Profitwise News and Views

September 2003 Issue

Seeds of Growth Conference Summary

Seeds of Growth Sustainable Community Development: What Works, What Doesn't, and Why

Session Four - Public Policy - Enterprise Zones and CDFIs
"Where do community development financial institutions (CDFIs) fit into the idea of development finance, particularly in lower income communities?" That's the question posed by Lehn Benjamin, Julia Sass Rubin, and Sean Zielenbach in their paper, Community Development Financial Institutions: Current Issues and Future Prospects.

The paper begins with a brief survey of the history of community-based financial institutions, from the black-owned banks of the 19th and early 20th centuries to low-income community credit unions and the explosive growth of CDFIs after the 1995 creation of the CDFI Fund. Today, there are 635 CDFIs, most of which are unregulated, nonprofit organizations,¹ including: 71 banks, thrifts, and bank holding companies; 119 credit unions; 20 venture capital funds; and 425 loan funds (includes housing, business, and facility funds).

The whole reason behind the CDFI movement is the historical lack of access to affordable credit and investment capital in lower-income communities. The number of certified CDFIs tripled between 1996 and 2003. This is an increasingly large segment of the development finance industry. "So what exactly is the role of these groups?" asked Sean Zielenbach, research director at the Housing Research Foundation, who presented the paper.

The authors identify and illustrate three general approaches CDFIs take in identifying and serving their markets. Although some CDFIs emphasize one approach more than others, most CDFIs probably have elements of each of these approaches.

The first is a "community-controlled capital" model that developed in response to the practice of redlining.² In this model, CDFIs function as an alternative to conventional lenders, keeping money in the community and recycling it in the form of loans.

The second and most prevalent approach is one that complements conventional lenders, helping them reach deeper into underserved, lower-income markets. Zielenbach highlighted multibank CDCs, Neighborhood Housing Services (NHS), Local Initiatives Support Corporation (LISC), and Boston Community Capital to illustrate this approach. In this role, CDFIs both mitigate risks and develop the market for conventional lenders.

The third approach is the niche market model. In this model, CDFIs serve markets that are just too costly for the conventional lenders to serve efficiently. Micro-enterprise funds and unsecured working capital loans for nonprofits are examples.

CDFIs and Conventional Institutions
The paper asks the question "What differentiates CDFIs from conventional institutions?" There is anecdotal evidence of CDFIs serving harder to serve markets or unbankable borrowers. But it's not really known how much deeper CDFIs go into those markets than conventional institutions do. What really differentiates CDFI products and borrowers from those of conventional banks?

The authors cite two recent studies that try to answer these questions. The Woodstock Institute found that 80 to 90 percent of community development banks' loans go to lower-income markets, compared to about 40 to 50 percent of similarly-sized community banks. A CDFI Fund study found that CDFI financial performance is worse than that of their conventional peers, perhaps owing to their serving a presumably higher risk market. "But we don't know how high a risk," Zielenbach emphasized.

The role of the CDFIs in community revitalization is dictated, at least in part, by what markets banks can realistically serve. Zielenbach and his co-authors assert that the combination of credit scoring and Community Reinvestment Act (CRA) pressures have moved banks deeper into some markets: "About ten years ago it was difficult for banks to provide mortgages to individuals making 80 percent of area median income. You now have some cases where banks are making loans to individuals making 50 percent of area median income."

Even as banks move deeper into some underserved markets, it is not clear that they serve heretofore underserved individuals as well as or better than CDFIs. It is also not clear if banks' extension into these markets has the "spillover" benefits that CDFIs provide. This all seems to beg two questions: Are there intrinsic benefits to the CDFI model? And what is the real impact of CDFIs, anyway?

In order to better analyze the impact of CDFIs and measure those impacts, CDFIs need to first define their goals and desired impacts better. "Too many organizations are simply saying that their mission is "to reduce poverty throughout Chicago," or "promoting economic development in western Michigan," Zielenbach said. He highlighted the Englewood neighborhood on the south side of Chicago, which he called a "basket case," in spite of $43 million in residential and small business lending by conventional, CRA-regulated financial institutions in 2000 alone. "Clearly," Zielenbach concluded, "there is a lot more to neighborhood development than simple financing."

Another impediment to thorough analysis of CDFI impact is figuring out what's reasonable to measure and being realistic about causality. While most organizations cite numbers of housing units created, jobs created, or jobs retained, the authors are emphatic that these are really, at best, indirect impacts of CDFI activities. "For the most part, CDFIs have assumed a direct causality between the financing that they provide and the creation of these outputs, counting both their own and their borrowers' activities, and ignoring other contributing factors...they themselves are not real estate developers or large employers. They do not build projects, but instead help to finance them."

The authors propose a few different ways of assessing impact. First, there is the level where CDFIs directly affect borrowers, projects, and organizational growth and development. How many clients now have any account with a regulated financial institution? Would the project be completed as quickly, efficiently, or cost-effectively without the CDFI? Can the CDFI provide a value added service that contributes to organizational development beyond the financing? These are important, but micro-level CDFI impacts.

What also needs to be accounted for are the impacts that projects financed by the CDFI have in the broader community. These are important indirect impacts. Zielenbach illustrated the point with a senior center, financed through the CDFI that, in addition to providing housing units, also provides a recreational opportunity for seniors in the community. "This can help build a community; it can help stabilize a community," Zielenbach said. "Is it directly tied to the CDFI's financing? No. Is it an indirect outcome? Yes."

Finally, the authors assert that the "bigger issue" in assessing CDFIs is their role in financial system development. How do CDFIs bring in other lenders? How do they help conventional lenders change the way they serve more customers? Are they helping to attract additional private capital to the market?

The paper concludes with questions about the cost, scale, and capital needed to sustain CDFIs, leading the reader to ponder how sustainable the CDFI model is. Zielenbach summarized those questions: "Probably 85 percent of CDFIs are not self-sufficient because they provide a range of services in addition to their financing. How do we pay for that? Too many CDFIs are tiny. Should we be thinking about mergers and consolidation in the CDFI industry? Can CDFI loans be standardized and securitized?"

The authors also look to the future. It's clear that CDFIs are not going to be able to turn around distressed neighborhoods or ensure adequate access to capital and credit by themselves. "CDFIs are a means to an end. We tend to look at them, as with all public policy efforts, as ends in and of themselves," Zielenbach concluded. "What is the role of conventional lenders and investors? The role of CDFIs? The role of the public sector? Obviously there's some subsidy that's going to be needed to create local infrastructures and make communities something in which financing makes sense. Where's the public sector here; where's the government? These are issues we need to think about."

In the next paper, Job Creation in California's Enterprise Zones: A Comparison Utilizing A Propensity Score Matching Model, Susanne O'Keefe, assistant professor of Economics at California State University, Sacramento, finds that California's Enterprise Zone program raised employment 2 to 3 percent per year.

California's Enterprise Zone program, established in 1985 and currently including 39 zones, uses hiring tax credits, machinery tax credits, net operating loss carry-forward, and net interest deductions for lenders to promote employment and economic growth in the distressed areas designated as enterprise zones.

The study focuses on the job creation aspect of the Enterprise Zone program. "What we would like to learn is, what would have happened in those enterprise zones if the program did not exist, which, of course, is an impossible question to answer," O'Keefe pointed out. "The best we can do is to try to control for characteristics of the zone by finding areas that are most similar to enterprise zones before the program began. I used the propensity score matching model to find areas most similar to enterprise zones."

O'Keefe developed a propensity score matching model by which she determined the probability of every census tract in California becoming an enterprise zone based on the observed characteristics of those tracts that did become enterprise zones. She then matched zones to tracts with the closest propensity score that lay outside the zone but within the same county. Table 1 shows the results of this matching exercise.

Growth in the enterprise zones was then compared to growth in the matched areas. The study finds employment growth of 22.7 percent in enterprise zones (342,377 new jobs) as compared to 14.6 percent employment growth in "matched" census tracts between 1992 and 1999.

Using firm level data from the California Employment Development Department that provided employment and payroll at each firm between 1992 and 1999, the study compares the aggregated data for enterprise zone census tracts to matched tracts. The study also compared employment growth at firms located within enterprise zones for firms that are observed for two consecutive years - the firm level analysis.

From this analysis, O'Keefe concluded, "Employment in the enterprise zones grew more quickly than the matched areas. Monthly earnings and number of firms grew more quickly in the matched areas than in the enterprise zones. Because there is a tax incentive for hiring workers for around $8 an hour, there could be an incentive to create more relatively low-wage jobs as compared to high-wage jobs, slowing monthly earnings growth. But, when the regression analysis is done, the earnings and number of firms findings are not statistically significant."

So the study concludes that both the census tract and establishment level analyses support the finding that, while enterprise zone designation does not significantly affect the number of firms or average earnings, zone designation does raise employment by 2 to 3 percent each year.

Julia Sass Rubin and Gregory M. Stankiewicz provide a legislative history of the New Markets initiatives that places the initiatives in the historical context of federal community development initiatives of the last half of the 20th century as well as the philosophical context of the Clinton and Bush administrations. Their paper, Evaluating the Impact of Federal Community Economic Development Policies on Targeted Populations: The Case of the New Markets Initiatives of 2000, raises concerns about the initiatives' prospects for success.

The Clinton administration created the New Markets programs in 1998, and Congress passed the programs in an agreement between President Bill Clinton and Speaker of the House Dennis Hastert in 2000. In spite of three tours of distressed areas that were reminiscent of the tours taken by Lyndon Johnson and Robert Kennedy 40 years earlier, President Clinton insisted that the New Markets initiatives were not designed for poverty alleviation.

Instead, the program was "sold" as a non-inflationary, pro-growth program. "The program is based on the assumption that investors need a little extra incentive. It's not a question of discrimination or bad deal flow, but these are just deals that need a little extra push," according to Rubin, a post-doctoral fellow at the Taubman Center for Public Policy at Brown University. "That's what the New Markets Tax Credit (NMTC) and the New Markets Venture Capital (NMVC) programs are designed to do."

The NMVC program is modeled on the Small Business Administration's (SBA) previous experiences in community development investments and venture capital programs. Rubin and Stankiewicz detail the elements of the program including the SBA's certification process for NMVC company designation and the matching grants and loans, 80 percent of which must be used to invest in low-income communities. The authors note that the program was eliminated in the fiscal 2003 budget and that proponents may not be able to revive it in the current political and budgetary climate. Even if the program could be revived, however, the authors are concerned about the mismatch of sources and uses of funds in the program.

"By utilizing debt to fund mostly equity investments, the NMVC program duplicates the mismatch between sources and uses of capital that had led to financial problems for most of the participants in the federal Small Business and Specialized Small Business Investment Company programs...although the five year interest deferment provided by the SBA is helpful, it still is unlikely that a NMVC company could invest its capital in equity and then exit those investments quickly and profitably enough to pay back its loan to the SBA."

The paper also details the NMTC program. Modeled on the Low-Income Housing Tax Credit program, the NMTC investors in certified community development entities (CDEs) will receive a 39 percent tax credit over seven years for their investments, 85 percent of which must be made in low-income communities.

As with the venture capital initiative, the authors find reasons to be concerned about the efficacy of the tax credit portion of the New Markets programs. First, the program's general economic development objective is ambiguous and the CDE's ability to raise capital quickly is more heavily weighted than community impact in the application process. This leads the authors to wonder whether the resulting investments will supplant rather than supplement investments that would have been made with or without the program's incentives.

Secondly, the NMTC program allows "double-dipping" with most federal and state programs according to the authors.3 The authors are also concerned that the NMTC program may encourage a "race to the bottom" in raising capital.

The CDEs have used the NMTC selection process to demonstrate their ability to raise capital, using letters of intent from potential investors. These letters are nonbinding, allowing investors to withhold funding until a review of specific deals is conducted. Under these situations, preference would go to the most lucrative deals instead of those that would have the greatest community development impact.

Rubin and Stankiewicz also warn readers about unrealistically high expectations with regard to program impacts. They also believe that, while business equity is what is most needed in new markets, it is the least likely to get funded through the NMTC program while commercial real estate deals are more likely to get funded under the program.

Finally, the program faces the double threat of additional funding cuts due to growing federal budget deficits and a reduction in demand for credits due to the elimination of "double" taxation of corporate dividends.

In closing her remarks, Rubin proposed potential program modifications that would help address these concerns. Those proposals include allowing more near-equity and debt investing and increasing the weighting for community development objectives in the application processes. She also proposed eliminating "double-dipping" opportunities and encouraging more equity investments by addressing issues related to the seven-year minimum holding period for equity investments. Finally, and perhaps most importantly, she wants the program to encourage investment that supplements rather than supplants investments that likely would have happened anyway, specifically by limiting organizations' ability to invest in related entities.

Paul Pryde, president of Washington, DC-based Capital Access Corporation, was the discussant for this panel. Noting that next year will mark the 40th anniversary of the War on Poverty, Pryde identified two shifts in the way the federal government has attempted to alleviate the effects or causes of unemployment and low incomes. The first shift went from the original direct provision of services by the federal government to an economic development approach - "creating the capacity to produce among low-income residents and making low-income communities more viable places to live and work," Pryde said.

The second shift was away from economic development to market-oriented solutions, which leads to the present day. Under the market-oriented approach, "we have a variety of ways that government tries to use its taxing, regulatory, and spending authority to change the way that private lenders, investors, and citizens make decisions," Pryde said. "Each one of the papers we have discussed this morning represents a somewhat different approach to government's use of its powers to influence private decision-making."

Pryde applauded O'Keefe's paper on the California enterprise zones for showing, in a way that has not been done before, that enterprise zones work to increase employment. But why they work is still not known. "Some data suggest that young, small firms are a large generator of new jobs. Yet, in the early years, these same firms have very little tax liability and have very little capacity to use tax credits or any type of tax incentives. So you would suspect that," Pryde asserted, "to the extent that these are the firms that are creating jobs in California's enterprise zones, firm-level tax credits and deductions don't work very well." He suggests that, "perhaps the incentives that work best are those that get money into firms so that they can expand."

Pryde noted some of the same challenges for CDFIs that were covered in the Benjamin, Rubin, and Zielenbach paper. While highlighting the demographic changes that will be driving more commercial banks toward serving nontraditional markets, Pryde also noted that competition would come increasingly from nonbank financial service providers. "If you look at trends in the banking industry, in every credit market in which banks compete, they're losing market share - mortgages, consumer credit, and so forth - and they're losing it to specialty lenders - specialty finance companies and diversified financial services companies," he said. "So as CDFIs look out for competition, don't look just at banks. You have to look at other organizations that are increasingly going to look at these markets that CDFIs have historically served."

In addressing questions of scale for CDFIs, Pryde emphasized the need for CDFIs to adopt financial information technologies that have made it possible for commercial lenders to better serve historically underserved markets. As an example, Pryde pointed to securitization. "Securitization has made it possible for anybody to get a credit card and anybody to get a mortgage loan. Those are the most securitized assets in the country," Pryde said. "Securitization means pooling loans and selling bonds backed by those loans to investors. And because of the advances in securitization, mortgage credit and consumer credit is available to almost anybody. My belief is that the same technology can be used to make credit more widely available to low-income communities. There are challenges associated with it, but I think it is definitely a promising technique."

Pryde then turned his attention to the NMTC, agreeing with Rubin and Stankiewicz that not much is known about how the credits will work. But he took issue with the notion that the $15 billion allocation is a thin incentive. Based on the SBA's highest default rate in the last 20 years of 10 to 12 percent, Pryde believed that if the tax credits were used as a credit support mechanism, they could leverage $150 billion in new lending.

Pryde concluded his remarks suggesting that the problem of getting capital flowing to underserved markets has been turned on its head. "There seems to be a willingness to invest, but we've got challenges in getting the money to the borrowers who need the money," he said. "We have to ask ourselves, what really are the limits of private markets' ability to serve low-income communities. Because of the advances in financial information and technology," he asserted, "the private sector can serve markets that, 30 years ago, they weren't interested in and couldn't efficiently serve."

"So," he concluded, "the problem we face is not a financial problem. It's an intellectual problem. How do we use our creativity to figure out what the solutions are to a challenge that we have faced for the last decade or more?"

Notes
¹ "A little less than 30 percent are regulated financial institutions, and probably only 15 percent of all CDFIs are for-profit institutions."

² "The CRA sought to eliminate the more insidious practice of redlining, in which bankers refused to lend in certain geographic markets because of the perception of high risk in those communities. (The term "redlining" resulted from certain bankers' demarcation of high-risk neighborhoods by stark red lines on city maps.) Contributing factors to high-risk perceptions included large numbers of racial and ethnic minorities and high poverty and unemployment rates" (Benjamin, Rubin, and Zielenbach, 2003, p. 4)

³ "The seventh largest recipient, which received $110 million is an organization called Advantage Capital, a certified capital company which, in nine states, is eligible to receive 100 percent subsidy for making investments," Rubin said. "So in those states, they would now receive 139 percent subsidy for their investments. It's hard to argue that this money is actually inducing them to make new investments."