This is the accessible text file for GAO report number GAO-14-500 entitled 'New Markets Tax Credit: Better Controls and Data Are Needed to Ensure Effectiveness' which was released on August 11, 2014. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. United States Government Accountability Office: GAO: Report to the Ranking Member, Committee on Homeland Security and Governmental Affairs, U.S. Senate: July 2014: New Markets Tax Credit: Better Controls and Data Are Needed to Ensure Effectiveness: GAO-14-500: GAO Highlights: Highlights of GAO-14-500, a report to the Ranking Member, Committee on Homeland Security and Governmental Affairs, U.S. Senate. Why GAO Did This Study: In recent years, private investors have claimed more than $1 billion in NMTCs annually. The credits are combined with private loans and other public funds to support investments in low-income communities. GAO was asked to review the financial structure of NMTCs. This report assesses: (1) the complexity and transparency of NMTC financial structures and controls over the size of federal subsidies; (2) what is known about the types and amounts of fees and other costs of the financial structures; (3) what is known about the equity remaining in low-income community businesses after the 7-year credit period; and (4) what is known about NMTC project failure rates. GAO reviewed Treasury NMTC data and surveyed CDEs that allocated credits to 305 projects in 2010-2012. What GAO Found: The financial structures of New Markets Tax Credit (NMTC) investments have become more complex and less transparent over time. The increased complexity is due, in part, to combining the NMTC with other federal, state, and local government funds. Based on GAO's survey of Community Development Entities (CDEs) an estimated 62 percent of NMTC projects received other federal, state, or local government assistance from 2010 to 2012. While combining public financing from multiple sources can fund projects that otherwise would not be viable, it also raises questions about whether the subsidies are unnecessarily duplicative because they are receiving funds from multiple federal sources. In addition, in some cases the complexity of the structures may be masking rates of return for NMTC investors that are above market rates. For example, a study done for the Department of the Treasury (Treasury) found an investor apparently earning a 24 percent rate of return, which is significantly above market rates of return. In that case, the investor leveraged the NMTCs by using other public funds to increase the base for claiming the NMTC. Treasury and the Internal Revenue Service issued guidance about allowable financial structures in the early years of the NMTC program, but the guidance has not been updated to reflect the subsequent growth in complexity, such as the use of other public money to leverage the NMTC. Treasury also does not have controls to limit the risk of unnecessary duplication in government subsidies or above market rates of returns. Without such guidance and controls the impact of the NMTC program on low-income communities could be diluted. The costs of complex NMTC financial structures may not be fully reflected in fees charged by CDEs, and they could be reflected in other costs such as higher interest rates. Treasury has taken steps to ensure businesses are better informed about fees and other costs, but is not collecting these additional data itself. Without these data, Treasury is limited in its ability to analyze NMTC program benefits. GAO also found that the data on equity remaining in businesses after the 7-year credit period were unreliable because, in part, instructions on what to report are unclear. As a result, at this time it is not possible to determine how much equity remains in low-income community businesses after 7 years. Similarly, data on NMTC project failure rates were unavailable. GAO reviewed data of performance on loans from CDEs to low-income community businesses as an indicator of whether the businesses will be viable over the long term. However, data on loan performance were also incomplete because some reporting of this information by CDEs is optional. As a result, it is not possible to determine, at this time, the NMTC project failure rate with certainty. What GAO Recommends: Treasury should issue further guidance on how other government programs can be combined with NMTCs; ensure adequate controls to limit the risks of unnecessary duplication and above-market rates of return; and ensure that more complete and accurate data are collected on fees and costs, the equity remaining in the business after 7 years, and loan performance. Treasury agreed with GAO's recommendations to improve data collection on equity remaining and loan performance. Treasury said that a recently formed working group, that includes representatives from the Community Development Financial Institutions Fund and the Internal Revenue Service, is considering GAO's other recommendations. View [hyperlink, http://www.gao.gov/products/GAO-14-500]. For more information, contact James R. White at (202) 512-9110 or whitej@gao.gov. [End of section] Contents: Letter: Background: The Financial Structures of NMTC Investments Have Become More Complex and Less Transparent While Treasury Guidance Covers Only the Simpler Structures: Fees and Retentions Reduce the Available NMTC Equity but Lack of Transparency Makes the Size of the Reduction Uncertain: Data on NMTC Equity Remaining in the Low-Income Community Businesses Are Not Sufficiently Complete or Accurate: CDFI Fund Data on Distressed Projects Are Not Sufficiently Reliable: Conclusions: Recommendations for Executive Action: Agency Comments and Our Evaluation: Appendix I: Objectives, Scope, and Methodology: Appendix II: Other Federal Funding Used in New Markets Tax Credit Projects: Appendix III: Regression Analysis of Total Fees and Retentions Associated with Characteristics of the Project Using CDFI Fund Data, 2011-2012: Appendix IV: Comments from the Department of the Treasury: Appendix V: GAO Contact and Staff Acknowledgments: Related GAO Products: Tables: Table 1: Estimated Percentage of Projects Receiving Other Federal Funding (Projects Originating in 2010-2012): Table 2: New Markets Tax Credit (NMTC) Investment Retained and Fees Charged by Community Development Entities (CDEs), 2011-2012: Table 3: Project Characteristics Associated with Fees and Retentions Charged by Community Development Entities (CDEs), 2011-2012: Table 4: Other Federal Tax Credits Used in a Sample of New Markets Tax Credit (NMTC) Projects, 2010-2012: Table 5: Tax Exempt Bond Financing Used in a Sample of New Markets Tax Credit (NMTC) Projects, 2010-2012: Table 6: Grants or Direct Payments Used in a Sample of New Markets Tax Credit (NMTC) Projects, 2010-2012: Table 7: Direct Loans or Guaranteed Loans Used in a Sample of New Markets Tax Credit (NMTC) Projects, 2010-2012: Figures: Figure 1: Simplified Example of a New Markets Tax Credit (NMTC) Basic Model: Figure 2: Simplified Example of New Markets Tax Credit (NMTC) Investment Leveraged with Private Funds: Figure 3: Simplified Example of a New Markets Tax Credit (NMTC) Investment Combined with (but Not Leveraged by) Other Public Funds: Figure 4: Simplified Example of New Markets Tax Credit (NMTC) Investment Leveraged with Both Private and Public Funds: Figure 5: Number of Projects by Total Retentions and Fees Charged as a Percentage of Total New Markets Tax Credit (NMTC) Investment, 2011- 2012: Figure 6: Data Quality for Amounts of Equity Projected to Remain in the Low-Income Community Businesses, 2011-2012: Abbreviations: CDE: Community Development Entity: CDFI: Community Development Financial Institutions: CIIS: Community Investment Impact System: HTC: historic tax credit: IRS: Internal Revenue Service: LIC: low-income community: LIHTC: Low-Income Housing Tax Credit: NMTC: New Markets Tax Credit: QEI: qualified equity investment: [End of section] United States Government Accountability Office: GAO: 441 G St. N.W. Washington, DC 20548: July 10, 2014: The Honorable Tom Coburn, M.D. Ranking Member: Committee on Homeland Security and Governmental Affairs: United States Senate: Dear Dr. Coburn: Congress established the New Markets Tax Credit (NMTC) program as part of the Community Renewal Tax Relief Act of 2000 to encourage investors to invest in impoverished, low-income communities that traditionally lack access to capital.[Footnote 1] Conventional access to credit and investment capital for developing small businesses, creating and retaining jobs, and revitalizing neighborhoods is often limited in economically distressed communities or in communities with large low- income populations. The NMTC provides investors (individuals, financial institutions, other corporations, etc.) with a tax credit for investing in communities that are economically distressed or consist of low-income populations. The investors are able to claim a credit equal to 39 percent of eligible investment spread over 7 years. The NMTC program is administered by the Community Development Financial Institutions (CDFI) Fund in the Department of the Treasury (Treasury) which allocates tax credit authority--the amount of investment which investors use as the base for determining the amount of tax credits they are eligible to claim--to Community Development Entities (CDE) that apply for and obtain allocations. These CDEs in turn offer the tax credits to private investors in exchange for equity investments with the CDEs. The CDEs then invest the capital they raise in qualified low-income community businesses. The resulting NMTC financial structures include the CDEs, investors, and qualified low- income community businesses--as well as the transactions between them-- and may also include private loans and funding from other government programs. We have issued three reports on the status and initial results of the NMTC program.[Footnote 2] You asked us to follow up on these reports and review the NMTC financial structures and the amount of equity investment flowing through to qualified low-income community businesses. Our objectives were to assess: (1) NMTC financial structures in terms of their complexity, transparency, and their effect on the size of the federal subsidies going to NMTC projects, as well as controls to ensure that subsidies are not larger than necessary for the investment; (2) what is known about the types and amounts of fees and other costs that reduce the amount of equity reaching low-income community businesses; (3) what is known about the amount of equity left in the low-income community businesses after the 7-year credit period; and (4) what is known about the NMTC projects that are at risk of failing by becoming economically nonviable. Our assessment criteria for objective 1 were drawn from our prior work evaluating tax expenditures, which discuss complexity (or simplicity), transparency, and duplication, as well as from internal control standards about the efficient use of resources.[Footnote 3] For objective 1, we reviewed the NMTC literature and interviewed representatives of CDEs and researchers who have evaluated the program to determine how the financial structures have evolved, and conducted and analyzed the results of a survey of CDEs to determine how the NMTC has been combined with other federal tax benefits, grants, and other assistance.[Footnote 4] Our criteria for objectives 2, 3, and 4 were GAO criteria for valid data--that they be sufficiently accurate and complete to capture program performance. For objective 2, we analyzed data from the CDFI Fund on fees (actual and projected) that CDEs charged to projects with loans and investments originating in 2011 and 2012. (The CDFI Fund started collecting these additional data on fees in 2011 based on our recommendation.[Footnote 5]) For objective 3, we analyzed data from the CDFI Fund on the projected equity remaining in the business for projects beginning in 2011-2012. The CDFI Fund requires CDEs to project equity remaining in businesses for loans and investments originating in 2011 or later. For objective 4, we analyzed data reported by CDEs to the CDFI Fund that could indicate that a project is experiencing financial distress, such as whether a loan on a project is delinquent, charged off, or restructured. We conducted this performance audit from May 2013 to July 2014 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. See appendix I for more details about our scope and methodology. Background: The CDFI Fund is authorized to allocate tax credit authority to CDEs that manage NMTC investments in low-income community development projects. The CDEs are domestic corporations or partnerships with a primary mission of providing investment capital for low-income communities or low-income persons. Some CDEs are established by other public or private entities, such as local governments or financial institutions. Through the CDFI Fund, Treasury awards tax credit authority to CDEs through a competitive application process. After CDEs are awarded tax credit authority, they use it to attract investments from investors who then claim the NMTC. CDEs use the money raised to make investments in projects (one or more) in low-income communities. In the 11 rounds of allocations since 2003, Treasury, through the CDFI Fund, has made allocations to CDEs that total $40 billion. The base for claiming the credit on a project is called the qualified equity investment (QEI). It equals the credit authority allocated to a project by a CDE and generally covers a portion of the total project costs. The equity in the QEI includes money provided by NMTC investors but may also include money from private lenders or other government entities. The NMTC, which is equal to 39 percent of the QEI, is claimed over 7 years--5 percent in each of the first 3 years and 6 percent over each of the last 4 years. In recent years, private investors have claimed more than $1 billion in NMTCs annually. When the QEI does not cover the entire project cost, the NMTC financing is supplemented by other financing outside the NMTC allocation. CDEs are required to invest the funds they receive in qualified low- income community investments, which include, but are not limited to, investments in operating businesses and residential, commercial, and industrial projects. Although the range of activities financed by CDEs varies, about half of NMTC investments have been used for commercial real estate projects. The program expired in 2013 but legislation has been proposed to extend it, and the President requested a permanent extension in his fiscal year 2015 budget proposal.[Footnote 6] The Financial Structures of NMTC Investments Have Become More Complex and Less Transparent While Treasury Guidance Covers Only the Simpler Structures: NMTC investors have developed financial structures that increase the amount of other funding from either private or public sources that are used with the NMTC--a process that is called increasing the leverage on the investment. These structures can increase the amount of federal subsidy to a project and can result in projects being undertaken that would not otherwise have been started for lack of sufficient funding. However, they also increase the complexity of the financial structures by adding more parties and more transactions, which in turn reduces transparency and may increase the cost in terms of fees and other related transactions costs. An assessment of the NMTC program as a whole requires complete information on all the costs and benefits attributable to the NMTC, including these administrative and compliance costs.[Footnote 7] Figures 1-4 are simplified illustrations of these increasingly complex financial structures.[Footnote 8] Figure 1: Simplified Example of a New Markets Tax Credit (NMTC) Basic Model: [Refer to PDF for image: model illustration] NMTC Investor: $10 million equity: Equity provided to the CDE in exchange for tax credits. Community Development Entity: Receives $10 million NMTC allocation. Low interest rate only loan. Low-income Community Business: $10 million of total investment. Interest payments to CDE. NMTC and interest payments from CDE to NMTC Investor. In the most basic of the NMTC financing structures, an investor contributes funds directly to a community development entity (CDE). The CDE has received authority from the CDFI Fund to issue tax credits for a $10 million project. The investor invests $10 million for 7 years and receives a tax credit of 39 percent on the $10 million. At the end of the 7 years, all or a portion of the investor's equity is returned, and the low-income community (LIC) business is expected to operate on its own with financing from other sources, perhaps market rate loans. The return the investor earns is a combination of the tax credits plus interest payments made by the LIC business. The investor incurs some project related risks since the LIC business interest payments are not necessarily guaranteed and the business could go bankrupt and fail to repay the loan made by the CDE. The investor earns an annual rate of return of approximately 6 percent based on the credits alone which are claimed over the 7-year compliance period.[A] For the investor to receive a return equal to the market rate of return – which for this and the following example is assumed to be 7 percent – the CDE could charge an interest rate to the LIC business as low as 1 percent. Revenue Effects: The present value of the lost tax revenue to the federal government is $3.46 million – the present value of the $3.9 million in tax credits paid over 7 years evaluated at the federal government's 7-year borrowing rate of about 3 percent. Source: GAO analysis. GAO-14-500. [A] This rate of return represents the internal rate of return of the $10 million investment in terms of the value of the tax credits. The tax credit is 5 percent of the investment amount in each of the first 3 years and 6 percent in each of the last 4 years during the 7-year compliance period. [End of figure] Figure 2: Simplified Example of New Markets Tax Credit (NMTC) Investment Leveraged with Private Funds: [Refer to PDF for image: model illustration] NMTC Investor: $2.97 million equity investment to Investment Fund. Leveraged Lender: $7.03 million loan to Investment Fund. Community Development Entity Receives $10 million NMTC allocation from Investment Fund. Low-income Community Business: $10 million of total investment from CDE. Interest payments from Low-income Community Business to CDE. Interest payments and NMTCs from CDE to Investment Fund. Investment Fund: $3.9 million in NMTCs to NMTC Investor; Interest payments for 7 years to Leveraged Lender. In this model the equity is combined in an investment fund with a loan from a leveraged lender. The NMTC investor receives tax credits on the full amount of the equity plus the loan. The equity investment is generally not paid back to the investor at the end of the compliance period. The leverage lender receives interest payments throughout the 7-year compliance period and unless refinanced, the loan amount is repaid at the end of the compliance period. The borrowed funds and the equity investor's cash are used by the CDE to make an investment (usually in the form of loans) to the LIC business. The CDE then passes the tax credits and interest earned on the investment (less any fees) back to the investment fund, where they are distributed to the equity investor and leverage lender. In the example, the allocation is again $10 million but in this case the investor puts $2.97 million into the fund and the leveraged lender loans the remaining $7.03 million. The investor's $2.97 million is the present value to the investor of the stream of tax credits that will be generated by the investment evaluated at the 7 percent rate of return. The investor incurs limited project related risk since the credit can be claimed whether the low-income community business succeeds or not.[A] Effect on Complexity: The complexity of this structure creates added cost from the resources used to create entities like the investment fund and to negotiate the terms of the investment among an increasing number of participants. Revenue Effects: The present value of the tax revenue loss to the government is $3.46 million as in figure 1. Source: GAO analysis. GAO-14-500. [A] The investor's risk is largely limited to "recapture" risk where the IRS reclaims credits from prior years as well as future credits due to issues related to CDE compliance with IRS regulatory requirements. See GAO, New Markets Tax Credit: The Credit Helps Fund a Variety of Projects in Low-Income Communities, but Could Be Simplified, GAO-10-334 (Washington, D.C.: Jan. 29, 2010) for a description of the events that can trigger a "recapture" of the credit. [End of figure] Figure 3: Simplified Example of a New Markets Tax Credit (NMTC) Investment Combined with (but Not Leveraged by) Other Public Funds: [Refer to PDF for image: model illustration] NMTC Investor: $2.38 million equity investment to Investment Fund. Leveraged Lender: $5.62 million loan to Investment Fund. Community Development Entity Receives $8 million NMTC allocation from Investment Fund. HTC Investor: $2 million HTC equity to Low-income Community Business. Low-income Community Business: $10 million of total investment. Interest payments from Low-income Community Business to CDE. Low-income Community Business: $2 million HTCs to HTC Investor. Interest payments and NMTCs from CDE to Investment Fund. Investment Fund: $3.12 million in NMTCs to NMTC Investor; Interest payments to Leveraged Lender. The leveraged structure can also be used with other public subsidies such as grants and tax credits where the subsidies are not part of the leverage structure; i.e., the grant or tax credit is not combined with the NMTC equity in the investment fund to serve as the basis for NMTCs. In contrast to figure 2, the CDE has only an $8 million NMTC allocation for the $10 million project but, in this example, all $10 million of the project's expenditures also qualify for a federal tax credit for rehabilitating historic structures. The historic tax credit (HTC) rate is 20 percent which yields an HTC of $2 million. The NMTC that can be claimed on the $8 million allocation is $3.12 million which has a present value to the investor of $2.38 million (evaluated at the 7 percent rate of return) which the investor puts into the investment fund. The remainder of the $8 million required to claim the NMTC credit is provided by the leveraged loan of $5.62 million. The additional $2 million required to fund the $10 million project is provided directly to the LIC business as equity by the HTC investor. Effect on Complexity: The complexity in this case arises, in part, from the need to create additional financial entities to conform to the legal requirements of the different types of public funds. The HTC investor must use a legal entity called a master tenant (not shown in figure 3) to receive the HTCs and make the HTC equity investment in the LIC business. Revenue Effects: In this case, the revenue cost sums to $4.76 million – $2.76 million (the present value of the $3.12 million of NMTCs) plus the $2 million of HTCs. Source: GAO analysis. GAO-14-500. [End of figure] Figure 4: Simplified Example of New Markets Tax Credit (NMTC) Investment Leveraged with Both Private and Public Funds: [Refer to PDF for image: model illustration] NMTC Investor: $2.97 million equity investment to Investment Fund. HTC Investor: $2 million HTC equity to Investment Fund. Leveraged Lender: $5.03 million loan to Investment Fund. Community Development Entity Receives $10 million NMTC allocation from Investment Fund. Low-income Community Business: $10 million of total investment. Interest payments from Low-income Community Business to CDE. Interest payments and NMTCs and HTC from CDE to Investment Fund. Investment Fund: $3.9 million in NMTCs to NMTC Investor; $2 million in HTCs to HTC Investor. Interest payments to Leveraged Lender. Other government subsidies can also be part of the leveraged structure; i.e., the grant or tax credit can be combined with the NMTC equity in the investment fund to serve as the basis for NMTCs. In figure 4, the investor claims the NMTC on the value of HTCs as well as on the NMTC equity and leveraged loan. The NMTC investor community often refers to this practice as “twinning” the NMTC with other tax credits or public assistance. The project has a $10 million NMTC allocation (rather than the $8 million in figure 3) and, as before, $10 million of expenditures that qualify for the HTC yielding a $2 million HTC. The NMTC investor provides $2.97 million in equity but in this case the leveraged lender provides $5.03 million for a total of $8 million. By including the $2 million HTC, the investment fund is brought up to $10 million. Unlike figure 3, the NMTC investor claims the credit on the full $10 million. Effect on Complexity: As in the previous example, complexity arises from the need to create financial entities to conform to the legal requirements of the different types of public funds. But in this case, because the funds are pooled, additional effort is required to ensure that the funds are spent in accordance with regulations. For example, the HTC equity that qualifies for the NMTC in the investment fund must be used for the qualified rehabilitation expenditures that form the basis of the HTC. Revenue Cost: The present value of the $3.9 million in NMTCs is $3.46 million. The total foregone revenue to the government, including the HTC, is $5.46 million. The revenue cost is higher than in figure 3 because the HTC was used to leverage the NMTC. Source: GAO analysis. GAO-14-500. [End of figure] In 2003, the Internal Revenue Service (IRS) confirmed the ability to use private funding to leverage a NMTC structure.[Footnote 9] Also, in 2004, Treasury issued final regulations generally allowing the NMTC to be combined with other tax credits.[Footnote 10] However, neither Treasury nor IRS has specifically confirmed the ability to use other federal or state funding to increase the leverage on NMTC investments. In Practice, NMTC Financial Structures Are More Complex: Many NMTC projects have financial structures that are more complex than the simple examples in figures 1 through 4. According to CDFI Fund data, 21 percent of projects originating in 2010 through 2012 had financial structures involving more than one CDE. In addition, 21 percent of projects had four or more transactions involving financial flows (such as loans from the CDEs to the LIC business). The NMTC financial structures have become more complex over time. We estimated that 41 percent of NMTC investments were made using one of the leveraged models in 2006, while more recent industry estimates state that more than 90 percent used such leverage in 2013.[Footnote 11] The simplified structures in our examples also do not show some of the complexities that are required for the purpose of acquiring the tax benefit. For example, when the public funds are leveraged, as in figure 4, a pass-through entity that is disregarded for tax purposes is established to route the flow of funds back from the low-income community business to the investment fund to claim the NMTC. The complexity of the structures may reduce transparency by making it more difficult to trace the flow of private and public funds and the benefits from the tax subsidies. For example, investors can leverage other tax credits with the NMTCs to gain access to other tax benefits such as when the tax credit for solar equipment is leveraged to claim additional NMTCs and also claim accelerated depreciation deductions based on the solar equipment. However, as mentioned above, this complexity can have benefits because it may result in projects getting financing they could not get otherwise. For example, combining assistance from other government programs with the NMTC, as in figures 3 and 4, could finance projects that would not be viable if they had to rely on only the NMTC and private lenders for financing as in figure 2. A Majority of NMTC Projects Used Other Public Funds in 2010-2012: Based on our survey of CDEs with projects originating in 2010-2012, the use of other public sources of funds with the NMTC is widespread (as shown in Table 1). An estimated: * Sixty-two percent of all NMTC projects received other public funding--funds from federal, state or local public sources; * Thirty-three percent of all NMTC projects received other federal funding; and: * Twenty-one percent of all NMTC projects received funding from multiple other government programs. Among the most frequently used other federal sources by NMTC projects in our survey were historic tax credits (HTC) and tax-exempt bonds for private non-profit education facilities. In addition, for a number of NMTC projects in our survey, small business participants obtained loans guaranteed by the Small Business Administration. The other federal funds most often leveraged with the NMTC projects in our survey were the historic tax credits, Recovery Zone bonds and tax- exempt bonds for private non-profit education facilities. The state and local funds most frequently leveraged with the federal NMTC were state historic tax credits and state new markets tax credits[Footnote 12] (see appendix II for additional results from our survey). Table 1: Estimated Percentage of Projects Receiving Other Federal Funding (Projects Originating in 2010-2012): Type of Federal Funding: Tax credits; Percent of All NMTC projects: 8%; 95 percent confidence interval[A]: 5-12%; Funding amount: $353 million; 95 percent confidence interval: $59-$647 million. Type of Federal Funding: Tax-exempt bonds; Percent of All NMTC projects: 6%; 95 percent confidence interval[A]: 4-11%; Funding amount: $2.023 million; 95 percent confidence interval: $874 million-$3,172 billion. Type of Federal Funding: Grants; Percent of All NMTC projects: 11%; 95 percent confidence interval[A]: 7-17%; Funding amount: $449 million; 95 percent confidence interval: $258-$639 million. Type of Federal Funding: Loans; Percent of All NMTC projects: 11%; 95 percent confidence interval[A]: 7-16%; Funding amount: $740 million; 95 percent confidence interval: $271 million-$1.209 billion. Type of Federal Funding: Total; Percent of All NMTC projects: 33%; 95 percent confidence interval[A]: 27-39%; Funding amount: $3.565 billion; 95 percent confidence interval: $2.317-$4.813 billion. Source: GAO survey of NMTC projects originating in 2010-2012. GAO-14- 500. Note: Total percent of projects receiving other federal funding does not equal sum of components because some projects used more than one type of federal funding. [A] A 95 percent confidence interval means that if you were to determine an estimate for different probability samples, 95 out of 100 times the confidence interval would include the actual population value. In other words, the actual population value is between the lower and upper limits of the confidence interval 95 percent of the time. [End of table] The use of multiple other government programs raises questions about unnecessary duplication. The CDFI Fund currently asks CDEs whether other public funds are used to finance projects but does not ask for any details such as the information on specific programs providing those funds. Therefore, it is not possible at this time to determine whether unnecessary duplication is occurring because of a lack of data. Complex NMTC Financial Structures May Mask Investments with Rates of Return That Are Higher Than Necessary: The competitive or market return should be sufficient to attract enough investors to fund the project because it reflects the return on comparable investments with similar risk. In figures 1 through 4 above, we assumed for illustrative purposes that the competitive market rate of return was 7 percent. We also assumed that NMTC investors provided enough equity to the project (perhaps because of competition for the credits) that the credits claimed over the 7-year compliance period provided a market return. NMTC industry representatives have argued that competition for credits by potential investors and competition between CDEs for credit allocations works to ensure returns on NMTC investments are kept at market rates commensurate with NMTC investor's risk. The NMTC investor's level of risk depends on the financial structure that the investor uses. For example, when the NMTC investor uses a leveraged investment model, the investor does not generally share in the riskiness of the project. Unless the investor is also the leveraged lender or has some other financial interest in the project, it is not exposed to any risk from project failure because such an event generally does not stop the investor from claiming the credit. Some evidence suggests that some investors may receive returns that are above-market and therefore more than the necessary subsidy required to attract the funds. In a case study reported by the Urban Institute, an investor appeared to put in about $500,000 of NMTC equity to claim $1.2 million of NMTCs representing a return of about 24 percent compounded annually.[Footnote 13] The NMTC was leveraged entirely with $2.5 million of federal and state HTCs without use of a conventional leveraged loan in the NMTC structure. As a result, 83 percent of the qualified equity investment on which an investor claimed NMTCs was provided by other federal and state tax credit programs. However, the Urban Institute study authors said that, because of the complex financial structure, they could not rule out the possibility that the investor supplied other, non-NMTC funds to the project and at a lower rate of return. They noted that the NMTC financing was combined with a conventional loan outside the NMTC structure, owner equity, and a loan from the local municipality. If the NMTC investor supplied some of these additional funds, it is possible that the investor's overall rate of return may have been lower and more in-line with the market return. Guidance and Controls Do Not Exist to Prevent Above-Market Rates of Return or Unnecessary Duplication and Costs: Internal controls should provide reasonable assurance that operations and use of resources are efficient. In the context of the NMTC program, the resources administered by Treasury include the tax benefits claimed by the NMTC investors. One NMTC control has been Treasury and IRS guidance on allowable financial structures. The Secretary of the Treasury has the authority to limit how other federal tax benefits are used with the NMTC and the Secretary has used this authority to prohibit its use with the Low-Income Housing Tax Credit (LIHTC).[Footnote 14] While an IRS revenue ruling, issued in 2003, allowed using debt to leverage NMTC equity, the ruling did not explicitly address using other public funds, such as other tax benefits, to leverage the NMTC as in figure 4. Controls do not exist to monitor and prevent unnecessary use of other public funds to supplement the NMTC. As already noted, Treasury, through the CDFI Fund, does not collect information about the specifics of other public funding. For the CDFI Fund, Treasury does not have controls to limit the risk of cases like the example from the Urban Institute study where other public funds were used to expand the NMTC base and apparently generate a 24 percent rate of return for the NMTC investor. We believe that such controls could take a variety of forms and would have to be assessed relative to any added compliance and administrative costs. A control that would provide greater clarity about tax subsidies could be achieved by requiring CDEs to report the NMTC investor's overall rate of return on the NMTC project. As suggested by the Urban Institute example, this information would allow an assessment of whether the NMTC investor is earning a market return commensurate with the risk on its entire investment. With this information, an additional control could be implemented that would require CDEs to justify rates of return above a certain threshold by explaining why this project was so risky that it required a greater- than-market rate of return. Other controls that could be considered include caps on rate of returns and mechanisms to ensure competition among NMTC investors sufficient to prevent above-market rates of return. The decision to adopt any of these controls would require that Treasury compare the benefits of the controls with any compliance costs from added complexity for taxpayers and administrative costs for Treasury from collection and evaluation of the data and monitoring the controls once they are put in place. The complexity of the financial structures creates a lack of transparency for taxpayers and IRS, and can increase both the risk of higher than needed NMTC rates of return and investment transactions costs. Combining multiple investment sources may help some NMTC projects to obtain sufficient financing to proceed. Indeed, NMTCs are often referred to as "gap financing" that can be added to other financing. Some projects may not require NMTC funding. The Urban Institute study estimated that about 20 percent of projects in the first 4 years of the program showed no evidence of needing NMTCs to proceed while about 30-40 percent did with 30 percent uncertain. However, even in the case where the NMTC funding is necessary, the intricate patterns of investment flows through NMTC structures, where multiple sources may be mingled and later dispersed, make it difficult to determine who is receiving subsidies and whether the return to NMTC investors is higher than necessary. In addition, the network of transactions in the NMTC financial structures increase costs both in terms of explicit CDE fees and other resources used to pay the legal and accounting costs necessary to establish the entities that make the transactions.[Footnote 15] Fees and Retentions Reduce the Available NMTC Equity but Lack of Transparency Makes the Size of the Reduction Uncertain: Fees and retentions directly reduce the amount of tax subsidized equity investment that is available to low-income community businesses but these costs do not represent the only way that equity can be reduced. The costs associated with financial structures could also appear in the form of higher interest rates, especially when the investor and leveraged lenders are related parties. In listening sessions organized by CDFI Fund officials, some CDEs reported that comparison of the fees that they charged to the fees charged by integrated investors/lenders were inaccurate because of the ability of integrated investors/lenders to receive compensation in other ways. The fees, interest, and other costs that can offset one another (a low fee may be offset by a high interest rate) also reduce transparency by making the net effect on the tax subsidized equity reaching the low- income community business hard to determine. Officials at the CDFI Fund are attempting to address this issue by requiring CDEs to provide a disclosure statement to low-income community businesses about the size of tax subsidized equity and how it is affected by fees and interest rates.[Footnote 16] The new requirement that CDEs disclose to the low-income community businesses all transactions costs, fees, and compensation could help those businesses understand the final net benefit to the project being financed with NMTCs. However, because Treasury does not require the CDFI Fund to collect the CDE disclosure statements itself, the CDFI Fund database has incomplete information about fees, interest, and other costs. Without such complete information the Treasury is limited in its ability to analyze the final net financial benefit of NMTC investments to low-income community businesses. CDE Fees and Retentions Reduce the NMTC Equity Available to the Low- Income Community Businesses: Our analysis shows that fees and retentions by the CDEs reduced the $8.8 billion of NMTC investment available to the businesses in 2011- 2012 by about $619 million or 7.1 percent. The initial reduction occurred as part of the NMTC investment that the CDEs retain to cover administrative costs before investing the remainder in the project. The CDEs then also charge fees over the course of the 7-year compliance period that further reduces the equity available to the project. These fees can take the form of front-end or origination fees at closing, on-going, or asset management fees during the compliance period, and closing fees at the end of the compliance period. shows the fees and retentions measured as a percentage of NMTC investment that reduce the equity that is available to the businesses. In addition, the projects may also incur third-party transaction costs for NMTC related accounting or legal services not provided by the CDEs. The CDEs are not required to report these other third-party transaction costs to the CDFI Fund. Table 2: New Markets Tax Credit (NMTC) Investment Retained and Fees Charged by Community Development Entities (CDEs), 2011-2012: Fee Types: NMTC investment retained by CDEs; Average fees: $0.2 million; Total fees for all projects: $170 million; Total fees as a percent of total NMTC investment[A]: 1.9%. Fee Types: CDE professional fees charged at or prior to close (front- end); Average fees: $0.2 million; Total fees for all projects: $183 million; Total fees as a percent of total NMTC investment[A]: 2.1%. Fee Types: CDE professional fees charged during compliance period (on- going); Average fees: $0.2 million; Total fees for all projects: $198 million; Total fees as a percent of total NMTC investment[A]: 2.3%. Fee Types: CDE professional fees charged at or after investment exit (back-end); Average fees: $0.1 million; Total fees for all projects: $67 million; Total fees as a percent of total NMTC investment[A]: 0.8%. Fee Types: CDE professional fees charged, Subtotal; Average fees: $0.6 million; Total fees for all projects: $448 million; Total fees as a percent of total NMTC investment[A]: 5.1%. Fee Types: Total; Average fees: $0.8 million; Total fees for all projects: $619 million; Total fees as a percent of total NMTC investment[A]: 7.1%. Source: GAO analysis of CDFI Fund data. GAO-14-500. Notes: Components may not sum to totals due to rounding. Dollar amounts are not adjusted for inflation. The fees contain both actual and projected amounts. The earlier in the project's history the fees are paid, the more likely the payments are to be actual rather than planned. The totals and subtotals include $2,000 in average fees and $1.4 million in total fees in a fourth category of CDE professional fees called "other" fees by the CDFI Fund not listed separately in this table. [A] The total qualified equity investment (QEI) for all projects originating in 2011-2012. [End of table] Higher CDE Fees and Retentions Are Associated with More Complex Financial Structures: Our analysis also shows that the amount of fees and retentions charged is strongly associated with the amount of NMTC investment in the project: the amount of NMTC investment accounts for about 50 percent of the variation in fees across projects. Although the CDFI Fund does not currently collect data that directly measure the complexity of investment structures, it does collect data on the number of investment transactions that occur on a NMTC project. These transactions often represent loans and investments to the business from different entities such as multiple CDEs, and can therefore be used as a proxy for more direct measures of complexity. Our regression analysis of these data shows that higher fees and retentions are associated with more complex structures as indicated by the number of transactions (see table 3 for details of this association between fees and retentions and financial structure, and other characteristics of the NMTC project). Fees and retentions are generally described by NMTC participants as reflecting mostly fixed costs, which would be more consistent with fees decreasing in proportion to the total size of the NMTC investment. Table 3: Project Characteristics Associated with Fees and Retentions Charged by Community Development Entities (CDEs), 2011-2012: Project characteristics: Amount of NMTC qualified equity investment; Effect on fees and retentions: Increased; Size of the effect: Change in dollar value of fees and retentions: $7,800 per $100,000 of additional NMTC-qualified equity investment. Project characteristics: Number of transactions as a proxy for financial structure complexity; Effect on fees and retentions: Increased; Size of the effect: Change in dollar value of fees and retentions: $4,800 per additional transaction. Project characteristics: Average interest rate; Effect on fees and retentions: Increased; Size of the effect: Change in dollar value of fees and retentions: $9,600 per 1.0 percentage point increase in the interest rate on NMTC- related loans to the project. Project characteristics: Purpose: business; Effect on fees and retentions: Increased; Size of the effect: Change in dollar value of fees and retentions: $29,000 more in fees and retentions charged on transactions for business purposes than those for rehabilitation, or other purposes. Project characteristics: Amount of non-public funds used on project; Effect on fees and retentions: Decreased; Size of the effect: Change in dollar value of fees and retentions: $90 less in fees and retentions per $100,000 of private funds invested in the project. Project characteristics: Poverty rate greater than 30 percent indicating distressed low-income community; Effect on fees and retentions: Decreased; Size of the effect: Change in dollar value of fees and retentions: $3,000 less in fees and retentions for projects in communities with poverty rates exceeding 30 percent. Legend: NMTC = New Markets Tax Credit. Source: GAO analysis of CDFI Fund data. GAO-14-500. Notes: The table summarized results for selected variables of a regression analysis of the relationship of total fees and retentions with other project characteristics. [End of table] CDEs That Reported They Charged No Fees or Retentions Illustrates the Program's Lack of Transparency: For about 20 percent of projects originating in 2011-2012, CDEs reported to the CDFI Fund that they charged zero fees or retentions on these projects (see figure 5). However, it seems unlikely that the services provided by the CDEs were uncompensated. While the payment may have come in different forms, a lack of transparency makes it hard to readily determine how much of the NMTC investment is being reduced and by what means. For example, NMTC program participants have suggested that these projects with no fee costs may be more integrated investments where banks or other large institutions may be lenders as well as investors. In this case, low fees may be offset by higher interest rates. We found some indirect evidence of this in another regression where the projects with no fees or retentions were analyzed relative to all the projects that had fees and retentions. Here, the analysis showed that the projects with higher average interest rates were more likely to charge no fees and retentions. For this subgroup of projects, the positive relationship between fees and interest rates that we found when we analyzed all the projects (as shown in table 3) is reversed. Figure 5: Number of Projects by Total Retentions and Fees Charged as a Percentage of Total New Markets Tax Credit (NMTC) Investment, 2011- 2012: [Refer to PDF for image: vertical bar graph] Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 0%; Number of projects: 142. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 1%; Number of projects: 42. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 2%; Number of projects: 15. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 3%; Number of projects: 27. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 4% Number of projects: 46. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 5%; Number of projects: 38. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 6%; Number of projects: 43. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 7%; Number of projects: 81. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 8%; Number of projects: 78. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 9%; Number of projects: 80. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 10%; Number of projects: 76. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 11%; Number of projects: 25. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 12%; Number of projects: 7. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 13%; Number of projects: 19. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 14%; Number of projects: 14. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 15%; Number of projects: 12. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 16%; Number of projects: 5. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 17%; Number of projects: 1. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 18%; Number of projects: 2. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 19%; Number of projects: 3. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 20%; Number of projects: 2. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 21%; Number of projects: 8. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 22%; Number of projects: 2. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 23%; Number of projects: 7. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 24%; Number of projects: 0. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: 25%; Number of projects: 1. Total Retentions and Fees Charged by CDEs as Percent of NMTC Investment: More than 25%; Number of projects: 18. Source: GAO analysis of CDFI Fund data. GAO-14-500. [End of figure] Data on NMTC Equity Remaining in the Low-Income Community Businesses Are Not Sufficiently Complete or Accurate: Projects that have considerable equity are more likely to have better loan-to-value ratios and are generally more likely to obtain loans with better terms than projects without their own equity. For this reason, the larger the amount of equity remaining in the project, the greater is the likelihood that the project will continue on its own without any further government subsidies. However, the data available from the CDFI Fund reflect only the equity left by NMTC investors and may not give a complete picture of the economic viability of the business because it does not include other forms of equity, such as those from retained earnings for a successful business. Furthermore, the CDFI data on equity left in the business are not sufficiently reliable because they are incomplete and not accurate enough to capture program performance. According to our standards for measuring program performance, several elements should be considered when examining the quality of agency performance data including accuracy and completeness. However, our review showed that about 60 percent of projects originating in 2011-2012 had inconsistencies that made these data unreliable. Examples of these inconsistencies are: * Incomplete data: One or more CDEs involved in the project did not report any values making it impossible to calculate an amount for the entire project. * Inaccurate data: Equity remaining was projected to equal or exceed 100 percent of all NMTC investments in the project. These amounts are not valid because they exceed the amounts of the original equity loan. * Zero values: Some CDEs may be reporting a zero value because they do not intend to leave equity in the project. But according to CDFI Fund officials, other CDEs may intend to leave equity in the project, but reported zero value for accounting reasons because the charge would not be recorded until a later date. Thus, the zero values may be understating the equity available to the low-income community businesses. We determined that one cause of data unreliability was the unclear instructions in the manual for entering data into the CDFI Fund's systems. The manual does not clearly explain the time period for which information should be reported, which may have led to the CDEs reporting according to different accounting rules. Incomplete and inaccurate data result in an inability to use the data to track an important indicator of the likely performance of the NMTC projects after the compliance period ends. As a result, it is not possible to determine from these data the amount of equity to remain in the low- income community businesses after the 7-year credit period. Figure 6: Data Quality for Amounts of Equity Projected to Remain in the Low-Income Community Businesses, 2011-2012: [Refer to PDF for image: vertical bar graph] Incomplete data: Number of projects: 201. Inaccurate data: Number of projects: 135. Zero values: Number of projects: 143. Expected values: Number of projects: 363. Source: GAO analysis CDFI Fund data. GAO-14-500. Note: CDFI Fund does not require data on equity remaining for any loans of investments originating before December 31, 2010. The figure is based on 842 projects with loans or investments originating in 2011- 2012, the most recent available data. [End of figure] CDFI Fund Data on Distressed Projects Are Not Sufficiently Reliable: CDFI Fund data on projects experiencing financial distress, such as the number of days a loan is delinquent, track program performance in that they indicate how likely a project is to continue in business during and after the credit period. These indicators of financial distress must be weighed against other NMTC program goals. Another program goal is to encourage investors to invest in projects that may be more risky because they are located in low-income communities. Regardless, reliable performance data are needed to administer the program. Our review of CDFI Fund data on the current status and performance of loans to NMTC projects showed inconsistencies that made these data not sufficiently reliable to determine the number and extent of projects experiencing financial distress. Examples of these inconsistencies include: * Incomplete data: Ninety-nine percent of projects reported current loan status (a mandatory field), but approximately 30 percent of projects omitted additional information for the other potential indicators of distress such as the number of times a loan has been restructured, the number of days the loan is currently delinquent, and the dollar amount of any loan that has been charged off (optional or conditionally-required fields). * Inaccurate data: Potential inaccuracies appear when we compared CDEs' descriptions of troubled projects from their NMTC allocation applications with indicators of distress for those same projects in CDFI Fund databases. In their 2012 applications, CDEs described 193 projects with delinquent, defaulted, or impaired loans in sufficient detail that we could identify those projects in the CDFI Fund data. But 49 of these projects showed no indications of distress in the data. The causes of data unreliability were unclear instructions and optional reporting. CDEs enter their data into the CDFI Fund's electronic database. How they enter the data is determined by instructions provided by the CDFI Fund. We found a lack of clarity in these instructions that prevented us from being confident that the data provided an accurate measure of distress. For example, the instructions did not clearly distinguish restructured loans from refinanced loans. However, refinancing may not be an indicator of distress. It can occur for a variety of economic reasons. In addition, the data were incomplete because reporting some information was optional. Without accurate and complete data, the CDFI Fund does not have sufficient information to track program performance related to the future viability of the NMTC funded projects. CDFI Fund officials told us that they are changing all the loan performance data points to make them mandatory. Difficulties in Determining Whether or Not a Project Has Failed: Project failures could significantly affect program performance by limiting its social and economic outcomes in low-income communities. However, determining when a project has failed is difficult. Projects that seem to be in difficulty based on the indicators of financial distress can become financially sound again. Some CDEs in their 2012 applications described successful restructurings of projects experiencing financial difficulty. But other projects are in situations where recovery seems unlikely or the CDEs have in fact written off the projects (see text boxes 1 and 2 for details of examples of both types of outcomes). The CDFI Fund is developing additional tools to collect better information on failed projects. As discussed above, their current measures of financial distress are inaccurate or incomplete. However, even if the data are improved, the financial distress measures may not accurately identify project failures because the projects can recover from distress. The CDFI Fund is attempting to rectify this measurement problem with a close-out report which is intended to collect additional information on the status of the business at the end of the 7-year compliance period, such as whether a business continues to operate or a real estate project has been put into service. Examples Illustrating Difficulties in Determining if a Project Has Failed: Text Box 1: Meat processing plant investment that failed. * An $8 million NMTC investment in a meat processing company involving two loans and one equity investment from one CDE; * According to news accounts, the investments were needed to address liquidity problems created by an expansion of this ongoing business that employed about 300 people; * According to the CDE's 2012 reapplication, the investments became impaired soon after the CDE made the investments when the leveraged lender limited the company's ability to borrow from its credit line, which forced the company into bankruptcy; * The CDE unsuccessfully sought alternative financing for the company but still suffered a loss of $5.2 million; * In the CDFI Fund data, one loan and one equity investment are reported as "charged off" for the full dollar amounts. The other loan is reported as "closed" but no amount charged off. Source: GAO analysis of CDFI Fund data and public sources. GAO-14-500. [End of text box] Text Box 2: Troubled college construction project that recovered: * A college construction project involving three CDEs and several NMTC loans; * In their 2012 reapplications for NMTC allocations, two of the CDEs reported that their loans for the project became delinquent for as much as 480 days because of construction delays and the economic recession of 2008; * The two CDEs reported that they worked with the borrower to rebalance the budget and bring in additional sources of financing; * At the time of their applications, the two CDEs reported that all the previous interest was paid and all subsequent interest payments had been on time; * The loans from the two CDEs show indications of poor performance (periods of delinquency) in CDFI Fund data; * The third CDE did not report any delinquent or impaired loans in its 2012 reapplication, and its loans do not show any indications of poor performance in CDFI Fund data. Source: GAO analysis of CDFI Fund data. GAO-14-500. [End of table] Conclusions: The potential impact of the NMTC in promoting economic development in designated low-income communities is diluted if the NMTC provides an above-market rate of return. Similarly, the impact of a combination of assistance from government programs is diluted if in the same cases the combination of assistance is unnecessarily duplicative. Treasury guidance and controls that are designed to limit these risks can help ensure the NMTC program realizes the greatest possible impact on low- income communities. Complete and reliable information is a vital component of assessing program effectiveness. While the complexity of the NMTC financial structures makes gathering information a challenge, there are several aspects of these structures where better information would aid in understanding the effectiveness of the program. These include the extent to which fees, interest rates, and other costs reduce the NMTC equity flowing to low-income community businesses, the amount of equity available to the low-income community businesses at the end of the 7-year compliance period, and the number of projects that failed or are at risk of failing. Recommendations for Executive Action: We recommend that the Secretary of the Treasury take the following actions: * Issue guidance on how funding or assistance from other government programs can be combined with the NMTC including the extent to which other government funds can be used to leverage the NMTC by being included in the qualified equity investment. * Ensure that controls are in place to limit the risk of unnecessary duplication at the project level in funding or assistance from government programs and to limit above market rates of return, i.e., returns that are not commensurate with the NMTC investor's risk. * Ensure that the CDFI Fund reviews the disclosure sheet that CDEs are required to provide to low-income community businesses to determine whether it contains data that could be useful for the Fund to retain. * Ensure that the CDFI Fund clarifies the instructions for reporting the amount of any equity which may be acquired by the low-income community business at the end of the 7-year NMTC compliance period. * Ensure that the CDFI Fund clarifies the instructions it provides to CDEs about reporting loan performance and make the reporting of that data mandatory. Agency Comments and Our Evaluation: We provided a draft of this product to Treasury for comment on June 11, 2014. In its written comments, reproduced in appendix IV, Treasury concurred with two of our recommendations and reported that the other recommendations are under consideration. The CDFI Fund also provided technical comments that were incorporated, as appropriate. Treasury said that it is considering our recommendations to issue further guidance on how other government programs can be combined with NMTCs, and to ensure that adequate controls are in place to limit the risks of unnecessary duplication and above-market rates of return. Treasury reported that our recommendations would be reviewed in consultation with a recently formed working group that includes representatives from the IRS and the CDFI Fund to discuss potential administrative or regulatory changes. Treasury said that it is considering our recommendation about reviewing data presented in the disclosure sheets that CDEs are required to provide low-income community businesses. As our report states, not all information on the disclosure sheets, such as third- party transactions costs, is reported to the CDFI Fund. This additional information on the disclosure sheets could be useful for the CDFI Fund to retain. Treasury agreed with our recommendation to clarify instructions for reporting any equity amounts that may be acquired by the low-income community business at the end of the compliance period. Treasury also agreed with our recommendation to clarify instructions to CDEs about reporting loan performance and make this data reporting mandatory. As agreed with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies to the appropriate congressional committees, the Director of the CDFI Fund, and other interested parties. In addition, the report will be available at no charge on the GAO website at [hyperlink, http://www.gao.gov]. If you or your staff have questions about this report, please contact me at (202) 512-9110 or at whitej@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix V. Sincerely yours, Signed by: James R. White: Director, Tax Issues: Strategic Issues: [End of section] Appendix I: Objectives, Scope, and Methodology: The objectives of this report were to assess: (1) New Markets Tax Credit (NMTC) financial structures in terms of their complexity, transparency, and effect on the size of the federal subsidies going to NMTC projects as well as controls to ensure that subsidies are not larger than necessary for the investment; (2) what is known about the types and amounts of fees and other costs that reduce the amount of equity reaching low-income community businesses; (3) what is known about the amount of equity left in the low-income community businesses after the 7-year credit period; and (4) what is known about NMTC projects that are at risk of failing by becoming economically nonviable. For our first objective, we reviewed the NMTC literature and interviewed representatives of community development entities (CDEs) and researchers who have evaluated the program to determine how the financial structures have evolved. To assess the complexity and transparency of NMTC investment structures, we applied criteria from our prior work on evaluating tax expenditures.[Footnote 17] We also applied criteria from federal government internal control standards to assess whether controls are present to ensure that subsidies are not larger than necessary for a NMTC project.[Footnote 18] To report on the number, types, and funding amounts of other federal programs used on NMTC projects, we designed and implemented a web- based survey to gather information on how projects were financed from the CDEs responsible for the project. Our survey population consisted of randomly selected NMTC projects with all loans and investments closing on or after January 1, 2010. Restricting the survey population to NMTC projects within our study's time period left us with a total of 1,265 projects in the population. We selected a stratified sample of 305 projects. From the first strata, defined as those projects for which there is an indication on the underlying population file that at least one of the funding sources was public dollars, we selected 126 projects. From the second strata, defined as all other projects, we selected 179 projects. Although some projects in the second strata could actually have public dollars, this stratification helps ensure that our sample has enough of these cases to produce estimates of that domain. The survey asked the CDEs a combination of questions that allowed for open-ended and close-ended responses with regard to federal, state, local, and private funding sources. We pre-tested the content and format of the questionnaire with four knowledgeable CDEs and made changes based on pre-test results. The survey was a web-based survey. We sent an activation e-mail for the survey on March 7, 2014, and closed the survey on April 4, 2014. The practical difficulties of conducting any survey may introduce errors, commonly referred to as non-sampling errors. For example, differences in how a particular question is interpreted, the sources of information available to respondents, how the responses were processed and analyzed, or the types of people who do not respond can influence the accuracy of the survey results. We took steps in the development of the survey, the data collection, and the data analysis to minimize these non-sampling errors and help ensure the accuracy of the answers that were obtained. A second independent analyst checked all the computer programs that processed the data. In instances where multiple CDEs responded about the same project, we manually merged the data together into a representative single project following the Community Development Financial Institutions (CDFI) Fund's practices, such as reporting the highest value as the default estimate. We obtained responses for 214 of the 305 projects in our sample for an overall response rate of about 70 percent. Population estimates were produced by weighting the sample data from the responding projects to account for differing sampling rates for projects funded with public dollars and those funded without public dollars. We have treated the respondents as a stratified random sample and calculated sampling errors as an estimate of the uncertainty around the survey estimates. All percentage estimates based on this sample have 95 percent confidence intervals of within +/-7 percentage points of the estimate itself. For other numeric estimates, the 95 percent confidence intervals are presented along with the estimates themselves. We are 95 percent confident that each of the confidence intervals in this report will include the true values in the study population. To address the last three objectives, we analyzed data from the CDFI Fund on NMTC investments in low-income community businesses from 2003 through 2012. The CDFI Fund requires all CDEs that have been awarded NMTC allocations to submit an annual report detailing how they invested the qualified equity investment (QEI) proceeds in low-income communities. These reports must be submitted to the CDFI Fund by the CDEs, along with their audited financial statements, within 6 months after the end of their fiscal years. CDEs are required to report their NMTC investments in the CDFI Fund's Community Investment Impact System (CIIS) for a period of 7 years. Due to a time lag in reporting, NMTC investments reported in CIIS are less than the total amount allocated for the NMTC program.[Footnote 19] Given that the CDFI Fund requires CDEs to report information on project characteristics and financing once a year, CIIS data may not capture the most current information for all existing projects. However, the CIIS data that we used represent the most current available information as of December 2, 2013 on the status of the program. We interviewed CDFI Fund officials with knowledge of CIIS about the steps they take to ensure its accuracy. Based on GAO's criteria for valid data--that they be sufficiently accurate and complete to capture program performance--we also determined the data on fees and retentions we used in this report were sufficiently reliable for our purposes (see below for discussion of the data we found unreliable for our purposes).[Footnote 20] To describe the types and amounts of fees and other costs, we analyzed data that CDEs reported to the CDFI Fund on the NMTC investment proceeds they retained and fees they charged to investors, low-income community businesses, and others. We analyzed the fee data at the project level for 794 projects with loans and investments originating in 2011 and 2012. The CDFI Fund began requiring that CDEs report data on QEI proceeds that the CDEs retained and their other fees starting with the 2011 reporting period, and only for loans and investments originating after December 31, 2010.[Footnote 21] We used fees data reported in basis points and data on the qualified low-income community investment (QLICI) amounts to calculate the total dollar amounts of fees. We compared our final results to other data on fees submitted by CDEs when applying for NMTC allocations in 2010 and 2011. We attempted to compare these costs to costs on non-NMTC investments by reviewing industry data and academic studies on fees, and interviewing industry and tax credit experts. We also reviewed written and oral comments made by CDEs and industry experts to the CDFI Fund in response to a November 7, 2011, Federal Register notice soliciting comments on several possible NMTC program changes. The CDFI Fund specifically requested comments on whether additional rules, restrictions, and requirements should be imposed related to fees and expenses charged by CDEs. The CDFI Fund also held three listening sessions in December 2012 and January and February 2013 with a total of 45 CDE and industry experts. We reviewed the transcripts of these listening sessions and concluded from these reviews that the types of projects funded by the NMTC are so varied that we could not conduct a valid comparison of fee costs with those on non-NMTC projects. We attempted to determine the amount of equity that remains in low- income community businesses by analyzing data reported by CDEs to the CDFI Fund on equity projected to remain in NMTC projects at the end of the 7-year credit period. We analyzed these data at the project level for 842 projects with loans and investments originating in 2011 and 2012. Starting with the 2011 reporting period, the CDFI Fund began requiring that, if applicable, CDEs report the projected amount of any equity or debt investment which may be acquired by the low-income community business as the result of a put/call option or other arrangement for loans and investments originating after December 31, 2010. After reviewing all 2,249 transactions for all 842 projects, we determined that data were usable for only 363 projects (about 40 percent), and therefore the data could not be used to give sufficiently reliable descriptions of the equity remaining in the project for our purposes. We found the following problems with the data that made the values unreliable for our report. * For 201 projects, no data were reported by one or more of the CDEs involved in those projects. We concluded that in these cases we could not determine: (1) how many of these projects were non-leveraged where the failure to project a residual value might be expected; (2) how many indicated a true intention not to leave any equity in the project; and (3) how many were simply errors or omissions. * For another 143 projects, one or more CDEs involved reported a total projected residual value of $0. We concluded in these cases that the implications of zero for equity remaining were ambiguous. According CFDI Fund officials, some CDEs may in fact do not intend to leave any residual equity for the business to obtain at the end of the 7-year period. However, other CDEs may project $0 remaining now when in fact they intend to designate the amount of equity at a later date. These CDEs may be reporting a zero value at this time due to an individual CDE's internal practices or accounting rules particular to the CDE's form of incorporation. That is, the CDEs would not report a projected or final residual value until the put/call option was exercised by the business. * Of the remaining 498 projects, 135 projects had data showing that the total projected residual value for these projects appeared to be overstated. Some data showed that one or more CDEs reported projected residual values greater than or equal to the original equity investment values. In most of these cases, the CDEs appeared to have incorrectly reported a projected residual value for both leveraged loans and equity investments. Based on the typical leveraged model structure, the CDEs should have only reported the residual value of the equity investments. In other cases, one or more CDEs involved in a project reported the projected residual value of the equity investment twice--once as value for the equity investment, and then repeated it as a value for the leveraged loan. As a result of these reporting errors or inconsistencies, the total projected residual value for these projects appeared to be overstated. To determine the NMTC projects at risk of failing by becoming economically nonviable, we analyzed data reported by CDEs to the CDFI Fund that could indicate that a project is experiencing financial distress. We used indicators of financial distress that are available from CDFI Fund data, such as whether a loan on a project is delinquent, charged off, or restructured that could show increased risk of business failure. These indicators, in most cases, could not be used to conclude that a business has failed in the sense of being economically nonviable as the CDFI Fund does not currently have data on the ultimate disposition of NMTC projects. We analyzed the number of projects that showed indications of financial distress between 2003 and 2012, and the dollar amounts invested in these financially distressed projects. We tested the reliability of these optional fields by reviewing CDEs applications for 2012 NMTC allocations. In the application instructions, CDEs were asked to discuss any delinquent, defaulted, or impaired loans or equity investments from prior NMTC investments. In the 2012 applications, we counted 281 projects with delinquent, defaulted, or impaired loans or investments. Of these, 193 projects were described in sufficient detail that we could then match those projects to transaction-level data in CIIS. However, 49 of those projects did not show any indications of financial distress in CIIS. For some projects involving multiple CDEs, one or more CDEs may have described the project in their 2012 applications as having delinquent, defaulted, or impaired investments, but only one CDE then reported any of the optional distress indicator data in CIIS. In other cases, CDEs described several projects in the 2012 applications as having delinquent, defaulted, or impaired investments, but then those CDEs did not report any distress indicators for these projects in CIIS. In the end, we concluded that the CDFI Fund CIIS data on indicators of financial distress were insufficient for our purposes, largely due to the fact that nearly all of the distress indicators were optional data fields. We conducted this performance audit from May 2013 to July 2014 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. [End of section] Appendix II: Other Federal Funding Used in New Markets Tax Credit Projects: In our survey of New Markets Tax Credit (NMTC) projects, we asked Community Development Entities (CDEs) what other federal sources of assistance were used (in addition to the NMTC) to fund the project. [Footnote 22] CDEs could select from a pre-populated list of federal sources comprised of federal tax credits, public bond financing exempt from federal tax, grants or direct payments from federal agencies, and direct or guaranteed loans from federal agencies. Our survey also permitted CDEs to write in other federal sources not included in the pre-populated list. CDEs were also asked if the NMTC was "twinned" or "enhanced" with any of these federal sources inside the NMTC structure, i.e., did NMTC investors claim NMTCs based on these additional amounts of federal assistance. The NMTC investor community often refers to this practice as "twinning" the NMTC with other tax credits or other public assistance, but for purposes of this report, we define this as leveraging other public sources with the NMTC. The four tables below list the types of other federal assistance (from our pre-populated list of federal assistance and write-in descriptions) that CDEs reported as being used to finance our sample of NMTC projects, and whether these other federal sources were leveraged with the NMTC. Table 4: Other Federal Tax Credits Used in a Sample of New Markets Tax Credit (NMTC) Projects, 2010-2012: 10 percent credit for rehabilitation of structures (other than historic): Federal source was leveraged with NMTC one or more times: [Empty]. 20 percent credit for rehabilitation of historic structures (Historic Tax Credit): Federal source was leveraged with NMTC one or more times: yes. Business Energy Investment Tax Credit (ITC): Federal source was leveraged with NMTC one or more times: yes. Credit for qualified school construction bonds; Federal source was leveraged with NMTC one or more times: yes. Brownfield credits: Federal source was leveraged with NMTC one or more times: [Empty]. Source: GAO survey of NMTC projects originating in 2010-2012. GAO-14- 500. [End of table] Table 5: Tax Exempt Bond Financing Used in a Sample of New Markets Tax Credit (NMTC) Projects, 2010-2012: Exclusion of interest for airport, dock, and similar bonds: Federal source was leveraged with NMTC one or more times: [Empty]. Exclusion of interest on bonds for private nonprofit educational facilities: Federal source was leveraged with NMTC one or more times: yes. Exclusion of interest on bonds for private nonprofit hospital facilities: Federal source was leveraged with NMTC one or more times: yes. Exclusion of interest on Gulf Opportunity Zone Revenue Bonds: Federal source was leveraged with NMTC one or more times: [Empty]. Exclusion of interest on Recovery Zone Bonds: Federal source was leveraged with NMTC one or more times: yes. Federal Energy 179D (tax deduction): Federal source was leveraged with NMTC one or more times: [Empty]. Various public purpose state and local bonds[A]: Federal source was leveraged with NMTC one or more times: yes. Source: GAO survey of NMTC projects originating in 2010-2012. GAO-14- 500. [A] A number of survey respondents indicated receiving project financing from a variety of state and locally-issued bonds. [End of table] Table 6: Grants or Direct Payments Used in a Sample of New Markets Tax Credit (NMTC) Projects, 2010-2012: Department of Housing and Urban Development (HUD): Federal source was leveraged with NMTC one or more times: [Empty]. Community Development Block Grants/Entitlement Grants (CDBG): Federal source was leveraged with NMTC one or more times: yes. HUD Economic Development Initiative: Federal source was leveraged with NMTC one or more times: [Empty]. HUD Financial Grant: Federal source was leveraged with NMTC one or more times: [Empty]. HUD Capital Fund Education and Training Community Facilities Program: Federal source was leveraged with NMTC one or more times: yes. Neighborhood Stabilization Program (NSP): Federal source was leveraged with NMTC one or more times: yes. Public Housing Development Funds: Federal source was leveraged with NMTC one or more times: yes. Department of Energy: Federal source was leveraged with NMTC one or more times: [Empty]. Energy Efficiency and Conservation Block Grant (EECBG): Federal source was leveraged with NMTC one or more times: [Empty]. Department of Health and Human Services: Federal source was leveraged with NMTC one or more times: [Empty]. Grants to Health Center Programs: Federal source was leveraged with NMTC one or more times: yes. Office of Head Start (OHS): Federal source was leveraged with NMTC one or more times: [Empty]. Department of the Treasury: Federal source was leveraged with NMTC one or more times: [Empty]. 1603 Program: Payments for Specified Energy Property in Lieu of Tax Credits: Federal source was leveraged with NMTC one or more times: [Empty]. U.S. Treasury - Neighbor Works: Federal source was leveraged with NMTC one or more times: [Empty]. Department of Education: Federal source was leveraged with NMTC one or more times: [Empty]. Impact Aid grant: Federal source was leveraged with NMTC one or more times: yes. Department of Education unspecified grant: Federal source was leveraged with NMTC one or more times: [Empty]. Department of Defense: Federal source was leveraged with NMTC one or more times: [Empty]. Department of Defense unspecified grant: Federal source was leveraged with NMTC one or more times: [Empty]. Environmental Protection Agency (EPA): Federal source was leveraged with NMTC one or more times: [Empty]. EPA Grant: Federal source was leveraged with NMTC one or more times: [Empty]. Federal Communications Commission (FCC): Federal source was leveraged with NMTC one or more times: [Empty]. FCC Grants for Rural Healthcare: Federal source was leveraged with NMTC one or more times: [Empty]. Federal Home Loan Bank: Federal source was leveraged with NMTC one or more times: [Empty]. Federal Home Loan Bank grants: Federal source was leveraged with NMTC one or more times: [Empty]. Department of Veterans Affairs (VA): Federal source was leveraged with NMTC one or more times: [Empty]. VA tenant improvement payment: Federal source was leveraged with NMTC one or more times: [Empty]. Unspecified: Federal source was leveraged with NMTC one or more times: [Empty]. Rural Community Assistance Partnership (RCAP): Federal source was leveraged with NMTC one or more times: [Empty]. American Recovery Reinvestment Act (ARRA): Federal source was leveraged with NMTC one or more times: [Empty]. Source: GAO survey of NMTC projects originating in 2010-2012.GAO-14- 500. [End of table] Table 7: Direct Loans or Guaranteed Loans Used in a Sample of New Markets Tax Credit (NMTC) Projects, 2010-2012: Department of Agriculture: Federal source was leveraged with NMTC one or more times: [Empty]. Business and Industry Loans: Federal source was leveraged with NMTC one or more times: [Empty]. Community Facilities Loans: Federal source was leveraged with NMTC one or more times: [Empty]. Department of Housing and Urban Development: Federal source was leveraged with NMTC one or more times: [Empty]. Section 108 Community Development Loan Guarantees: Federal source was leveraged with NMTC one or more times: [Empty]. Supplemental Loan Insurance Multifamily Rental Housing: Federal source was leveraged with NMTC one or more times: [Empty]. Small Business Administration (SBA): Federal source was leveraged with NMTC one or more times: [Empty]. SBA 7(a) Guaranteed Loan: Federal source was leveraged with NMTC one or more times: [Empty]. SBA 504 loan program: Federal source was leveraged with NMTC one or more times: [Empty]. SBA unspecified guaranteed loan: Federal source was leveraged with NMTC one or more times: [Empty]. Department of Education: Federal source was leveraged with NMTC one or more times: [Empty]. Charter School Credit Enhancement Grant: Federal source was leveraged with NMTC one or more times: [Empty]. Department of Energy: Federal source was leveraged with NMTC one or more times: [Empty]. ARRA Energy Loan Program--Green Energy Loan Fund: Federal source was leveraged with NMTC one or more times: [Empty]. Other: Federal source was leveraged with NMTC one or more times: [Empty]. Department of Housing and Community Development (state) CDBG-ARRA: Federal source was leveraged with NMTC one or more times: yes. Source: GAO survey of NMTC projects originating in 2010-2012. GAO-14- 500. [End of table] [End of section] Appendix III: Regression Analysis of Total Fees and Retentions Associated with Characteristics of the Project Using CDFI Fund Data, 2011-2012: Variable Name: Average interest rate on loans; Coefficient: .0968962; Std. Error: .0360374; t: 2.69; P>|t|: 0.007. Variable Name: Dollar amount of NMTC qualified equity investment (QEI); Coefficient: .0782438; Std. Error: .0014382; t: 54.40; P>|t|: 0.000. Variable Name: Number of transactions (financial structure complexity); Coefficient: .0482567; Std. Error: .0189266; t: 2.55; P>|t|: 0.011. Variable Name: Poverty rate > 30% (distressed community); Coefficient: -.302405; Std. Error: .1086737; t: -2.78; P>|t|: 0.005. Variable Name: Dollar amount of non-public funds used on project; Coefficient: -.0009428; Std. Error: .0002418; t: -3.90; P>|t|: 0.000. Variable Name: Purpose: business; Coefficient: .2869305; Std. Error: .1392813; t: 2.12; P>|t|: 0.040. Variable Name: Purpose: construction; Coefficient: -.0928775; Std. Error: .1401169; t: -0.66; P>|t|: 0.507. Variable Name: Origination year: 2012; Coefficient: -.1675759; Std. Error: .1076841; t: -1.56; P>|t|: 0.120. Variable Name: Constant; Coefficient: -.5996888; Std. Error: .1900319; t: -3.16; P>|t|: 0.002. Source: GAO analysis of CDFI Fund data. GAO-14-500. Notes: N=2,018; R-squared = 0.6028. Dollar amounts are scaled in $100,000 units. [End of table] The table reports the results of an ordinary least squares regression with the dependent variable equal to the amount of fees and retentions charged. Omitted category variables are: project type equal to "rehabilitation or other" and origination year equal to "2011." The regression was also estimated using various functional forms including quadratic and log forms of the regression equation. These specifications either were rejected as statistically insignificant or, in the case of some of the log specifications, resulted in a substantial decrease in the explanatory power of the regression as measured by its R-squared. This regression was also estimated with fees and retentions as separate dependent variables, and, in these cases, the results were different for certain variables of interest. With fees as the dependent variable, neither the interest rate nor the number of transactions was statistically significant at the 95 percent confidence level. However, the size of the qualified equity investment remained positive and significant. With retentions as the dependent variable, the retentions were positively and significantly related to interest rates but had no statistically significant relationship to the number of transactions. [End of section] Appendix IV: Comments from the Department of the Treasury: Department of The Treasury: Community Development Financial Institutions Fund: Washington, D.C. 20220: Memorandum: Date: July 1,2014: To: James R. White: Director, Tax Issues: Government Accountability Office: From: [Signed by Jeff Berg for] Dennis Nolan: Acting Director: Community Development Financial Institutions Fund: Subject: Draft Audit Report - New Markets Tax Credit: Better Controls and Data Are Needed to Ensure Effectiveness (GAO-14-500): On June 11, the Government Accountability Office (GAO) issued a report entitled "New Markets Tax Credit: Better Controls and Data Are Needed to Ensure Effectiveness." The New Markets Tax Credit (NMTC) Program is administered by the Community Development Financial Institutions Fund (CDFI Fund). The draft report provided the following recommendations to the Secretary of the Treasury: 1. Issue guidance on how funding or assistance from other government programs can be combined with the NMTC including extent to which other government funds can be used to leverage the NMTC by being included in the qualified equity investment. 2. Ensure that controls are in place to limit the risk of unnecessary duplication at the project level in funding or assistance from government programs and to limit above market rates of return, i.e. returns that are not commensurate with the NMTC investor's risk. 3. Ensure that the CDFI Fund reviews the disclosure sheet that Community Development Entities (CDEs) are required to provide to low- income community businesses to determine whether it contains data that could be useful for the CDFI Fund to retain. 4. Ensure that the CDFI Fund clarifies the instructions for reporting the amount of any equity which may be acquired by the low-income community businesses at the end of the 7 -year NMTC compliance period. 5. Ensure that the CDFI Fund clarify the instructions it provides to CDEs about reporting loan performance and make the reporting of that data mandatory. The CDFI Fund's responses to each recommendation are discussed below. Due to the limited time provided for the CDFI Fund to review the report's findings and methodology, specific comments on those items are not provided in this response letter. GAO Recommendation #1: Issue guidance on how funding or assistance from other government programs can be combined with the NMTC including extent to which other government funds can be used to leverage the NMTC by being included in the qualified equity investment. Auditee Response: Recommendation is Under Consideration. Management Response: The CDFI Fund recently formed a working group to discuss potential administrative or regulatory changes to the NMTC Program. The working group includes representatives from the Treasury Department's Internal Revenue Service; Office of Tax Policy; and Office of Small Business, Community Development and Affordable Housing. This recommendation, which considers broad policy issues, will be reviewed in consultation with the working group and other relevant Treasury officials. GAO Recommendation #2: Ensure that controls are in place to limit the risk of unnecessary duplication at the project level in funding or assistance from government programs and to limit above market rates of return, i.e. returns that are not commensurate with the NMTC investor's risk. Auditee Response: Recommendation is Under Consideration. Management Response: Assessing the depth of the subsidy necessary to mitigate risk and finance projects in low-income communities requires careful consideration of the purposes that various types of public support play in project implementation. In addition, it is important to consider how NMTC-financed projects leverage private funds to supply public goods and access to capital and credit in low-income communities that are chronically underserved by mainstream financial institutions. This recommendation will be reviewed in consultation with the working group discussed in our response to Recommendation #1, above. GAO Recommendation #3: Ensure that the CDFI Fund reviews the disclosure sheet that CommUnity Development Entities (CDEs) are required to provide to low-income community businesses to determine whether it contains data that could be useful for the CDFI Fund to retain. Auditee Response: Recommendation is Under Consideration. Management Response: The CDFI Fund currently collects most of the data contained in the disclosure sheet provided by CDEs to low-income community businesses through the Community Investment Impact System (CIIS), the CDFI Fund's data impact monitoring system. In 2010, in response to a prior GAO report (GAO-10-334), the CDFI Fund added thirteen questions to the CIIS which closely mirror the disclosure sheet provided to borrowers. In an effort to minimize the reporting burden for CDEs, the CDFI Fund is assessing technical approaches that would eliminate manual reconciliation of the data that it collects in the CIIS with the data reported in the disclosure sheets. In response to this recommendation, the CDFI Fund will consult with OMB with respect to the requirements of the Paperwork Reduction Act (PRA) to assess whether this recommendation implies collecting the same data twice: through the disclosure sheet that is provided by the CDE to the borrower; and through the data reported by the CDE to the CDFI Fund on each transaction's fees, interest and terms, and other costs. GAO Recommendation #4: Ensure that the CDFI Fund clarifies the instructions for reporting the amount of any equity which may be acquired by the low-income community businesses at the end of the 7-year NMTC compliance period. Auditee Response: Management concurs with the Recommendation. Management Response: The CDFI Fund will address this issue in the next version of the CIIS scheduled for release in August 2014. The CIIS update will implement a new close-out form at the end of the seven-year compliance period to collect data on the equity left in low-income businesses and status of businesses and projects. The close-out data will complement the current data collected on the "projected residual value left in the qualified business." In conjunction with the release of the new version of the CIIS, the CDFI Fund will provide additional guidance in the CIIS reporting instructions on completing the close-out form and conduct training sessions for users. GAO Recommendation #5: Ensure that the CDFI Fund clarify the instructions it provides to CDEs about reporting loan performance and make the reporting of that data mandatory. Auditee Response: Management concurs with the Recommendation. Management Response: The CDFI Fund is addressing this recommendation. The next version of the CIIS will require all loan performance data to be mandatory. In addition, the CDFI Fund will provide additional guidance in the CIIS reporting instructions that will be released with the new version of the CIIS in August 2014. Thank you again for the opportunity to review and comment upon your draft report. We appreciate your efforts and the collaborative relationship that you fostered during the course of your review. [End of section] Appendix V: GAO Contact and Staff Acknowledgments: GAO Contact: James R. White, (202) 512-9110, or whitej@gao.gov. Staff Acknowledgments: In addition to the contact named above, Kevin Daly, Assistant Director; Amy Bowser; Cathy Hurley; Mark Kehoe; Jill Lacey; Edward Nannenhorn; Mark Ramage; Wayne Turowski; and Elwood White made key contributions to this report. [End of section] Related GAO Products: 2014 Annual Report: Additional Opportunities to Reduce Fragmentation, Overlap, and Duplication and Achieve Other Financial Benefits. [hyperlink, http://www.gao.gov/products/GAO-14-343SP]. Washington, D.C.: April 8, 2014. Community Development: Limited Information on the Use and Effectiveness of Tax Expenditures Could Be Mitigated through Congressional Attention. [hyperlink, http://www.gao.gov/products/GAO-12-262]. Washington, D.C.: February 29, 2012. Efficiency and Effectiveness of Fragmented Economic Development Programs Are Unclear. [hyperlink, http://www.gao.gov/products/GAO-11-477R]. Washington, D.C.: May 19, 2011. Opportunities to Reduce Potential Duplication in Government Programs, Save Tax Dollars, and Enhance Revenue. [hyperlink, http://www.gao.gov/products/GAO-11-318SP]. Washington, D.C.: March 1, 2011. New Markets Tax Credit: The Credit Helps Fund a Variety of Projects in Low-Income Communities, but Could Be Simplified. [hyperlink, http://www.gao.gov/products/GAO-10-334]. Washington, D.C.: January 29, 2010. HUD and Treasury Programs: More Information on Leverage Measures' Accuracy and Linkage to Program Goals Is Needed in Assessing Performance. [hyperlink, http://www.gao.gov/products/GAO-08-136]. Washington, D.C.: January 18, 2008. Tax Policy: New Markets Tax Credit Appears to Increase Investment by Investors in Low-Income Communities, but Opportunities Exist to Better Monitor Compliance. [hyperlink, http://www.gao.gov/products/GAO-07-296]. Washington, D.C.: January 31, 2007. New Markets Tax Credit Program: Progress Made in Implementation, but Further Actions Needed to Monitor Compliance. [hyperlink, http://www.gao.gov/products/GAO-04-326]. Washington, D.C.: January 30, 2004. [End of section] Footnotes: [1] Pub. L. No.106-554, § 121, 114 Stat. 2763 (2000) codified at 26 U.S.C. § 45D. [2] See list of related GAO products at the end of this report. [3] For more detail on these criteria, see GAO, Tax Expenditures: Background and Evaluation Criteria and Questions, [hyperlink, http://www.gao.gov/products/GAO-13-167SP] (Washington, D.C.: Nov. 29, 2012); GAO, Understanding the Tax Reform Debate: Background, Criteria and Questions, [hyperlink, http://www.gao.gov/products/GAO-05-1009SP] (Washington, D.C.: Sept. 1, 2005); and GAO, Standards for Internal Control in the Federal Government, [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1] (Washington, D.C.: Nov. 1, 1999). [4] All estimates based on our survey are subject to sampling error. For percentage estimates, the 95 percent confidence intervals are within +/-7 percentage points of the estimate itself. [5] GAO, New Markets Tax Credit: The Credit Helps Fund a Variety of Projects in Low-Income Communities, but Could Be Simplified, [hyperlink, http://www.gao.gov/products/GAO-10-334] (Washington, D.C.: Jan. 29, 2010). [6] At the time of this report's publication, there are many bills pending in Congress that would extend the expired NMTC. For example, the Senate had begun floor debate on S.2260, which would extend the NMTC through 2015. The House of Representatives had yet not begun considering H.R.3939 or H.R.4365, which would permanently extend the credit. [7] For a discussion of the criteria appropriate for evaluating tax expenditures such as the NMTC, see GAO, Tax Expenditures: Background and Evaluation Criteria and Questions, [hyperlink, http://www.gao.gov/products/GAO-13-167SP] (Washington, D.C.: Nov. 29, 2012). [8] The simplified examples focus on the revenue effects of the NMTC and are not intended to be estimates of the credit's total revenue cost. For example, they do not reflect other tax liabilities that investors may incur on the credit claimed that would reduce the revenue cost. For estimates of NMTC revenue costs, see U.S. Congress Joint Committee on Taxation, Estimated Revenue Effects of the Revenue Provisions Contained in the "Family and Business Tax Cut Certainty Act of 2012," as reported by the Senate Committee on Finance, Aug. 29, 2012, JCX-71-12. [9] In Rev. Rul. 2003-20, the IRS ruled that for purposes of determining the NMTC allowable under Section 45D, the amount of the QEI made by an investment fund may include cash from a nonrecourse loan that the investment fund invests as equity in a qualified CDE. In Rev. Rul. 2010-17, the IRS ruled that the amount of QEI could also include cash from a recourse loan. [10] In Notice 2002-64, IRS allowed the NMTC to be combined with other federal programs except the Low-Income Housing Tax Credit (LIHTC). Treasury issued final regulations in 2004 incorporating the guidance from the 2002 IRS Notice--see Treas. Reg. § 1.45D-1(g)(3). Investors interpreted the Treasury regulations, together with the 2003 Revenue Ruling allowing leveraging of private debt, to permit leveraging with other federal programs such as historic tax credits where the value of these credits was used to increase the base of the NMTC by being loaned to an investment fund. Treasury has not amended its regulations or issued specific guidance to address this practice. [11] GAO, Tax Policy: New Markets Tax Credit Appears to Increase Investment by Investors in Low-Income Communities, but Opportunities Exist to Better Monitor Compliance, [hyperlink, http://www.gao.gov/products/GAO-07-296] (Washington, D.C.: Jan. 31, 2007); and New Markets Tax Credit Coalition, New Markets Tax Credit Progress Report 2014, (Washington, D.C.: June 10, 2014). [12] According to data from an accounting and consulting firm, 15 states currently have state new markets tax credit programs and 36 states have state historic tax credit programs. Novogradac & Company LLP, Tax Credit Resource Centers, accessed June 10, 2014, [hyperlink, http://www.novoco.com/resource_center/index.php]. [13] Urban Institute, New Markets Tax Credit (NMTC) Program Evaluation: Final Report, a report prepared for U.S. Department of the Treasury, Community Development Financial Institutions (CDFI) Fund, April 2013. [14] 26 U.S.C. § 45D(i)(1); Treas. Reg. § 1.45D-1(g)(3). [15] These risks of higher costs led, in part, to our prior suggestion that Congress consider a pilot test whereby some NMTCs would be converted to grants. [16] As part of their 2012 NMTC Allocation Agreements with the CDFI Fund, CDEs are now required to prepare a separate stand-alone document that discloses "any and all direct and indirect" NMTC-related transactions costs related to the loan or investment (such as third- party legal and accounting costs), and any compensation (fees) that the CDE is assessing the low-income community business prior, to, during, and at the conclusion of the 7-year NMTC term. The CDFI Fund provided a sample disclosure statement to CDEs based on commonly-used leverage model transactions and typical costs identified with NMTC transactions. The CDEs are required to provide the low-income community businesses a final disclosure statement at closing. [17] GAO, Tax Expenditures: Background and Evaluation Criteria and Questions, [hyperlink, http://www.gao.gov/products/GAO-13-167SP] (Washington, D.C.: Nov. 29, 2012) and Understanding the Tax Reform Debate: Background, Criteria and Questions, [hyperlink, http://www.gao.gov/products/GAO-05-1009SP] (Washington, D.C.: Sept. 1, 2005). [18] GAO, Standards for Internal Control in the Federal Government, [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1] (Washington, D.C.: Nov. 1, 1999). [19] U.S. Department of the Treasury, Community Development Financial Institutions Fund, "New Markets Tax Credit Public Data Release: 2003- 2011 Summary Report," July 2, 2013. [20] GAO, Managing for Results: Assessing the Quality of Program Performance Data, [hyperlink, http://www.gao.gov/products/GAO/GGD-00-140R], (Washington, D.C.: May 25, 2000). [21] Seventy-three projects were excluded because they had loans and investments originating both before and after December 31, 2010, and 49 projects were excluded because one or more CDEs appeared to have reported fees data incorrectly by reporting fees as a percentage of the QEI rather than as basis points as instructed by the CDFI Fund. [22] As noted in appendix I, our survey population consisted of randomly selected NMTC projects with all loans and investments closing on or after January 1, 2010. Restricting the survey population to NMTC projects within our study's time period left us with a total of 1,265 projects in the starting population from which we drew our sample. [End of section] GAO's Mission: The Government Accountability Office, the audit, evaluation, and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO's commitment to good government is reflected in its core values of accountability, integrity, and reliability. Obtaining Copies of GAO Reports and Testimony: The fastest and easiest way to obtain copies of GAO documents at no cost is through GAO's website [hyperlink, http://www.gao.gov]. Each weekday afternoon, GAO posts on its website newly released reports, testimony, and correspondence. To have GAO e-mail you a list of newly posted products, go to [hyperlink, http://www.gao.gov] and select "E-mail Updates." Order by Phone: The price of each GAO publication reflects GAO's actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO's website, [hyperlink, http://www.gao.gov/ordering.htm]. Place orders by calling (202) 512-6000, toll free (866) 801-7077, or TDD (202) 512-2537. Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information. Connect with GAO: Connect with GAO on facebook, flickr, twitter, and YouTube. Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts. Visit GAO on the web at [hyperlink, http://www.gao.gov]. To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Website: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]; E-mail: fraudnet@gao.gov; Automated answering system: (800) 424-5454 or (202) 512-7470. Congressional Relations: Katherine Siggerud, Managing Director, siggerudk@gao.gov: (202) 512-4400: U.S. Government Accountability Office: 441 G Street NW, Room 7125: Washington, DC 20548. Public Affairs: Chuck Young, Managing Director, youngc1@gao.gov: (202) 512-4800: U.S. Government Accountability Office: 441 G Street NW, Room 7149: Washington, DC 20548. [End of document]