ap- pendix section 5.A3 (see Note 1).
12. See appendix section 5.A6 and 5.A7 for details (see Note 1).
13. Wrigley, J. (1989). “Do Young Children Need Intellectual Stimulation? Experts’
Advice to Parents, 1900–1985.” History of Education Quarterly 29(1): 41–75.
14. Levy, F. & Murnane, R. J. (1992). “U.S. Earnings Levels and Earnings Inequality: A Review of Recent Trends and Proposed Explanations.” Journal of Economic Literature 30(3): 1333–81.
15. McLanahan, S. (2004). “Diverging Destinies: How Children Are Faring Under the Second Demographic Transition.” Demography 41(4): 607–27.
16. Jargowsky, P. A. (1996). “Take the Money and Run: Economic Segregation in U.S. Metropolitan Areas.” American Sociological Review 61(6): 984–98; Rear- don, S. F. & Bischoff, K. (2011). “Income Inequality and Income Segregation.” American Journal of Sociology 116(4): 1092–153; Reardon, S. F. & Bischoff, K. (2011). “Growth in the residential segregation of families by income, 1970- 2009.” US2010 Project Census Brief. Russell Sage Foundation. Available at
Community Investments, Summer 2012 – Volume 24, Number 2
39

Federal reserve Bank oF san Francisco
101 MARKET STREET
SAN FRANCISCO, CA 94105
ADDRESS SERVICE REQUESTED AU 10252
Free subscriptions and additional copies are available upon request from the Community Development Department, Federal Reserve Bank of San Francisco, 101 Market Street, San Francisco, California 94105, or call (415) 974-2765.
Change-of-address and subscription cancellations should be sent directly to the Community Develop- ment Department. Please include the current mail- ing label as well as any new information.
The views expressed are not necessarily those of the Federal Reserve Bank of San Francisco or the Federal Reserve System. Material herein may be reprinted or abstracted as long as Community Investments is credited. Please provide Laura Choi in the Community Development Department with a copy of any publication in which such material is reprinted.
Call for Papers: 2013 Community Development Research Conference
Resilience and Rebuilding for Low-Income Communities: Research to Inform Policy and Practice
The Community Affairs Officers of the Federal Reserve System invite paper submissions for the eighth Federal Reserve Community Development Research Conference, to be held April 11-12, 2013, in Washington DC. The goal of the conference is to highlight new action-oriented and academically rig- orous research on resiliency and rebuilding initiatives that can directly inform community development policy and practice. The deadline is September 15, 2012.
For more information, visit
www.frbatlanta.org/documents/commdev/research_conf_2013_call_for_papers.pdf
|