CohnReznick Low-Income Housing Tax Credit Program at Year 25 by Reznick Group

The Low-Income Housing Tax Credit Program at Year 25:

An Expanded Look at Its Performance A CohnReznick LLP Report December 2012

Introduction This is the second in a series of periodic reports issued by CohnReznick LLP that addresses the performance of properties financed with low-income housing tax credits (housing tax credits). The housing tax credit program was enacted as a way to stimulate the development of affordable housing for low- to moderate-income families. Over the past 26 years, the program has been studied by various interested parties such as the General Accounting Office (GAO) and various congressional oversight committees to determine whether: • The program was meeting the original intent of Congress • The program should be made permanent • The program could be made more efficient. Given the scrutiny that federal income tax expenditures are undergoing, this is an appropriate time to examine whether housing tax credit properties are meeting their financial obligations and the needs of the markets they serve. While the report was not undertaken for this particular purpose, the data collected provides most of the information that would be required to assess the housing tax credit program. To compile and analyze the data required for the assessment, CohnReznick requested the participation of 40 investment sponsors and the nation’s largest institutional investors. Nearly every sponsor of housing tax credit investments, and many of the nation’s largest investors, participated in the survey. For a complete list of study participants, please refer to Appendix A. With the assistance of Integratec, our affiliated real estate services and software solutions company, CohnReznick examined data collected from the financial statements of 17,118 apartment properties. While CohnReznick examined operating data for every housing credit property without regard to the year in which the property was placed in service, focus was on the manner in which housing credit properties performed during the economic downturn from 2008 to 2010. For a more extensive discussion of the methodology employed to collect and analyze property data, please refer to Appendix B. This report represents an expansion of CohnReznick’s first housing credit property study that was published in August 2011. A phased approach allowed CohnReznick to supply much needed recent industry data while still operating within the timeframe necessary to perform a current yet increasingly rigorous analysis of the data collected. In addition to the inclusion of 800 additional properties, the enclosed report contains an expanded analysis of the August 2011 study and provides: • Additional analysis of the various trends CohnReznick identified in different parts of the country • Fundamental reasons why the housing tax credit portfolio exhibited better financial results in 2008–2010 and why certain properties underperform • How the performance of housing tax credit funds has improved for the benefit of investors over the years. We are grateful to the many firms that supported CohnReznick’s effort in promoting a deeper understanding of the housing tax credit program, its strengths, opportunities for improvement and the critical role the program plays in the development of affordable housing.

CohnReznick LLP December 2012

A CohnReznick Report | 1

Report Restrictions The initial report on the performance of low-income housing tax credit properties (â&#x20AC;&#x153;The Low-Income Housing Tax Credit Program at Year 25: A Current Look at its Performanceâ&#x20AC;?) was published by Reznick Group, P.C. in August 2011. In October 2012, Reznick Group, P.C. combined with J.H. Cohn, LLP to form CohnReznick, LLP, which is now the 11th largest accounting, tax and advisory firm in the United States. Accordingly, this follow-up report is issued by CohnReznick, LLP. CohnReznick has used information gathered from the housing credit industry participants listed in Appendix A to compile this study. The information provided to us has not been independently tested or verified. As a result, we have relied exclusively on the study participants for the accuracy and completeness of their data. No study can be guaranteed to be 100% accurate, and errors can occur. CohnReznick does not warrant the completeness or the accuracy of the data submitted by study participants and thus does not accept responsibility for your reliance on this report or any of the information contained herein. The information contained in this report includes estimations, approximations and assumptions and is not intended to be legal, accounting or tax advice. Please consult a lawyer, accountant or tax advisor before relying on any information contained in this report. CohnReznick disclaims any liability associated with your reliance on any information contained herein. To ensure compliance with the requirements imposed by the IRS, we inform you that any U.S. federal tax advice contained in this communication (including any attachments) is not intended or written to be used, and cannot be used, for the purpose of (i) avoiding penalties under the Internal Revenue Code or (ii) promoting, marketing or recommending to another party any transaction or matter addressed herein.

2 | The Low-Income Housing Tax Credit Program

Table of Contents Chapter 1: Executive Summary..

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Chapter 2: Portfolio Performance. . . . . . . . . . . . . . . . . . . 2.1. Physical Occupancy. . . . . . . . . . . . . . 2.2. Debt Coverage Ratio. . . . . . . . . . . . . 2.3. Per-Unit Cash Flow. . . . . . . . . . . . . . . . 2.4. Possible Explanations for Improved Financial Metrics. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . 13 . . . . . . . . . . . . . . . . 15 . . . . . . . . . . . . . . . . 16 . . . . . . . . . . . . . . . . 17 . . . . . . . . . . . . . . . . 18

Chapter 3: Portfolio Performance by Segments.. . . . . . . . . . . . . . . . . . . 3.1. Segmentation Analysis – by Property Age. . . . . . . . 3.2. Segmentation Analysis – by Property Size . . . . . . . . 3.3. Segmentation Analysis – by Investment Type. . . . . 3.4. Segmentation Analysis – by Credit Type. . . . . . . . . . 3.5. Segmentation Analysis – by Development Type.. . 3.6. Segmentation Analysis – by Tenancy Type.. . . . . . . 3.7. Geographic Segmentation Analysis . . . . . . . . . . . . . 3.7.1 Geographic Segmentation Analysis – by Region. . 3.7.2. Geographic Segmentation Analysis – by State.. . . 3.7.3 Geographic Segmentation Analysis – by Top 10 MSAs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 4:

Nonperforming Properties. . . . . . . . . . . . . . . . 4.1. Operating Underperformance. . . . . 4.1.1. Underperformance in 2010. . . . . . . . 4.1.2. Historical Trend (2008–2010). . . . . . . . 4.1.3. Chronic Underperformance. . . . . . . 4.1.4. Magnitude of Underperformance. . 4.1.5. Underperformance – by State.. . . . . 4.2. Foreclosure. . . . . . . . . . . . . . . . . . . . . . . 4.3. Technical Underperformance. . . . . .

Chapter 5:

Fund Investment Performance. . . . . . . . . . 5.1. Introduction. . . . . . . . . . . . . . . . . . . . . 5.2. Fund Yields. . . . . . . . . . . . . . . . . . . . . . 5.3. Yield Variance Analysis. . . . . . . . . . . 5.4. Housing Credit Variance Analysis. .

. . . 25 . . . 25 . . . 27 . . . 28 . . . 29 . . . 30 . . . 31 . . . 31 . . . 32 . . . 36 . . . 39

. . . . . . . . . . . . . . . . 42 . . . . . . . . . . . . . . . . 43 . . . . . . . . . . . . . . . . 43 . . . . . . . . . . . . . . . . 45 . . . . . . . . . . . . . . . . 46 . . . . . . . . . . . . . . . . 47 . . . . . . . . . . . . . . . . 49 . . . . . . . . . . . . . . . . 54 . . . . . . . . . . . . . . . . 56

. . . . . . . . . . . . . . . . . 57 . . . . . . . . . . . . . . . . . 57 . . . . . . . . . . . . . . . . . 59 . . . . . . . . . . . . . . . . . 60 . . . . . . . . . . . . . . . . . 61

A CohnReznick Report | 3

Chapter 6:

Portfolio Composition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1. Portfolio Composition – by Property Age. . . . . . . 6.2. Portfolio Composition – by Property Size. . . . . . . . 6.3. Portfolio Composition – by Investment Type. . . . 6.4. Portfolio Composition – by Credit Type. . . . . . . . . 6.5. Portfolio Composition – by Development Type.. 6.6. Portfolio Composition – by Tenancy Type.. . . . . . 6.7. Portfolio Composition – by Region. . . . . . . . . . . . . 6.8. Portfolio Composition – by State. . . . . . . . . . . . . . . 6.9. Portfolio Composition – by MSA. . . . . . . . . . . . . . .

Appendices:. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix A. Acknowledgments. . . . . . . . . . . . . . . . . . . . . Appendix B. Survey Methodology. . . . . . . . . . . . . . . . . . . Appendix C. Glossary.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix D. Property Performance – by State. . . . . . . . Appendix E. Property Underperformance – by State. . Appendix F. Property Performance – by MSA. . . . . . . .

4 | The Low-Income Housing Tax Credit Program

. . . . . 64 . . . . . 64 . . . . . 65 . . . . . 67 . . . . . 68 . . . . . 68 . . . . . 69 . . . . . 70 . . . . . 72 . . . . . 72 . . . . . 74 . . . . . 74 . . . . . 75 . . . . . 80 . . . . . 82 . . . . . 84 . . . . . 87

Index of Figures Figure 2.0.1

Overall Portfolio Composition.. . . . . . . . . . . . . . . . . . . . . . . . . . . Overall Portfolio Performance (2008–2010). . . . . . . . . . . . . . . . . . 2.0.3 Overall Portfolio Performance (2008–2010) – Comparison of August 2011 and Current Report. . . . . . . . . . . . 2.1 Median Physical Occupancy (2008–2010). . . . . . . . . . . . . . . . . . 2.2 Median Debt Coverage Ratio (2008–2010).. . . . . . . . . . . . . . . . . 2.3 Median Per-Unit Cash Flow (2008–2010). . . . . . . . . . . . . . . . . . . . 2.4.1 Net Equity Price by Year Placed in Service. . . . . . . . . . . . . . . . . . 2.4.2 Net Equity Price vs. Hard Debt Ratio by Year Placed-in-Service – 9% Housing Tax Credit Properties. . . . . . . . . . 2.4.3 Net Equity Price vs. Hard Debt Ratio by Year Placed-in-Service – 4% Housing Tax Credit Properties. . . . . . . . . . 2.4.4 Hypothetical Housing Tax Credit Project Debt Service Calculation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1(a) 2010 Median Physical Occupancy by Year Placed-in-Service. . . 3.1.1(b) 2010 Median Debt Coverage Ratio by Year Placed-in-Service. . . 3.1.1(c) 2010 Median Per Unit Cash Flow by Year Placed-in-Service. . . . . 3.1.1(D) 2008/2009 Median Debt Coverage Ratio by Year Placed-in-Service. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1(E) 2008/2009 Median Per Unit Cash Flow by Year Placed-in-Service. 3.2 Operating Performance by Project Size. . . . . . . . . . . . . . . . . . . . 3.3 Operating Performance by Investment Type. . . . . . . . . . . . . . . . 3.4 Operating Performance by Credit Type. . . . . . . . . . . . . . . . . . . . 3.5 Operating Performance by Development Type. . . . . . . . . . . . . . 3.6 Operating Performance by Tenancy Type. . . . . . . . . . . . . . . . . . 3.7.1 Portfolio Distribution by Region. . . . . . . . . . . . . . . . . . . . . . . . . . . 3.7.1(a) Operating Performance by Region. . . . . . . . . . . . . . . . . . . . . . . 3.7.1(b) 2010 Median Physical Occupancy by Region. . . . . . . . . . . . . . . 3.7.1(c) 2010 Median Debt Coverage Ratio by Region . . . . . . . . . . . . . . 3.7.1(d) 2010 Median Per Unit Cash Flow by Region. . . . . . . . . . . . . . . . . 3.7.2(a) 2010 Median Physical Occupancy by State. . . . . . . . . . . . . . . . . 3.7.2(b) 2010 Median Debt Coverage Ratio by State. . . . . . . . . . . . . . . . 3.7.2(c) 2010 Median Per Unit Cash Flow by State. . . . . . . . . . . . . . . . . . . 3.7.3 Operating Performance by Top 10 MSAs. . . . . . . . . . . . . . . . . . . 4.1.1 2010 Underperformance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.1(a) 2010 Per Unit Cash Flow Performance by Property Size. . . . . . . . . 4.1.2 Underperformance (2008–2010).. . . . . . . . . . . . . . . . . . . . . . . . . 4.1.3 Chronic Underperformance.. . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.4(a) Distribution of 2010 Physical Occupancy. . . . . . . . . . . . . . . . . . . 4.1.4(b) Distribution of 2010 Debt Coverage Ratio.. . . . . . . . . . . . . . . . . . 4.1.4(c) Distribution of 2010 Per Unit Cash Flow. . . . . . . . . . . . . . . . . . . . . 4.1.4(D) Operating Deficit Funding Sources for 2010 Property Deficits. . . . 4.1.5(a) 2010 Occupancy Underperformance by State (Percent Below 90% Physical Occupancy). . . . . . . . . . . . . . . . . . 4.1.5(b) 2010 Debt Coverage Ratio Underperformance by State (Percent Below 1.00 DCR). . . . . . . . . . . . . . . . . . . . . . .

Figure 2.0.2 Figure

Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure

Figure

. . . 14 . . . 15 . . . . .

. . . . .

15 16 17 18 19

. . . 20 . . . 21 . . . .

. . . .

21 25 26 26

. . . . . . . . . . . . . . . . . . . . . . . .

27 27 28 29 30 30 31 32 33 34 35 36 37 38 39 40 44 44 45 46 47 48 48 49

. . . 49 . . . 50

A CohnReznick Report | 5

Figure 4.1.5(c)

Figure Figure Figure Figure Figure Figure Figure

Figure Figure Figure Figure Figure

Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure

2010 Per Unit Cash Flow Underperformance by State (Percent Below $0 Per Unit Cash Flow). . . . . . . . . . . 4.1.5(d) Chronic Occupancy Underperformance by State (Below 90% Occupancy in All Three Years 2008–2010). . . . . . 4.1.5(e) Chronic DCR Underperformance by State (Below 1.00 DCR in All Three Years 2008–2010). . . . . . . . . . . . 4.1.5(f) Chronic Cash Flow Underperformance by State (Below $0 Per Unit Cash Flow in All Three Years 2008–2010). . . . . . . . . 4.2.1 Cumulative Foreclosure Rate by Year.. . . . . . . . . . . . . . . . . . 5.1.1 Total Surveyed Gross Equity by Fund Type (Post 1994 Funds). . 5.2(a) Gross Equity Price vs. Fund Yield by Year. . . . . . . . . . . . . . . . 5.2(b) Surveyed Housing Tax Credit Fund Yield vs. 10-Year Treasury Security Rate (after tax equivalent). . . . . . . . . . . . . 5.3 Fund Yield Variance by Year. . . . . . . . . . . . . . . . . . . . . . . . . 5.4.1 Housing Credit Delivery Variance by Investment Type. . . . . . 5.4.2 Initial Years’ Housing Credit Delivery Variance by Year Fund Closed. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 Net Equity by Year Placed-in-Service. . . . . . . . . . . . . . . . . . . 6.1(a) Percent Net Equity by Property Age (Years since Placed-in-Service, as of 12/31/2010). . . . . . . . . . 6.2 Average Project Size by Year Placed in Service. . . . . . . . . . . 6.2(a) Average Project Size by Net Equity, Credit Type and Year Placed-in-Service. . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Percent Net Equity by Investment Type. . . . . . . . . . . . . . . . . 6.4 Percent Net Equity by Credit Type. . . . . . . . . . . . . . . . . . . . . 6.5 Percent Net Equity by Development Type. . . . . . . . . . . . . . . 6.6 Percent Net Equity by Tenancy Type. . . . . . . . . . . . . . . . . . . 6.7 Portfolio Composition by Region. . . . . . . . . . . . . . . . . . . . . . 6.7(a) Percent Net Equity by Region. . . . . . . . . . . . . . . . . . . . . . . . 6.7(b) Average Project Size by Region. . . . . . . . . . . . . . . . . . . . . . . 6.8 Percent Net Equity by State. . . . . . . . . . . . . . . . . . . . . . . . . . 6.9 Net Equity Concentration among Top 10 MSA’s. . . . . . . . . . .

6 | The Low-Income Housing Tax Credit Program

. . . . . . 51 . . . . . . 52 . . . . . . 53 . . . .

. . . .

54 55 58 59

. . . . . . 60 . . . . . . 61 . . . . . . 62 . . . . . . 62 . . . . . . 64 . . . . . . 65 . . . . . . 65 . . . . . . . . . .

. . . . . . . . . .

66 67 68 69 69 70 71 71 72 73

Chapter 1:

Executive Summary Background

he Low-Income Housing Tax Credit program reached the 25th anniversary of its enactment in 2011. Adopted in the midst of dramatic changes to the Internal Revenue Code in 1986, the program has since enjoyed a strong level of bipartisan support in the United States Congress. Following are some of the many features that make the housing tax credit program unique.

• The cost of the housing tax credit program to the federal government is fixed and determinable by statute. The program is subject to a volume limit that permits its cost, unlike most tax expenditures, to be calculated with precision, thus ensuring that it cannot become a “runaway” government program. • Housing tax credits are divided among the states based on their respective populations. The determination of which projects are to be awarded housing credit allocations is made by state housing credit agencies pursuant to a set of highly transparent procedures. As a result of its local control, the program has proven to be adaptable enough to serve changing housing needs as established by the states rather than by the federal government. • For the last 15 years, the demand for housing tax credits has exceeded supply almost every year. This imbalance between the supply and demand for housing credits has resulted in a highly efficient use of tax credit dollars as a tool to finance the construction of new affordable housing and the rehabilitation of older affordable housing complexes. • Over the course of the past decade, the occupancy level in housing tax credit properties has consistently been approximately 96%. Given the normal turnover of rental units, this means that housing credit properties are effectively fully occupied. An inventory of 25,000+ fully occupied properties is directly attributable to the number of U.S. households that are “rent burdened.” The traditional measure for severe rent burden is when more than 50% of household income is required for housing. In 2010, 20.2 million U.S. households fell into this category.1 • In addition to the housing credit program, there are other federal housing programs designed to maintain the affordability of rents for low-income tenants. The housing credit program is unique because it functions as a capital subsidy to stimulate the production of new affordable housing and represents a public-private partnership. The housing tax credit program has been, by any measure, a resounding success. A number of previous studies of the housing tax credit program have documented the program’s favorable record of: • Serving income-qualified tenants at restricted rents • Operating with an exceedingly low number of properties being lost to foreclosure • Maintaining high levels of occupancy.

Source: Joint Center for Housing Studies of Harvard University. “The State of the Nation’s Housing 2012.”

A CohnReznick Report | 7

The housing tax credit program has developed a strong track record for delivering quality housing to low-income families, meeting the expectations of institutional investors and maintaining a cumulative foreclosure rate that is less than 1%, which is more fully described in the report. State housing credit agencies are statutorily obligated to award only enough housing tax credits to make potential developments financially feasible, and the allocators have been effective at ensuring that projects to which they award housing credits have not been overfinanced. With statutory rent restrictions constraining the income potential of housing credit projects, one consequence of this statutory obligation is that housing tax credit properties are underwritten with very little margin for error in generating sufficient net operating income. Accordingly, when operating expenses are higher than projected or when rents are marginally lower than expected, housing tax credit properties may produce just enough or slightly less cash flow than is needed to service their mortgage debt. In recent years, as much as 35% of all housing tax credit properties have operated below break-even, albeit often by fairly small amounts.2 At the same time, studies have confirmed that the rate of foreclosure for housing tax credit properties has been very low in the aggregate. The apparent contradiction between these two data points – break-even or slightly below break-even net operating income in conjunction with an overall low foreclosure rate – is discussed in this report. The tension between these two economic realities has left some investors with the impression that housing credit investments are riskier than was previously understood. The concern held by some parties with respect to this issue has been further exacerbated by the challenges to the national economy occasioned by the near meltdown of the financial services sector and the negative economic developments that followed it. Addressing the question of whether the number of struggling housing credit properties may have escalated and whether the incidence of foreclosures has increased in recent years was a central objective of CohnReznick’s August 2011 report. Our analysis of the data suggests that there has been no material deterioration in housing credit property performance and that certain operating performance metrics significantly improved from 2008 to 2010.

Survey Findings CohnReznick achieved strong industry participation in its efforts to compile operating data for this report. Yielding a 95% overall response rate, 32 organizations chose to participate in our August 2011 study and an additional six participated in this study. All analytics have been updated to include the additional property data from the new participants. Cumulatively, study respondents provided CohnReznick with operating data for 17,118 housing credit properties, 90% of which achieved stabilized operations as of December 31, 2010. On average, stabilized properties in the surveyed pool CohnReznick analyzed had operated for approximately seven years as of December 31, 2010. This pool represents approximately $73 billion in housing tax credits approximately $62 billion in equity contribution from investors to finance property development.3

2 3

Source: Ernst & Young. “Understanding the Dynamics V.” The net equity and credit figures are slightly understated in value and slightly mismatched as a result of missing data from either or both data fields. We estimate that the Housing Credit Net Equity figure may be understated by approximately $1 billion, and the Total Housing Credits figure may be understated by approximately $1.7 billion.

8 | The Low-Income Housing Tax Credit Program

Our analysis of housing credit property performance is based on the three most important metrics for measuring property operations: • Physical occupancy • Debt coverage ratio • Per-unit net cash flow. The August 2011 study included data for 16,356 properties; the current report contains data for 17,118 properties, representing an increase of 762 properties (4.5% of the total pool). In comparison to CohnReznick’s August 2011 study, the expansion in the number of properties included in our data sample had rather limited impact on the change in overall industry performance. We attribute this limited impact to the large scale of the August 2011 data sample, which provided a solid statistical basis to support the findings. The limited impact of increasing the data sample is also partly attributed to the fact that trends identified in the August 2011 study appear to be common across individual participants’ respective property portfolios. CohnReznick reports the following operating results from the data collected from respondents: • Housing tax credit properties typically require economic occupancy of at least 87–89% (or physical occupancy of approximately 89-91%) to attain break-even operations. In recent years, the median occupancy in housing credit properties has reliably been approximately 96%. Notwithstanding the national recession and sharp increase in unemployment, median occupancy in housing credit properties was 96.4%, 96.3% and 96.6% in 2008, 2009 and 2010, respectively. As previously noted, high occupancy rates are another indicator of the tremendous imbalance between the increasing demand and short supply of affordable housing properties. Many survey respondents we interviewed following publication of our August 2011 report noted that unfavorable economic conditions led to enlarged tenant bases across properties in their affordable housing portfolios. • The median debt coverage ratio (DCR) for housing tax credit properties has hovered between 1.13 and 1.15 for a significant portion of the past decade. The data indicate that debt coverage ratios climbed from these levels to 1.21 in 2009 and 1.24 in 2010. Similar to positive trends in the DCR, annual net cash flow per apartment unit was $250 in 2008, $341 in 2009 and reached $419 in 2010. As more fully described in Section 2.4, CohnReznick made significant efforts aimed at answering the question “What made this improvement possible?” including designing quantitative analysis to test our suppositions and interviewing industry experts to draw on their experience. Out of many possible explanations for the growth in debt coverage ratios and cash flow, we point out two important factors: better expense underwriting practices and a more favorable mix of debt to equity, which characterizes most of the properties developed over the past five years. • CohnReznick prepared further analyses of the manner in which metric variances are distributed by segmenting properties according to construction type, property age and property size, in addition to other segmentation categories. CohnReznick has not observed material differences in operating results based on segmentation comparisons. Historically, the most significant difference in property performance has been attributed to certain geographical areas, particularly the Midwest. As a result, this report focuses on the influence that geography has on property performance by region, state and metropolitan

A CohnReznick Report | 9

statistical area (MSA). CohnReznick found through careful analysis of the performance data that low-income housing tax credit properties in certain areas tend to have more favorable operating histories than others. Nonetheless, with rare exceptions, properties in virtually all markets collectively showed improved financial performance from 2008 to 2010. In fact, 2010 is the first year in which every one of the 50 states reported a median debt coverage ratio of greater than 1.00 for the entire state. However, certain markets remain fragile and report a disproportionately larger share of both underperforming properties and persistent operating deficits. In addition to regional and state-level performance metrics, the report includes performance data generated for MSAs for which a meaningful sample size could be obtained. Operating data at the MSA level show more volatility because of a smaller sample size. Though CohnReznick is making the study’s MSA data available, we caution readers to refrain from drawing conclusions about an MSA based solely on the enclosed MSA information. We encourage readers to contact CohnReznick professionals to assess the data provided in the appropriate context. Operating performance by MSA is included as Appendix F. • In addition to studying operating performance for the entire cohort of surveyed properties, CohnReznick adhered to current industry practice for isolating “performing” as opposed to “underperforming” properties. We identify properties as underperforming when they report occupancy levels below 90% and/or debt coverage ratios below 1.0. It was within the group of underperforming properties that CohnReznick observed the most significant change in operating results from 2008 to 2010. The subset of underperforming properties that report occupancy challenges has been perhaps the most volatile of the data points we analyzed on a year-to-year basis. Before 2008, the percentage of properties reporting below 90% occupancy (on a net equity versus property count basis) ranged from a low of 11.5% to a high of 18%. However, the percentage of properties reporting below 90% occupancy dropped to 11.9% in 2008, increased slightly to 12.6% in 2009 and decreased to 9.5% in 2010. There are certain markets, particularly in parts of the Midwest, where the inherent rent advantage between rents charged by housing tax credit versus market-rate properties has been nominal. CohnReznick has observed continued chronic occupancy challenges in some of these markets. In a few such states, over 20% of the housing credit portfolio operated with physical occupancy below 90% during 2010, representing more than twice the national percentage. Paralleling the overall increase in occupancy in 2010, the percentage of properties reporting negative cash flow and/or negative debt coverage actually decreased from 2008 to 2010. The percentage of properties operating below break-even, which has traditionally been the statistic of greatest concern for investors, has historically been as high as 35%; based on the data collected, this percentage was 33.4% in 2008, 27.8% in 2009 and 24.7% in 2010. The decrease in properties operating below break-even from 2008 to 2010 is clearly a favorable trend, all the more so because it was achieved during an economic downturn. As one might expect, in those states where an above-average number of properties report occupancy challenges, an above-average number of properties report negative cash flow. For instance, while housing credit portfolios in Georgia and Indiana both followed the national trend of improved financial performance, more than 40% of their respective surveyed portfolios incurred operating deficits in 2010. We note, however, that no state reported a median DCR of below 1.0 in 2010.

10 | The Low-Income Housing Tax Credit Program

The vast majority of housing tax credit properties that slip into one of the underperforming categories do so for just a year and return to profitable operation in the following year. Properties that report occupancy and cash flow challenges for three or more consecutive years (characterized as “chronic” underperformers) are therefore fairly unusual. In some cases, these properties are unable to struggle back to breakeven despite changing property management companies, funding large deficits for multiple years or trying to restructure property debt. These properties are deemed to have “structural” deficits because of serious physical plant issues, high area crime rates or similar issues that cannot easily be corrected. Measured against the total pool of underperforming properties, CohnReznick’s data suggest that the percentage of properties reporting chronic underperformance for each consecutive year from 2008 to 2010 was 13.9% for negative debt coverage and just 3.9% for those properties with occupancy rates below 90%. There is an apparent contradiction between the percentage of properties reporting deficits in comparison to the remarkably low foreclosure rate for housing tax credit properties. However, because underperforming properties tend to underperform for only short periods of time and because deficits tend not to be significant in most underperforming properties, it is easier to understand why the cumulative foreclosure rate remains less than 1%. • Historically, a great deal of attention has been given to the relatively small number of housing tax credit properties foreclosed upon by their lenders. It appears that this particular data point may have been understated, in part because some of the larger syndicators were using their own capital to support troubled properties in order to avoid foreclosure. This practice became less prevalent in the years from 2002 to 2006, when investor equity became relatively easy to obtain. As a result, the rate of foreclosures in housing tax credit properties has increased in small increments in recent years. The respondents CohnReznick surveyed reported that 98 of the total property count of 17,118 experienced foreclosure through the end of 2010, an aggregate foreclosure rate of 0.57% measured by property count. Approximately 50% of the stated foreclosures were reported to have occurred between 2008 and 2010. Thus, although operating performance generally improved, the rate of foreclosure from 2008 to 2010 still increased, suggesting that challenging economic conditions may have disproportionately affected chronically underperforming properties during those years. Clearly, the number of foreclosures has been underreported as a result of incomplete data. Over the past 10 years, a minimum of eight syndication firms closed operations or became inactive. CohnReznick believes, on the basis of anecdotal evidence, that some of those firms experienced a disproportionately higher incidence of foreclosures. We were not able to pinpoint the number of foreclosed properties syndicated by these firms, nor were we able to ascertain the total number of properties that had been syndicated by these firms. As a result, any attempt to estimate the impact that the property portfolios syndicated by these firms might have on the industry foreclosure rate would require speculation on our part. Rather than abandon the methodology CohnReznick adopted to undertake this study, and compromise its results, we have confined the scope of our observations to the results we received from study respondents.

A CohnReznick Report | 11

Based upon respondents’ data, while the number and rate of foreclosures increased incrementally from 2008 to 2010, the incidence of foreclosures in housing tax credit properties continued to compare very favorably with the foreclosure rate of market rate multifamily properties and other real estate asset groups. Based on the data we collected, tax credit properties were foreclosed, on average, just shy of year 11 of the 15-year compliance period. As such, while foreclosure is a catastrophic event, the financial impact to investors is much less significant than it is to the property’s lenders. • Virtually all surveyed properties have been syndicated to investors through one of several types of investment funds: direct, proprietary, multi-investor and others. In addition to operational data for property-level investments, survey respondents were asked to supply CohnReznick with performance data for every low-income housing tax credit fund that they syndicated to date. This analysis was intended to assess the track record of low-income housing tax credit funds in terms of delivering the originally projected yield and housing credits to investors. On a weighted average basis, survey respondents reported a positive 6.0% variance in meeting yield targets, i.e., the actual yield was 6% higher than the projected yield at investment closing. That being said, the composition of yield is as, if not more, important than the yield figure alone. This is discussed in greater detail in Chapter 5 of the report. • Consistent with CohnReznick’s industry experience, the relative variance in housing credit delivery versus the projected credits over the entire 10-year housing credit period tends to be very small. While developers and syndicators tend to overestimate timing of credit delivery, the data suggest that syndicators have become more accurate in their forecasts of tax credit timing than they were in the housing credit program’s early years. Whether improved DCR and cash flow metrics can be sustained in the coming years will depend on a number of factors, including whether the industry continues to benefit from the historically low interest rate environment that it has enjoyed in recent years. CohnReznick is committed to conducting similar studies periodically in order to supply the industry with current and reliable data.

12 | The Low-Income Housing Tax Credit Program

Chapter 2:

Portfolio Performance

ohnReznick solicited data from 40 currently active housing tax credit syndicators and a number of the nation’s largest housing credit investors, hereinafter referred to as “data providers” or “respondents.” Thirty-two organizations chose to participate in the August 2011 study, and an additional six participated in the current study, resulting in a 95% overall response rate (see Appendix A). In an effort to avoid the administrative burden of reconciling property investments held in shared portfolios, we collected only direct investment and fund investment performance data from investor participants. All data was provided by the respondents to CohnReznick on a voluntary and strictly confidential basis.

This report summarizes the operating and financial data collected from the respondents for housing tax credit property investments located in each of the 50 states, the District of Columbia, Guam, the U.S. Virgin Islands and Puerto Rico. After adjusting for property investments where equity investments in the same property were held in multiple funds, the data gathered to support the August 2011 report represented 16,356 housing tax credit properties. An additional 762 properties have expanded the dataset for a total of 17,118 housing tax credit properties represented in this report. We believe this data sample represents approximately 70% of the entire inventory of housing tax credit properties that are actively managed by syndicators and/or investors. The gap between CohnReznick’s data sample and 100% of all housing tax credit properties is largely attributable to investments made by defunct syndicators, and properties that have reached the expiration of their respective compliance periods and subsequently “cycled out” of the program.

A CohnReznick Report | 13

As can be observed in Figure 2.0.1, the 17,118 properties in CohnReznick’s data sample collectively represent approximately $62 billion in net equity investments and approximately $73 billion in housing tax credits. Of the 17,118 properties, 15,399 (90%) reached “stabilized operations” as of December 31, 2010. We define stabilized operations as properties that have completed construction, achieved 100% tax credit qualified occupancy (i.e., all of the tax credit units have been occupied by income-eligible tenants) and the property has closed its permanent financing. While the definition of stabilized operations differs slightly among industry participants, CohnReznick defines stabilized operations using the industry’s consensus definition and does not believe these slight differences are significant enough to distort our analysis. As can be observed in Figure 2.0.1, the 15,399 stabilized properties collectively represent approximately 83% of the survey sample on a housing credit net equity basis. Per property, stabilized investments included in this report averaged 72.4 apartment units, $3.4 million in net equity investment and $3.9 million in total housing credits.

Overall Portfolio Composition

Figure 2.0.1

Survey Total

Number of Properties

Stabilized Properties

% of Stabilized

17,118

15,399

90.0%

1,264,353

1,114,928

88.2%

Housing Credit Net Equity

$ 62,363,416,612

$ 51,711,320,179

82.9%

Total Housing Credits

$ 73,155,492,616

$ 60,255,560,568

82.4%

Number of Units

CohnReznick measured the real estate performance of the surveyed properties by using a number of operating and financial metrics, including: • Physical occupancy, defined as the number of units occupied divided by the number of units available within a property • Debt coverage ratio, defined as net operating income less required replacement reserve deposits divided by mandatory debt service payments • Per-unit cash flow, defined as the amount of cash flow generated by each property after deducting debt service payments and required replacement reserve contributions • Incidence of noncompliance • Incidence of foreclosure. This chapter summarizes the 2008–2010 operating performance data of the 15,399 stabilized properties. Figure 2.0.2 summarizes 2008–2010 operating results measured by median physical occupancy, DCR and per-unit cash flow data for the entire stabilized portfolio. While physical occupancy remained consistently strong from 2008 to 2010, DCR and perunit cash flow trended upward in 2008, 2009 and 2010.

14 | The Low-Income Housing Tax Credit Program

Overall Portfolio Performance (2008–2010) 2008

Figure 2.0.2 2009

2010

Median Physical Occupancy

96.4%

96.3%

96.6%

Median Debt Coverage Ratio

1.15

1.21

1.24

Median Per Unit Cash Flow

$250

$341

$419

As previously discussed, the expansion of CohnReznick’s dataset from the August 2011 to the current report has had a limited impact on the overall industry performance data because of the sheer size of the August 2011 report. However, for purposes of comparison, Figure 2.0.3 below represents the data differences between the two reports.

Overall Portfolio Performance (2008–2010) – Comparison of August 2011 and Current Report Median Physical Occupancy

Median Debt Coverage Ratio

2008

2009

2010

2008

2009

August 2011 report

96.4%

96.3%

96.6%

1.15

1.19

Current report

96.4%

96.3%

96.6%

1.15

1.21

2010

Figure 2.0.3 Median Per Unit Cash Flow 2008

2009

2010

1.24

$246

$335

$412

1.24

$250

$341

$419

2.1. Physical Occupancy Syndicators and investors alike generally underwrite housing tax credit property investments based on the assumption that “effective” or “economic” occupancy will be 93%. The assumed economic loss of 7% takes into account the periodic turnover of units, the ability to lease such units and losses resulting from rent skips and/or collection problems. While physical occupancy may be calculated at 95%, it is common for housing tax credit properties to lose an additional 1–2% of gross potential rent because of collection problems. Figure 2.1 summarizes the median physical occupancy data for the stabilized properties CohnReznick surveyed for calendar years 2008 through 2010. The data suggest that, notwithstanding the recent recession, the troubled housing sector and increased unemployment, median occupancy remained consistently robust, with only minor fluctuations from year to year. The 2008 median occupancy rate of 96.4% decreased slightly to 96.3% in 2009 but rebounded and subsequently increased to 96.6% in 2010.

A CohnReznick Report | 15

Median Physical Occupancy (2008–2010) 2008

Median Physical Occupancy

Figure 2.1 2009

96.4%

2010

96.3%

96.6%

In contrast to the housing credit portfolio, the U.S. Census Bureau recently published data suggesting that conventional multifamily properties were negatively impacted by the recession. The Bureau reported that the national multifamily rental vacancy rate climbed from 9.6% in the fourth quarter of 2007 to 10.1% in the fourth quarter of 2008 and increased further to 10.7% in 2009 before returning to the pre-recession level of 9.4% during the fourth quarter of 2010.4 The reasons for the difference in occupancy levels are numerous, but the major driver of the consistently high occupancy rates in housing tax credit properties is that the United States simply does not have enough low-income housing units to satisfy the national demand for affordable housing. In fact, the recent downturn in the economy may have created a more pressing need for low-income housing than ever before. During this downturn, housing tax credit property production contracted to half its pre-recession level as a result of the halving of the housing credit market. In a recently published report, the National Low Income Housing Coalition estimated the deficit of rental units that are both affordable and available for extremely low-income households (those earning up to 30% of area median income [AMI]), to be 6.8 million units in 2010.5 We note that only physical occupancy data have been presented in this report. Economic occupancy, while meaningful, is not monitored by a significant portion of data providers and was thus excluded from the survey. Furthermore, while physical occupancy was relatively consistent across the country, economic losses may have varied significantly, thereby contributing to differing financial performance among housing credit properties across various geographic segments.

2.2. Debt Coverage Ratio The term “debt coverage” relates to the relationship between net income (effective gross rental income less operating expenses and replacement reserve deposits) and mandatory debt service payments. Thus, for example, an apartment project that reports net rental income of $115,000 and $100,000 of annual mandatory debt service is considered to have a 1.15 DCR. Most lenders require housing tax credit properties to generate a debt coverage ratio of at least 1.15 (the industry standard) before agreeing to retire a property’s construction loan and extend long-term permanent financing. In addition, some lenders require higher coverage ratios for properties demonstrating lower real estate quality.

4 5

Source: U.S. Census Bureau American Housing Survey. http://www.census.gov/hhes/www/housing/hvs. Source: National Low Income Housing Coalition. http://nlihc.org/sites/default/files/HousingSpotlight2-1.pdf.

16 | The Low-Income Housing Tax Credit Program

The properties CohnReznick surveyed experienced a steady increase in DCR from 2008 to 2010, at a pace that was more pronounced than the positive trend in occupancy rates. In 2008, median DCR was 1.15, which is consistent with previous studies based on various industry sources6 and coincides with the current industry standard. Furthermore, the median DCR increased to 1.21 in 2009 and increased significantly again to 1.24 in 2010, a surprising result for some industry observers given the national recession, increased unemployment and the turmoil in certain housing markets.

Median Debt Coverage Ratio (2008–2010) 2008

Median Debt Coverage Ratio

Figure 2.2 2009

1.15

2010

1.21

1.24

The improvement in 2008–2010 DCR was pervasive. The same positive trend was identified across virtually every state, property type and financing type and was reflected in the data supplied by almost all individual data providers. Enterprise Community Investment, Inc. (Enterprise), one of the nation’s largest nonprofit housing credit syndicators and a participant in our study, published a November 2010 report examining the operating performance of 1,545 housing tax credit properties in its own portfolio.7 Enterprise’s report concluded that, “Whether using the weighted average method (calculating one DCR for the entire aggregate portfolio) or portfolio median, DCR results were consistently in the 1.05 to 1.12 range between 2004 and 2008. In 2009, there was an 8% increase in DCR to 1.20.”

2.3. Per-Unit Cash Flow The level of cash flow that a property generates (expressed here in terms of annual cash flow per apartment unit) closely tracks the property’s DCR; however, to the extent that a property only has “soft” debt, DCR measurements are less relevant. Soft debt refers to mortgage loans made by government agencies that require current payments only to the extent that the project has sufficient cash flow (or in some cases, do not require any payments until the maturity of such loans even if there is surplus cash flow). Accordingly, the number of properties reporting per-unit cash flow was larger than the number of properties reporting debt coverage. In the same way that DCRs improved from 2008 to 2010, the data suggest that median cash flow per unit increased year over year from 2008 to 2010. Since 2002, cash flow after paying hard debt service was minimal averaging between $200 and $250 per unit per annum (PUPA), or $20 per unit per month. However, the 2008 median cash flow per unit of $250 increased to $341 in 2009 and to $419 in 2010. As previously noted, while these annual increases may appear dramatic, they represent growth in net income per apartment of less than $10 per month.

S ource: Ernst & Young. “Understanding the Dynamics V,” reporting the median DCR of 1.14 for 12,064 housing credit properties in calendar year 2006. 7 Source: Enterprise Community Investment, Inc. “Asset Management LIHTC Portfolio Trends Analysis – November 2010.” 6

A CohnReznick Report | 17

Median Per Unit Cash Flow (2008–2010) 2008

Median Per Unit Cash Flow

Figure 2.3 2009

$250

2010

$341

$419

Along with improved financial and operating performance, the incidence of underperforming properties with respect to per-unit cash flow decreased consistently from 2008 to 2010. The most significant improvement was the lower percentage of housing tax credit properties operating below break-even. Based on the data collected, the percentage of properties operating below breakeven was 33.7% in 2008, 28.3% in 2009 and 25.2% in 2010.

2.4. Possible Explanations for Improved Financial Metrics Since publishing the August 2011 report, CohnReznick committed to answering the question “how is this improvement possible?” Accordingly, CohnReznick interviewed industry experts to draw on their experience and designed quantitative analyses to test possible factors that point to the improvement in performance metrics of housing tax credit properties. There are many possible factors that could have contributed to improved operating and financial performance of a typical housing tax credit property. These factors include: • Higher rental rates • Lower occupancy turnover or collection losses • Lower hard debt service levels • Lower than projected operating expenses or better expense underwriting practices. However, none of these factors can be singled out as a principal or overriding source for improved operations. Of the various causes explored, CohnReznick found that more efficient expense underwriting and more favorable debt-to-equity ratios are the two primary contributors to improved performance. Lower hard debt service: In CohnReznick’s August 2011 report, we speculated that the marked increase in the number of properties carrying lower leverage was an important contributing factor to improved DCRs and cash flow. As housing tax credit prices have trended upward, the overall portfolio reflects an increasing number of properties that have been financed with little to no hard debt. This is not surprising, since a more favorable equity-to-debt mix is a direct result of higher tax credit pricing. Figure 2.4.1. illustrates the evolution of net equity pricing over the past 20 years, measured by the amount of capital investors committed to, in accordance with a pre-negotiated pay-in schedule, in order to receive one dollar of tax credit. At the inception of the housing credit program, equity was raised principally from relatively small investments by individual investors through public offerings. Beginning in 1992 and 1993, a corporate

18 | The Low-Income Housing Tax Credit Program

equity market began to develop, and the housing tax credit program was made permanent in 1993. Institutional investors began to understand the asset class, and syndicators quickly came to realize that raising capital from institutional investors was a more efficient way to raise equity. At the national level, housing tax credits initially traded at net prices as low as $0.50 per dollar of credit, steadily increased to $0.85 per dollar of credit in 2004 and skyrocketed to close to $1.00 at the height of the equity market between 2006 and 2007. However, the exit of Fannie Mae and Freddie Mac and a precipitous decline in the profitability of the largest financial services companies resulted in a meltdown of the housing credit equity market. As a direct consequence, housing tax credit prices fell sharply to $0.70-$0.75 before increasing gradually again at the end of 2010 and in earnest in 2011.

Net Equity Price by Year Placed-in-Service

Figure 2.4.1

housing credit net equity price

$1.00 $0.90 $0.80 $0.70 $0.60 $0.50 $0.40 1991

1993 1992

1995 1994

1997 1996

1999 1998

2001 2000

2003 2002

2005 2004

2007 2006

2009 2008

2010

Readers should note the following: • Housing tax credit prices presented in Figure 2.4.1 are described as the “net equity price” because they reflect the direct amount of equity per dollar of credit that will be invested to finance the development of these properties. We refer to them as “net” prices because they do not include the costs of raising capital such as fees paid to compensate syndicators for their services, brokerage commissions and similar costs often collectively referred to as “the load.” The amount of load can vary significantly depending on an investor’s choice of investment vehicles (multi-investor versus proprietary versus direct investment) and the individual syndicator’s business practices. • The years depicted are a function of the year in which the properties are placed in service, as opposed to when the underlying investments are closed and the housing credit prices are determined. Given the development timeline of a typical housing tax credit property, the prices in Figure 2.4.1 naturally reflect a 1- to 2-year lag in market price.

A CohnReznick Report | 19

• Finally, while the housing tax credit prices in Figure 2.4.1 are median prices reported by survey respondents, a price disparity as wide as 35 cents between properties in certain markets can be observed based on whether the property is located within the Community Reinvestment Act (CRA) assessment area where one or more investors compete for investment in the same property. The CRA and its effect on housing tax credit pricing will be analyzed in a separate report that will be published in the first quarter of 2013. Figures 2.4.2 and 2.4.3 illustrate the impact of higher tax credit pricing on the debt-toequity mix (expressed as a hard debt ratio) of housing tax credit properties placed in service between 1997 and 2011. The right Y-axis shows the median hard debt ratio of the surveyed housing tax credit properties, while the left Y-axis shows the median housing tax credit price for properties placed in service within the same time period. We presented the 4% housing tax credit properties separately from 9% housing tax credit properties because 4% properties tend to be much more heavily leveraged than 9% properties. Though not perfect, a strong inverse relationship exists between a given property’s tax credit price and its level of hard debt. Using 9% tax credit properties as an example, in the late 1990s when median housing tax credit equity pricing was in the mid- to high 70 cents range, one-third of the permanent financing of housing tax credit properties was provided by conventional “hard” debt. In recent years, as housing tax credit prices began to benefit from much lower leverage, only one-fifth of permanent financing is made up of hard debt.

Figure 2.4.2

$1.00

40.0%

$0.95

35.0%

$0.90

30.0%

$0.85

25.0%

$0.80

20.0%

$0.75

15.0%

$0.70

10.0% 5.0%

$0.65 1997

1998

1999

2000

2001

2002

2003

Net Equity Price

20 | The Low-Income Housing Tax Credit Program

2004

2005

2006

2007

Hard Debt Ratio

2008

2009

2010

Hard Debt ratio

housing credit net equity price

Net Equity Price vs. Hard Debt Ratio by Year Placed-in-Service – 9% Housing Tax Credit Properties

Figure 2.4.3

$1.00

70.0%

$0.95

60.0%

$0.90

50.0%

$0.85

40.0%

$0.80

30.0%

$0.75

20.0%

$0.70

10.0%

Hard Debt ratio

housing credit net equity price

Net Equity Price vs. Hard Debt Ratio by Year Placed-in-Service – 4% Housing Tax Credit Properties

0.0%

$0.65 1997

1998

1999

2000

2001

2002

2003

Net Equity Price

2004

2005

2006

2007

2008

2009

2010

Hard Debt Ratio

The following hypothetical example illustrates the scale of the potential impact on debt coverage from lower debt burdens using the following assumptions: • Number of units = 72 • Per unit total development cost = $200,000 • Total development cost = $14,400,000 • Property bears a conventional first mortgage equal to 18% of cost (which is the median hard debt for surveyed projects placed in service in 2009).

Hypothetical Housing Tax Credit Project Debt Service Calculation

Figure 2.4.4

Scenario A (33% leveraged)

Total development costs

Hard debt ratio Total hard debt

Scenario B (18% leveraged)

$14,400,000

33%

Interest rate

$4,752,000

$14,400,000 18%

6.50%

$2,592,000 6.50%

Annual debt service

$360,430

$196,598

Per unit per month debt service

$417

$228

A CohnReznick Report | 21

Compared with the hard debt burden typical of a housing credit property developed in 1997 (33% leverage – Scenario A), the reduction in debt service (Scenario B) equates to a substantial debt service savings of $189 per unit per month. It is worth noting that by 2010, greater than 34% of the surveyed properties were placed in service within the previous five years. As a group, these properties are clearly benefiting from lower levels of hard debt. Notwithstanding the data reflecting the favorable effect of lower leverage, a property’s hard debt ratio has, as a single statistic, little bearing on a property’s overall performance. Indeed, cash flow levels are often considerably higher in larger properties financed with tax-exempt bonds and 4% tax credits. The average 4% property in our survey has 116 apartments units, almost twice the size of the average 9% property with 59 units. Smaller properties have fewer units over which to distribute their fixed costs. As a result, they are more sensitive to debt levels and perform more predictably with lower levels of debt. Operating expenses: Reduced insurance premiums and lower utility bills resulting from individual metering and/or energy efficiencies are two frequently cited reasons for unexpected operating expense savings in recent years. One of the nation’s largest syndicators, and a participant in this study, believes “the increase in DCR is a result of both revenue and expense improvements.”8 Across this syndicator’s portfolio, median total revenue increased 4.5% from 2008 to 2009, relative to the 2% underwritten rate, while expenses increased only 2.2% relative to the 3% underwritten rate. Given the combination of higher-than-underwritten rental income and lower-than-underwritten operating expenses, the syndicator’s portfolio realized a 9.2% increase in net operating income. The syndicator explained that expense savings were most notable in insurance (decrease of 6.4%), real estate taxes (decrease of 5.1%) and utilities (increase of only 1.6%) between 2008 and 2009. While the types of cost savings realized in the syndicator’s portfolio are consistent with other industry participants, we searched for additional confirmation of this observation and found that the National Apartment Association (NAA) reported that in the subsidized housing subset, average operating expenses of $4,441 per unit in 2008 fell to $4,319 per unit in 2009 and rose to $4,856 per unit in 2010. The NAA has an extensive database of multifamily properties consisting of 1.1 million units and does an annual survey of operating expenses.9 The NAA database includes 60,326 units identified as “subsidized affordable housing units.” Unfortunately, there is no way to break this subset down further into housing credit properties versus properties developed under pre-housing credit subsidy programs (Secs. 8, 221 (d)(4), 236, etc.). Nonetheless, NAA’s survey and findings are useful for corroborating our own analysis. Perhaps more important than lower operating expenses, CohnReznick’s industry experience and interviews with survey respondents led us to believe that the housing tax credit industry, as a whole, has come a long way in improving its underwriting of operating expenses. Syndicators, for instance, indicated that the availability of benchmarked data from their own portfolios, state credit allocation agencies and industry data providers have helped them improve their expense underwriting.

8 9

Source: Enterprise Community Investment, Inc. “Asset Management LIHTC Portfolio Trends Analysis – November 2010.” Source: National Apartment Association. “2011 Survey of Operating Income & Expenses in Rental Apartment Communities.”

22 | The Low-Income Housing Tax Credit Program

Other contributing factors: Lenders of permanent financing for housing tax credit properties typically require a 1.15 to 1.20 DCR at conversion of the loan from a construction loan to a permanent loan. During the recent equity market meltdown, when the housing credit industry was able to attract only half of its prerecession level of equity, lenders and investors began to require more stringent underwriting terms, including a DCR of 1.20 to 1.25 and more favorable guarantee and reserve protections. Accordingly, while a 1.24 median DCR may seem high, higher DCRs suggests a greater ability to generated projected cash flow needs. Many industry participants believed that favorable interest rates and the abundance of subsidies, like the Section 1602 program made available in the last few years, played a part in the improvement of property financial performance. However, the full impact of many of these newer programs cannot yet be measured with operating performance data, as the properties developed under these programs have only just begun to achieve stabilized operations. Factors found to have minimal to no bearing on improved financial metrics: While high physical occupancy is an indicator of the demand for affordable housing, it is not an influential one with respect to the 2008-2010 improvement in performance metrics, as occupancy remained within a very narrow band from 96% to 96.9% over the past decade. Higher rental rates: Because of the sheer volume, CohnReznick did not attempt to collect rent rate data. However, participantsâ&#x20AC;&#x2122; senior asset management staff were interviewed from a cross-section of data providers regarding their portfolioâ&#x20AC;&#x2122;s ability to achieve rent increases during the survey years.

A CohnReznick Report | 23

Survey respondents concurred that the decrease in workforce employment, the decreasing homeownership rate and an overriding sense of financial instability led to an enlarged renter pool and increased competition for rental housing from 2008–2010. However, none of the interviewed respondents observed a direct correlation between higher demand and rent increases over the last few years. In some markets, affordable housing properties reduced rents to guard against pricing pressure from market rate rentals or condominium conversions. In other markets, affordable housing properties experienced less pricing pressure because of decreased competition from homeownership, and thus either eliminated concession offerings or implemented rent increases without jeopardizing occupancy. Lower occupancy turnover and collection losses: Because a significant portion of survey respondents do not track economic occupancy or have such data readily available, CohnReznick collected only physical occupancy data. After an initial analysis of the occupancy data and other metrics, we speculated that turnover rates and turnoverrelated costs might have decreased. Interestingly, upon further research and interviews with industry participants, this has not proven to be the case. While tenant retention efforts coupled with declining homeownership rates, particularly among first-time homebuyers, may have resulted in reduced turnover levels in select properties, many asset managers reported that turnover rates had increased for a variety of reasons, including tenants losing or switching jobs and tenants moving more frequently to reap the benefits of a property’s first month’s concessions in a practice referred to as “concession shopping.” We lack a basis for predicting whether the improvement in property operations is a trend likely to sustain itself or is a short-term phenomenon. However, since the data suggest that the improvements in expense underwriting and the favorable impact of lower leverage are more clearly visible in “younger” properties, it is reasonable to assume that more favorable debt coverage ratios will be maintained for the foreseeable future. CohnReznick is committed to conducting similar studies periodically to supply the industry with current and reliable data.

24 | The Low-Income Housing Tax Credit Program

Chapter 3:

Portfolio Performance by Segments

ousing credit investors and lenders frequently question whether property investments in certain geographical areas, construction types, tenancy types, financing types or other segmented criteria tend to perform better than others. This chapter reviews the 2008–2010 trends in occupancy, debt coverage and cash flow according to property age, property size and other attributes. However, the single most important expansion from the segmentation analysis in CohnReznick’s August 2011 report is the geographic segmentation performance data we herein present by region, state and metropolitan statistical area.

3.1. Segmentation Analysis – by Property Age The following graphs illustrate how the 2010 operating and financial performance data may have differed based on the year in which a property was originally placed in service. CohnReznick chose not to present data for properties placed in service during 2009 and 2010 due to the relatively small size of the stabilized sample during the aforementioned years. For purposes of this report, we have used “placed-in-service” date and “property age” interchangeably. Based on Figure 3.1.1(A) below, occupancy by property age is clustered within the 95.5% to 97.5% range, indicating that property age has not been a material driver of occupancy rates as the difference of this range is minimal.

2010 Median physical Occupancy

2010 Median Physical Occupancy by Year Placed-in-Service

Figure 3.1.1(a)

98% 97% 96% 95%

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

Surveyed properties display a wider spread in DCR and per-unit cash flow along the age spectrum; however, CohnReznick observes that there is no linear relationship suggesting that older properties tend to underperform newer ones financially.

A CohnReznick Report | 25

2010 Median Debt Coverage Ratio by Year Placed-in-Service

Figure 3.1.1(B)

2010 Median Debt Coverage Ratio

1.50 1.40 1.30 1.20 1.10 1.00

1990

1992

1994

1996

1998

2000

2002

2004

2010 Median Per Unit Cash Flow by Year Placed-in-Service

2006

2008

Figure 3.1.1(C)

2010 Median per unit cash flow

$700 $600 $500 $400 $300 $200

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

In general, housing tax credit properties placed in service during the past five years generated greater than average cash flows. This is consistent with our findings, as these younger properties have realized the benefit of low levels of leverage and relatively inexpensive financing. A survey respondent noted in its report that “another factor in the DCR improvement is that the newer projects entering the portfolio are larger in size and have healthier net operating income (NOI) than the portfolio as a whole.”10 Further, some respondents noted that the positive improvement in DCRs and cash flows may be attributed to the fact that there tends to be fewer “surprises” in the first several years of a property’s operation once it is stabilized and when its debt has been converted to permanent financing (and resized properly if necessary). Some respondents believe that as properties age, deferred maintenance could become a compounded issue over time and cause decreased DCRs and cash flow.

Source: Enterprise Community Investment, Inc. “Asset Management LIHTC Portfolio Trends Analysis – November 2010.”

26 | The Low-Income Housing Tax Credit Program

The following two figures illustrate that stronger financial performance metrics were observed in nearly every age group of surveyed properties from 2008 to 2009.

2008/2009 Median Debt Coverage Ratio by Year Placed-in-Service

Figure 3.1.1(d)

2008/2009 Median debt coverage ratio

1.45

1.35

1.25

1.15

1.05

1990

1992

1994

1996

1998

2009 DCR

2000

2002

2004

2008 DCR

2008/2009 Median Per Unit Cash Flow by Year Placed-in-Service 2008/2009 Median Per Unit Cash Flow

2006

Figure 3.1.1(e)

$600 $500 $400 $300 $200 $100

1990

1992

1994

1996

2009 Per Unit Cash Flow

1998

2000

2002

2004

2006

2008 Per Unit Cash Flow

3.2. Segmentation Analysis â&#x20AC;&#x201C; by Property Size Figure 3.2 presents the operating performance data of surveyed housing credit properties grouped by size, e.g., number of apartment units per property. On average, stabilized housing tax credit properties are composed of 72 apartment units per property. Those containing between 51 and 100 apartment units per property were found to have operating performance metrics that most closely mirror that of the entire surveyed portfolio.

A CohnReznick Report | 27

Operating Performance by Project Size Number of Apartment units per property

Median Physical Occupancy

Figure 3.2

Median Debt Coverage Ratio

2008

2009

2010

2008

2009

0–25

96.7%

96.2%

96.7%

1.16

1.21

26–50

96.6%

96.5%

96.7%

1.16

51–100

96.6%

96.8%

97.0%

101–200

96.0%

201–300

95.0%

301 or more Overall

2010

Median Per Unit Cash Flow 2008

2009

2010

1.28

$199

$271

$371

1.22

1.24

$221

$303

$354

1.16

1.21

1.24

$286

$395

$460

96.1%

1.14

1.20

1.24

$357

$490

$590

94.6%

95.0%

1.11

1.15

1.19

$256

$344

$498

95.0%

95.4%

1.15

1.17

1.19

$258

$440

$490

96.4%

96.3%

96.6%

1.15

1.21

1.24

$250

$341

$419

3.3. Segmentation Analysis – by Investment Type Figure 3.3 summarizes the operating performance data for stabilized properties segmented by the source of a property’s equity financing, e.g., those acquired by public funds versus multi-investor funds, and so forth. As noted in Chapter 6, multi-investor fund investments account for the majority of the portfolio covered by this report. Consistent with the overall data findings, properties syndicated through multi-investor funds reported an increase in median occupancy during 2009 and 2010, at 96.4% to 96.6%, respectively. Based on our findings, property investments held by proprietary funds slightly outperformed the overall portfolio by the single indicator of occupancy, reflecting the highest 2010 median occupancy of 96.8% among all investment types. In contrast, public funds, consisting of older property investments, reported a somewhat lower 2010 median occupancy rate of 95.8%. Both DCR and per-unit cash flow levels for properties acquired by multi-investor funds mirrored the overall trend as well. However, with slight variations from year to year, direct investments tend to generate debt coverage and cash flow levels that are slightly lower than the overall median. We suspect that this is attributable to the fact that the average size of properties acquired directly is somewhat smaller than average. Other than this distinction, the differences in operating performance from one investment type to another are relatively immaterial.

28 | The Low-Income Housing Tax Credit Program

Operating Performance by Investment Type Median Physical Occupancy

Median Debt Coverage Ratio

2008

2009

2010

Direct

95.8%

95.7%

96.2%

1.08

1.12

Multi-investor

96.4%

96.6%

1.15

Proprietary

96.8%

96.5%

96.8%

Public

95.8%

Overall

96.4%

96.3%

investment Type

2008

2009

2010

Figure 3.3 Median Per Unit Cash Flow 2008

2009

2010

1.18

$114

$247

$347

1.21

1.25

$240

$345

$420

1.16

1.20

1.24

$262

$352

$414

95.8%

1.16

1.18

1.30

$232

$212

$340

96.6%

1.15

1.21

1.24

$250

$341

$419

Fund-level performance metrics, measured in terms of yield and credit delivery variances, are discussed in Chapter 5 of this report.

3.4. Segmentation Analysis – by Credit Type The data reflected in Figure 3.4 summarize the operating performance data for stabilized properties segmented by credit type. Data providers were presented with the options to classify the tax credit type for each property by 9% versus 4% housing tax credits and subsequently further separate the 9% credit properties into two subcategories: 9% new construction properties and 4% & 9% acquisition and rehabilitation properties. However, many respondents did not represent that their properties were classified as “4% & 9%” property; thus, the sample size for “4% & 9%” tax credit types was very small. For purposes of this report, CohnReznick merged the subset of acquisition/rehabilitation properties that qualify for both 4% and 9% credits into the 9% category. As shown below, the median occupancy rate for stabilized 9% credit properties was on par with 4% credit properties from 2008 to 2010. We have not observed meaningful differences between the operating performance of 4% versus 9% properties in terms of DCR. However, the 4% properties we surveyed reported consistently higher levels of cash flow than their 9% counterparts. We attribute this to the fact that properties financed with tax-exempt bonds are generally larger and thus have the ability to distribute their fixed costs over a wider base of apartments. The surveyed 4% properties averaged 116 units per property while the 9% surveyed properties averaged 59 units per property.

A CohnReznick Report | 29

Operating Performance by Credit Type Median Physical Occupancy

Median Debt Coverage Ratio

2008

2009

2010

4% Tax Credits

96.5%

96.3%

96.7%

1.14

1.19

9% Tax Credits

96.5%

96.4%

96.6%

1.15

Overall

96.4%

96.3%

96.6%

1.15

Credit Type

Figure 3.4

2008

2009

2010

Median Per Unit Cash Flow 2008

2009

2010

1.23

$318

$404

$506

1.21

1.25

$220

$329

$399

1.21

1.24

$250

$341

$419

3.5. Segmentation Analysis â&#x20AC;&#x201C; by Development Type Stabilized new construction properties account for the majority of the properties we surveyed. As might be expected, based on historical results, newly constructed properties consistently reported stronger operating performance among all development types, followed by rehabilitated properties and, finally, historic rehabilitation properties (i.e., properties qualifying for both housing and historic rehabilitation credits). Historic rehabilitation properties, which account for 480 properties, reported median occupancy of 95.9% in 2010, representing the lowest among all development types. Per-unit cash flow generated by historic rehabilitation properties was also substantially below average. The fact that historic buildings adapted for use as low-income housing do not perform as well as other property types should not be surprising. Historic buildings formerly used as school houses or for manufacturing are often slower to lease and, because their physical plants are less efficient, tend to experience higher operating and maintenance costs.

Operating Performance by Development Type Median Physical Occupancy

Median Debt Coverage Ratio

Development Type

2008

2009

2010

Historic Rehab

95.7%

95.0%

95.9%

1.05

1.15

1.16

New Construction

96.7%

96.4%

96.8%

1.16

1.20

Rehab

96.0%

96.2%

96.4%

1.16

Mixed

96.0%

95.4%

96.1%

Overall

96.4%

96.3%

96.6%

30 | The Low-Income Housing Tax Credit Program

2008

2009

2010

Figure 3.5 Median Per Unit Cash Flow 2008

2009

2010

$129

$121

1.23

$273

$334

$422

1.22

1.27

$238

$379

$447

0.88

1.11

1.08

$(113)

$211

$251

1.15

1.21

1.24

$250

$341

$419

3.6. Segmentation Analysis – by Tenancy Type Based on CohnReznick’s experience, housing tax credit properties set aside for senior tenants have historically reported somewhat stronger operating results than properties rented to other types of tenants. The results of our survey were consistent with that trend: seniors-only properties (25% of the total) outperformed the overall portfolio consecutively across 2008 to 2010 and consistently by all measures (occupancy, DCR and per-unit cash flow). These results are not surprising, given that senior properties traditionally report lower turnover ratios as well as lower operating expenses. However, the strong performance of properties serving tenants with special needs is not as intuitive. It has been our experience that special needs properties, while among the most challenging to manage, can generate higher levels of operating income because they tend to attract multiple sources of subsidy and are commonly undertaken by nonprofit syndicators, many of whom have chosen to underwrite these projects conservatively and have dedicated staff managing these properties.

Operating Performance by Tenancy Type Median Physical Occupancy

Median Debt Coverage Ratio

2008

2009

2010

Family

96.0%

1.14

1.18

Senior

97.7%

97.4%

97.5%

1.20

Special Needs

97.0%

Other

96.1%

96.4%

Overall

96.4%

96.3%

Tenancy Type

Figure 3.6

2008

2009

2010

Median Per Unit Cash Flow 2008

2009

2010

1.22

$227

$314

$400

1.27

1.29

$300

$414

$458

1.32

1.37

1.45

$397

$529

$551

96.7%

1.17

1.23

1.22

$111

$271

$310

96.6%

1.15

1.21

1.24

$250

$341

$419

3.7. Geographic Segmentation Analysis While housing credit investments provide investors with tax credit benefits, they are ultimately equity investments in operating real estate. A major component of the success of any real estate investment is its geographic location. A well-conceived development can succeed nearly anywhere; however, CohnReznick found through careful analysis of the performance data that low-income housing tax credit properties in certain areas have a more favorable operating history than others. Nonetheless, with rare exceptions, properties in virtually all markets showed improved financial performance during the survey period. In section 3.7 we present property operating data detailed by: • Twelve regional areas we selected • State • The top ten metropolitan statistical areas, ranked by number of properties represented in the survey.

A CohnReznick Report | 31

We further present in Appendix D the property operating and financial performance data organized by MSA where a meaningful sample size could be obtained. CohnReznick stresses that geographic location, while a strong factor in determining an individual propertyâ&#x20AC;&#x2122;s success, is just one of a number of factors that will ultimately lead to success or failure of a given low-income housing tax credit property. An analysis of property operating performance by location classification, i.e., urban, suburban and rural, was conducted but not included in the report. We chose to exclude this particular analysis from the report because of the lack of consistency among data providers in how they define and apply location classification terms to their own portfolios.

3.7.1. Geographic Segmentation Analysis â&#x20AC;&#x201C; by Region CohnReznick separated survey properties in the 50 states, the District of Columbia, Guam, the U.S. Virgin Islands and Puerto Rico into 12 regions with similar geographic profiles that most ideally classified the country. The regions are as follows:

Portfolio Distribution by Region Region Number

Constituent States

Figure 3.7.1 % of stabilized portfolio

Region 1

CA, OR, WA

Region 2

AK, HI

0.3%

Region 3

ID, MT, WY

1.3%

Region 4

AZ, CO, NM, NV, UT

4.7%

Region 5

MN, ND, SD

3.4%

Region 6

IA, KS, NE, MO

7.0%

Region 7

IN, IL, MI, OH, WI

Region 8

AR, OK, TX

Region 9

AL, FL, GA, LA, MS

10.6%

Region 10

KY, NC, SC, TN, VA, WV

11.2%

Region 11

CT, DC, DE, MA, MD, ME, NH, NJ, NY, PA, RI, VT

22.4%

Region 12

GU, PR, VI

<0.1%

32 | The Low-Income Housing Tax Credit Program

14.4%

16.0% 8.0%

Figure 3.7.1(A) illustrates the 2008 to 2010 operating performance (occupancy, DCR and per-unit cash flow) data for stabilized properties in the surveyed portfolio segmented by region, in descending order of the sample size of housing credit properties.

Operating Performance by Region Median Physical Occupancy

Figure 3.7.1(A)

Median Debt Coverage Ratio

2008

2009

2010

2008

2009

Region 12

99.6%

99.7%

99.4%

1.19

1.23

Region 11

97.0%

97.1%

97.2%

1.19

Region 10

96.4%

96.3%

97.0%

Region 9

95.2%

94.8%

Region 8

95.8%

Region 7

2010

Median Per Unit Cash Flow 2008

2009

2010

1.23

$438

$485

$521

1.22

1.34

$238

$420

$511

1.15

1.14

1.24

$209

$328

$402

95.6%

1.16

1.06

1.20

$187

$221

$292

95.0%

95.8%

1.16

1.06

1.22

$218

$302

$362

95.3%

95.5%

95.8%

1.02

1.09

1.16

$42

$201

$304

Region 6

95.0%

95.3%

95.8%

1.13

1.12

1.18

$226

$233

$250

Region 5

96.8%

96.7%

97.1%

1.20

1.22

1.33

$470

$614

$604

Region 4

96.5%

96.0%

96.7%

1.17

1.14

1.26

$375

$444

$548

Region 3

95.4%

94.8%

95.0%

1.04

1.00

1.13

$130

$162

$264

Region 2

96.7%

96.8%

97.0%

1.21

1.28

1.30

$735

$1,169

$959

Region 1

97.6%

97.0%

97.4%

1.28

1.27

1.32

$615

$663

$658

Overall

96.4%

96.3%

96.6%

1.15

1.21

1.24

$250

$341

$419

Region 12: GU, PR, VI; Region 11: CT, DC, DE, MA, MD, ME, NH, NJ, NY, PA, RI, VT; Region 10: KY, NC, SC, TN, VA, WV; Region 9: AL, FL, GA, LA, MS; Region 8: AR, OK, TX; Region 7: IN, IL, MI, OH, WI; Region 6: IA, KS, NE, MO; Region 5: MN, ND, SD; Region 4: AZ, CO, NM, NV, UT; Region 3: ID, MT, WY; Region 2: AK, HI; Region 1: CA, OR, WA.

Seven of the 12 regions reported 2010 median occupancy rates that were higher than the 96.6% overall portfolio median. The highest-performing region measured by occupancy rate was Region 12, which includes Puerto Rico, the U.S. Virgin Islands and Guam. The 101 stabilized properties in this region reported near 100% median occupancy, a rate significantly more favorable than the other regional occupancy rates and greater than the

A CohnReznick Report | 33

national median. The survey data support our experience that properties located in Puerto Rico, the U.S. Virgin Islands and Guam consistently operate at or close to 100% occupancy because the scarcity of affordable housing in these areas. Regional occupancy data suggest that there has not been a meaningful variance from the overall portfolio trend. Given the disproportionately smaller size of properties in Region 3, it is not surprising that it reported the lowest median occupancy of 95%, as a few vacant units can have a more drastic impact on the occupancy rates of smaller relative to larger properties. Figures 3.7.1(B) – (D) illustrate each region’s 2010 median occupancy rate, DCR and perunit cash flow on a national map. Regions are colored such that each performance range is indicated with a different color.

2010 Median Physical Occupancy by Region

95.0%

Figure 3.7.1(B)

97.1% 97.2%

97.4% 95.9% 95.8% 97.0%

96.7%

95.8%

95.7% and below 96.6% 95.7% and below 95.8% to 95.8% to 96.6%

96.7% to 97.0% 96.7% to 97.0%

95.6%

97.1% to 97.1% to 97.4% 97.4%

97.4% above 97.4% andand above

Not surprisingly, regions with median occupancy rates that were greater than the national portfolio median are the same ones that report more favorable financial performance, measured by each region’s respective median debt coverage ratio and per unit cash flow. Properties located on the East and West Coast, representing Regions 1 and 11, were found to have the strongest occupancy performance. Inasmuch as these two regions have the

34 | The Low-Income Housing Tax Credit Program

largest representation of properties in the survey sample, their performance has had the largest influence on overall national portfolio performance. The Southeast and Midwest states reported occupancy rates that were slightly below the portfolio median. Region 7 and Region 3 had the least favorable DCRs, although both made significant improvements in the last few years and are closer to the nationwide median DCR. In 2008, housing credit properties located in Regions 3 and 7 operated just above DCR break-even on a regional level. In 2010, the two regions reported improved median DCRs of 1.13 and 1.16, respectively.

2010 Median Debt Coverage Ratio by Region

1.13

Figure 3.7.1(C)

1.34

1.33

1.32 1.16 1.18 1.24

1.26

1.22

1.71 and 1.71 below and below

1.18 to 1.19 1.18 to 1.19

1.20 to 1.25 1.20 to 1.25

1.2

1.26 1.26toto1.30 1.30

1.30 and above 1.30 and above

Across the national housing tax credit portfolio, the 2008 median per-apartment-unit cash flow was $250 and is generally consistent with what it has been for most of the previous decade.11 In 2010, the national overall median per-unit cash flow was $419, reflecting a significant increase from prior years. While per-unit cash flow growth was common throughout the regions from 2008 to 2010, this trend was most pronounced in the two

Source: Ernst & Young. â&#x20AC;&#x153;Understanding the Dynamics V,â&#x20AC;? citing the trend of DCR over the last decade.

A CohnReznick Report | 35

traditionally underperforming regions (Regions 3 and 7), measured solely by DCR and cash flow. Surveyed housing tax credit properties in Region 7 reported having generated $42 of median per-unit cash flow during 2008, $201 during 2009 and $304 during 2010. Similarly, properties in Region 3 reported having generated $130 of median per-unit cash flow during 2008, $162 during 2009 and $264 during 2010. Consistently reporting high levels of cash flow, Regions 1 and 2 continued to do so during the survey period and reported 2010 median per-unit cash flows in excess of $500.

2010 Median Per Unit Cash Flow by Region

$264

Figure 3.7.1(D)

$604 $511

$658 $304 $250 $402

$548

$362

$0 to $100 $0 to $100

$101 to $250 $101 to $250

$251 to to $500 $500 $251

$292

$501 to $1,000 $501 to $1,000

$1,001 and above $1,001 and above

3.7.2. Geographic Segmentation Analysis â&#x20AC;&#x201C; by State CohnReznick further segmented operating data for the surveyed properties according to their location in the 50 states, Puerto Rico, the U.S. Virgin Islands and Guam. California, New York, Texas, Florida and Illinois collectively accounted for more than 42% of the overall portfolio based on the volume of equity investment represented by surveyed properties in the five states. Properties located in South Dakota, Hawaii, Delaware, Guam and the U.S. Virgin Islands are the bottom five states in terms of overall net equity investment, representing less than 1% of the overall portfolio.

36 | The Low-Income Housing Tax Credit Program

Figure 3.7.2(A) illustrates each state’s 2010 median occupancy rate. The states were grouped and color-coded based on each state’s median occupancy percentage. As previously discussed, occupancy generally decreased slightly between 2008 and 2009 and increased in 2010.

2010 Median Physical Occupancy by State

Figure 3.7.2(A)

ME MT

ND MN

IA NE NV

UT CA

NH VT MA RI CT

NJ DE MD DC

KY NC TN

OK NM

AR MS

LA FL

90.0% and90.0% belowand below 90.1% to90.1% 95.6%to 95.6%

95.61% to 96.7% 95.61% to 96.7%

96.71% 96.71%to to99.0% 99.0%

99.1% and above 99.1% and above

Hawaii and the U.S. Virgin Islands reported median 2010 occupancy in excess of 99%. Note, however, that this result may be skewed by the relatively small sample size of less than 15 properties. The next–highest-performing territory was Puerto Rico, which reported 2010 median occupancy of 99.7% for the 88 properties in the pool. New York, New Jersey and California, which cumulatively total more than 3,200 stabilized properties and more than $15 billion in net equity, reported occupancy equal to or greater than 97.5% in 2010. These states had a significant impact on the overall national occupancy rate, and may be significantly responsible for the increase in 2010 median occupancy. Were these states removed from the national portfolio, the 2010 median occupancy rate would decrease from 96.6% to 96.1%.

A CohnReznick Report | 37

Figure 3.7.2(B) illustrates each state’s 2010 median debt coverage ratio. The majority of states reported 2010 median DCRs between 1.15 and 1.30. All states reported 2010 median DCR in excess of 1.10 except for Georgia and Idaho, which reported statewide median debt coverage ratios of 1.05 and 1.04, respectively. Of particular significance is that 2010 is the only year of the three-year survey period in which none of the 50 states, Puerto Rico, Guam or the U.S. Virgin Islands reported overall median DCR of less than 1.00. In 2008, three states – Idaho, Indiana and Ohio – reported below break-even median DCRs ranging from 0.85 to 0.99, followed by two states whose median DCRs were just above 1.00. In 2010, Indiana and Ohio properties operated with median DCRs in excess of 1.13, and Idaho had a 1.04 median DCR. Based on our analysis of the data, DCRs in all states appear to be improving, and the fact that no states were operating below break-even in 2010 is an encouraging sign for the entire low-income housing tax credit industry. Whether the same level of positive performance can be sustained in the future remains to be seen.

2010 Median Debt Coverage Ratio by State

Figure 3.7.2(B)

ME MT

ND MN

IA NE NV

UT CA

KY NC TN

OK NM

AR MS

LA FL

1.00 or below 1.00 or below

1.011.01 to 1.10 to 1.10

38 | The Low-Income Housing Tax Credit Program

NH VT MA RI CT

to 1.20 1.111.11 to 1.20

1.21 to 1.30 1.21 to 1.30 1.31 and above

1.31 and above

NJ DE MD DC

Hawaii, Guam and the U.S. Virgin Islands each had 2010 median per-unit cash flow of greater than $1,400. This is not a surprising result, as properties in these states reported near 100% median occupancy in each of the survey years. Idaho, Georgia and Connecticut had median cash flow per unit of less than $200 and were the three lowest-performing states, measured solely by per-unit cash flow. Despite the general trend of improvement that coincides with the overall portfolio trend from 2008 to 2010, these three statesâ&#x20AC;&#x2122; cash flows were less than half of the overall portfolio median.

2010 Median Per Unit Cash Flow by State

Figure 3.7.2(C)

ME MT

ND MN

NH VT MA RI CT

NJ DE MD DC

KY NC TN

OK NM

AR MS

LA FL

$0 to $100 $0 to $100

$101$101 to $250 to $250

$251 $251to to$500 $500

$501 $1000 $501 to to $1,000

$1001 above $1,001 andand above

3.7.3. Geographic Segmentation Analysis â&#x20AC;&#x201C; by Top 10 MSAs Figure 3.7.3 summarizes the operating performance data for stabilized properties segmented by the top 10 MSAs, which were selected based on the aggregate net equity invested in the properties located within each MSA. However, Appendix F includes property operating and financial performance data for all metropolitan statistical areas where a meaningful sample size could be obtained. The top 10 MSAs collectively represent 31.9% of the total stabilized net equity surveyed.

A CohnReznick Report | 39

Nearly all top 10 MSAs, apart from Miami and Detroit, had median occupancy rates that were consistent with or greater than the national portfolio median. As previously discussed, occupancy rates decreased slightly from 2008 to 2009 and increased in 2010 while remaining consistently strong within a very narrow band of 96% to 97%. Apart from Philadelphia, Miami and Detroit, the top 10 MSAs reported median debt coverage ratios that were in line with or greater than the national portfolio medians.

Operating Performance by Top 10 MSAs Median Physical Occupancy

Figure 3.7.3

Median Debt Coverage Ratio

2008

2009

2010

2008

2009

2010

New York-Northern New Jersey-Long Island, NY-NJ-PA

97.4%

97.7%

97.8%

1.21

1.44

1.46

$251

$557

$675

Los Angeles-Long Beach-Santa Ana, CA

98.0%

97.5%

97.9%

1.53

1.52

1.44

$982

$1,028

$1,042

San FranciscoOakland-Fremont, CA

97.2%

97.4%

97.5%

1.19

1.23

1.21

$628

$706

$688

Chicago-JolietNaperville, IL-IN-WI

96.4%

97.0%

96.5%

1.16

1.22

1.24

$294

$418

$531

PhiladelphiaCamden-Wilmington, PA-NJ-DE-MD

96.7%

97.0%

96.7%

1.09

1.12

1.16

$80

$120

$211

WashingtonArlington-Alexandria, DC-VA-MD-WV

97.0%

96.5%

96.8%

1.17

1.24

1.25

$581

$736

$764

Miami-Fort LauderdalePompano Beach, FL

96.7%

95.9%

95.6%

1.17

1.15

1.18

$281

$303

$474

Seattle-TacomaBellevue, WA

97.2%

96.4%

97.0%

1.24

1.23

1.19

$441

$474

$395

Boston-CambridgeQuincy, MA-NH

96.5%

97.0%

97.2%

1.15

1.21

1.25

$378

$639

$723

Detroit-WarrenLivonia, MI

94.0%

94.1%

95.0%

0.85

1.00

0.99

-$275

-$54

-$14

Overall

96.4%

96.3%

96.6%

1.15

1.21

1.24

$250

$341

$419

40 | The Low-Income Housing Tax Credit Program

2008

2009

2010

Median Per Unit Cash Flow

Detroit’s performance metrics place it at the bottom of the top 10 MSAs. The financial performance of housing tax credit properties in Detroit, while improving, is still significantly less favorable than that of the other top 10 MSAs and the nationwide portfolio. The 200 stabilized properties located within the Detroit metropolitan area represent 29.4% of all the surveyed properties in the State of Michigan. While the Detroit properties account for a large portion of the Michigan properties in the surveyed portfolio, the Detroit properties are faring significantly worse than the balance of Michigan’s portfolio. Perhaps more so than any other major American city, Detroit has felt the effects of the recent national recession in the form of elevated unemployment rates stemming from decline of its manufacturing base and overall economy.12 However, the trend between 2008 and 2010 suggests that the operating performance of housing credit properties located in Detroit has been improving, albeit at a slower pace than the overall portfolio. While CohnReznick has not collected 2011 and early 2012 performance data, positive signs continue to be observed in the conventional rental market sector in Detroit. In ranking 44 major metropolitan markets, Marcus & Millichap identified Detroit as having moved up from the 42nd to 38th position among the top MSAs and projects that employment-generated demand will further increase occupancy by 60 basis points in 2012 to 95%, accompanied by a 3.2% increase in effective rents.13

ureau of Labor Statistics: The unemployment rate in Wayne County reached a high point of 18.2% in July 2009 and has since B improved, but remained high at 11.0% as of May 2012. 13 Source: Marcus & Millichap. “2012 National Apartment Report.” 12

A CohnReznick Report | 41

Chapter 4:

Underperforming Properties

iven the tremendous demand and historically high occupancy rates associated with affordable housing units, CohnReznick is often asked how such properties can fail. In its effort to provide discussion points related to the failure rate of affordable housing, CohnReznick analyzed the data obtained from respondents by isolating a cohort of properties as “underperforming” versus “performing.” Underperforming properties are those reporting any of the following criteria: • Physical occupancy levels below 90% • A debt coverage ratio below 1.00 • Insufficient cash flow to cover operating expenses. The properties identified as underperforming have been further segmented to identify those that have reported operating impediments versus those that have reported technical impediments. Operating underperformance refers to instances where a property suffers from low occupancy, operating deficits or physical plant issues such as deferred maintenance. Herein lies the similarity between housing tax credit properties and market-rate or any other real estate rental assets: Housing tax credit properties are effectively a real estate asset group unto themselves and therefore are measured, in some ways, in the same manner that their non-tax-incented counterparts are measured. Syndicators and investors commonly maintain what is referred to as a “watch list” in connection with their asset management procedures. Watch lists track assets meeting certain performance measures so that “problem” properties can be closely monitored. Watch list criteria can vary from syndicator to syndicator; however, most respondents adopted the criteria established by the Affordable Housing Investors Council (AHIC)14 as a baseline for measuring underperformance. Pursuant to AHIC standards, a property investment reporting below 90% economic occupancy or below 1.00 DCR should be placed on a watch list for close monitoring, in addition to being observed for other performance difficulties. 14

http://www.ahic.org.

42 | The Low-Income Housing Tax Credit Program

Because housing tax credit properties must conform to certain statutory requirements, they are also subject to rigorous compliance tests and layers of oversight by the IRS and state housing agencies. Because of the added burden of statutory requirements, housing credit properties bear higher administrative costs than non-tax-incented real estate counterparts. Accordingly, a property failing to comply with housing tax credit program requirements is characterized in the report as a property that is technically underperforming. There are limitations to CohnReznickâ&#x20AC;&#x2122;s analysis because, as in most studies preceding it, the focus is on stabilized properties. Thus, the report does not address construction or lease-up risks, nor does it offer indicators related to properties that were unable to come to fruition because of financing feasibility issues or other development-stage challenges. The fact that some housing tax credit properties underperformed can be attributed to a number of reasons. Specifically, low occupancy can be attributed to: soft market conditions, competitive properties in close proximity to the housing credit property, ineffective tenant screening resulting in high eviction rates and deteriorating property conditions rendering the property uninhabitable or inferior to its competition. Although this chapter explores the symptoms of underperformance of housing tax credit properties, diagnosing the underlying causes for underperformance is beyond the scope of this report as the sheer size of the reportâ&#x20AC;&#x2122;s sample renders a deeper dive infeasible.

4.1. Operating Underperformance In addition to the static information presented, the report presents analysis related to both the duration and magnitude of underperformance. Clearly, chronic underperformance deserves more attention than pure operating volatility, as persistent underperformance results in a more likely loss on investment, while operating volatility may result only in a temporary drop in occupancy or DCR. In addition, the distribution of underperformance is an interesting indicator. For instance, assuming all other indicators remain constant, it would be natural to be concerned about a portfolio where 35% of the properties report below 1.00 DCR with an average per-unit annual deficit of $100 â&#x20AC;&#x201C; in comparison to a portfolio where only 15% of the properties report below 1.00 DCR with annual deficits that are much higher. In practice, however, the magnitude of operating deficits has proven to be more important than the number of properties reporting deficits.

4.1.1. Underperformance in 2010 As reflected in Figure 4.1.1, calendar year 2010 operations indicate that 9.5% (measured by net equity) of the capital invested in stabilized housing tax credit properties operated at below 90% physical occupancy, 24.6% operated at or below break-even and 24.7% incurred operating deficits. As previously noted, the incidence of properties reporting negative cash flow generally corresponds to the incidence of properties reporting debt coverage below 1.00, with the exception of properties financed exclusively with soft debt. Furthermore, while approximately 9.5% of housing tax credit properties operated below 90% occupancy in 2010, 24.6% failed to achieve break-even operations during the same period. The aforementioned spread indicates that high occupancy does not necessarily guarantee strong financial performance. While low occupancy is often a key driver of operating deficits, these deficits may be the result of a multitude of issues, including spikes in operating expenses, rent concessions and higher than normal turnover.

A CohnReznick Report | 43

In our analysis of property size, CohnReznick isolated the cohort of underperforming properties as a percentage of the total number of properties (as opposed to a percentage of net equity). Comparison of the two columns in Figure 4.1.1 indicates that, as expected, properties with a higher number of units tend to withstand operating challenges more easily by distributing certain fixed costs among a larger number of units. In addition, equity investors tend to pay a premium to invest in larger properties, and premium pricing translates to lower levels of debt per apartment.

2010 Underperformance

Figure 4.1.1 % of Net Equity

Below 90% Physical Occupancy

% of Properties

9.5%

12.5%

Below 1.00 DCR

24.6%

27.5%

Below $0 Per Unit Cash Flow

24.7%

27.5%

CohnReznick performed a number of analyses to ascertain whether certain property characteristics, other than location, tend to have any bearing on high incidence of underperformance. Of the characteristics CohnReznick tested, other than property size, no other combinations (such as tenancy, development and credit type) had a material effect on property performance metrics. As shown in Figure 4.1.1(A), smaller-scale properties containing 50 or fewer apartment units were found to have a disproportionate share in the incidence of operating deficits.

2010 Per Unit Cash Flow Performance by Property Size

Figure 4.1.1(A)

Number of Units Per Property

301–10,000 201–300 101–200 51–100 26–50 0–25

20%

40% n % Underperforming

44 | The Low-Income Housing Tax Credit Program

60% n % Performing

80%

100%

4.1.2. Historical Trend (2008–2010) Industry observers have expressed concern about the potentially negative effect of national economic conditions on the health of housing tax credit inventory during the period from 2008 to 2010. However, the data that CohnReznick collected for 2008 to 2010 consistently suggest that this was not the case. For 2008 and 2009, the percentage of underperforming properties was largely consistent with that of prerecession years. As such, during 2008, 11.9% of the properties surveyed operated at below 90% occupancy, 32.2% reported below 1.00 debt coverage and 33.4% generated net operating income that was insufficient to cover their mandatory debt service payments and replacement reserve contributions. Consistent with the overall median performance data presented in Chapter 3, the number of underperforming properties decreased for two consecutive years after 2008, reaching what appears to be a historically low level in 2010.

Underperformance (2008–2010)

Figure 4.1.2

% of Net Equity 2008

2009

2010

Below 90% Physical Occupancy

11.9%

12.6%

9.5%

Below 1.00 DCR

32.2%

27.6%

24.6%

Below $0 Per Unit Cash Flow

33.4%

27.8%

24.7%

In addition to the factors discussed in Chapter 2.4, a variety of other reasons may explain why housing tax credit properties fared better than their market-rate counterparts during this period. • Demand for affordable housing, which has historically been in short supply, tends to move in the opposite direction of adverse economic conditions. • A significant portion of the housing tax credit properties surveyed either benefit from property-based rental assistance or serve tenants who possess rental assistance vouchers. For these tenants, regardless of the gap between their ability to pay pro forma rents, tenants are responsible for contributing no more than 30% of their adjusted gross income toward rent, with some or the entire gap being covered by rental assistance. The more subsidy a property has, the more insulated it becomes from adverse economic conditions. • Many housing tax credit properties benefit from below-market interest rates or soft financing. Properties receiving additional subsidies in the form of below-market rates or soft financing and increasingly high levels of housing credit equity permit project owners to charge rents that can be significantly below market rate. While housing tax credit properties are usually underwritten with relatively small operating cushions, housing tax credit properties are clearly benefiting from lower debt burdens than their market-rate competitors and earlier generations of housing tax credit properties.

A CohnReznick Report | 45

4.1.3 Chronic Underperformance To account for the fact that housing tax credit properties, like other types of real estate, are vulnerable to operating volatility in varying degrees, CohnReznick assessed the incidence of underperformance in consecutive years. We summarized properties with less than 90% physical occupancy consecutively using three time periods: 2008-2010, 2009-2010 and 2010. Across the entire portfolio, only 6.1% of properties reported below 90% occupancy during both 2009 and 2010, and an even more modest number, 3.9%, reported below 90% occupancy consecutively from 2008 to 2010. As with occupancy, properties reporting debt coverage below 1.00 and negative cash flow for sustained periods of time represent a more modest fraction of total properties than the ratio of properties reporting operating deficits for a single year.

Chronic Underperformance

Figure 4.1.3 % of Net Equity 2010

Below 90% Physical Occupancy

2010 and 2009

2010, 2009 and 2008

9.5%

6.1%

3.9%

Below 1.00 DCR

24.6%

17.3%

13.9%

Below $0 Per Unit Cash Flow

24.7%

16.9%

13.7%

46 | The Low-Income Housing Tax Credit Program

4.1.4. Magnitude of Underperformance CohnReznick plotted the distribution of properties reporting underperformance by occupancy rate, DCR and per-unit cash flow in order to ascertain the magnitude of underperformance. Of the 9.5% properties reporting below 90% occupancy during 2010, 7.6% are clustered within the 80% to 90% range. Measured by physical occupancy, only 1.9% of the surveyed stabilized properties were considered extreme underperformers reporting less than 80% occupancy for 2010.

Distribution of 2010 Physical Occupancy

Figure 4.1.4(A)

60% 49.9%

50% 40% 30% 20% 10% 0%

0.3%

0.4%

1.2%

< 60%

60% to 69%

70% to 79%

1.8% 80% to 84%

21.0%

19.7%

90% to 94%

95% to 97%

5.7% 85% to 89%

> 97%

Occupancy Ranges

A significant indicator of the magnitude of underperforming affordable housing properties is evidenced by the fact that less than 1% of housing tax credit properties placed in service have been lost to foreclosure. The low risk of foreclosure, given the fragile nature of housing tax credit property cash flows, can be understood by focusing on the incidence of chronic underperforming properties as well as the relatively nominal level of negative cash flow deficits. While 25% of surveyed housing credit properties experienced negative cash flow in 2010, as shown in Figure 4.1.3, these properties tended to recover financially fairly quickly, as temporary issues are managed by their owner-operators. Given the fact that only 14% of the surveyed properties report cash flow deficits that CohnReznick regards as material (i.e., more than $400 per unit), it appears that in most cases the propertyâ&#x20AC;&#x2122;s developer managed deficits through a combination of withdrawal from reserves, fee deferrals, short-term suspension of replacement reserve deposits and loans to the property under guarantee obligations. On rare occasions, syndicators call upon investors to make additional capital contributions.

A CohnReznick Report | 47

Distribution of 2010 Debt Coverage Ratio

Figure 4.1.4(B)

60% 50.5%

50% 40% 30% 20% 10% 0%

14.6% 8.2%

< .50

1.1%

2.0%

2.6%

.50 to .59

.60 to .69

.70 to .79

4.6%

6.1%

.80 to .89

.90 to .99

1.00 to 1.14

10.3%

1.15 to 1.25

> 1.25

DCR Ranges

Distribution of 2010 Per Unit Cash Flow

Figure 4.1.4(C)

70%

65.6%

60% 50% 40% 30% 20%

14.0%

10% 0%

< -$400

4.6%

6.2%

-$400 to -$199

-$200 to -$1

9.6%

$0 to $200

> $200

Per Unit Cash Flow Ranges

A subset of respondents provided information about deficit funding. Respondents were asked to indicate the main sources of deficit funding, as well as identify the main source of funding. While there are multiple funding sources for any given property, the table below identifies the most common sources of funding.

48 | The Low-Income Housing Tax Credit Program

Operating Deficit Funding Sources for 2010 Property Deficits

Figure 4.1.4(D)

8.4% 1.4% 46.4%

■ GP Advance. . . . . . . . . . . . . . 29.1% ■ Lower Tier Reserve. . . . . . . . 46.4%

14.7%

■ Accrual of Fees. . . . . . . . . . . . 8.4% ■ Upper Tier Reserves. . . . . . . . 1.4% ■ Other. . . . . . . . . . . . . . . . . . . . . . 14.7%

29.1%

4.1.5. Underperformance by State Following is a series of maps illustrating the percentage of 2010 underperformance by state.

2010 Occupancy Underperformance by State (Percent Below 90% Physical Occupancy)

Figure 4.1.5(A)

ME MT

ND MN

IA NE NV

UT CA

NH VT MA RI CT

NJ DE MD DC

KY NC TN

OK NM

AR MS

LA FL

and below 6.1% to6.1% to 10.0% 6.0% and 6.0% below 10.0%

10.1% to 16.0% 10.1% to 16.0%

16.1% 16.1%toto60.0% 60.0%

60.1% and and above 60.1% above

A CohnReznick Report | 49

2010 Debt Coverage Ratio Underperformance by State (Percent Below 1.00 DCR)

Figure 4.1.5(B)

ME MT

ND MN

IA NE NV CO

UT CA

NH VT MA RI CT

NJ DE MD DC

KY NC TN

OK NM

AR MS

LA FL

17.0% and17.0% belowand below 17.1% to17.1% 23.0%to 23.0%

50 | The Low-Income Housing Tax Credit Program

23.1% to 29.0% 23.1% to 29.0%

29.1% to 29.1% to 35.0% 35.0%

35.1% above 35.1% andand above

2010 Per Unit Cash Flow Underperformance by State (Percent Below $0 Per Unit Cash Flow)

Figure 4.1.5(c)

ME MT

ND MN

IA NE NV

UT CA

NH VT MA RI CT

NJ DE MD DC

KY NC TN

OK NM

AR MS

LA FL

15.0% and15.0% belowand below 15.1% to15.1% 22.0% to 22.0%

22.1% to 28.0% 22.1% to 28.0%

28.1% 28.1%toto34.0% 34.0%

34.1% and above 34.1% and above

States where a greater share of properties failed to achieve 90% occupancy also tend to struggle with a greater share of properties reporting below break-even operations, relative to the national portfolio. For instance, 24% of Idaho properties and 20% of Michigan properties were less than 90% occupied during 2010. These two states had 36% and 38% of their respective portfolios reporting less than a 1.0 debt coverage ratio. Properties in a few other states like Georgia and Indiana, while having slightly less occupancy underperformance challenges compared to Idaho or Michigan, appear to have more financial underperformance challenges. In both Georgia and Indiana, more than 40% of the survey properties incurred operating deficits in 2010, though at the state median level none of the 50 states reported below 1.0 DCR in 2010. The following three figures present the incidence of underperformance by state in three consecutive years, 2008-2010. As noted, 3.9% of all surveyed properties that had at least three years of stabilized operating history reported below 90% occupancy consecutively from 2008 to 2010. States with the most significant negative occupancy underperformance include Idaho (17.5%), Michigan (12.3%), West Virginia (11.1%) and Georgia

A CohnReznick Report | 51

(11.0%). The percentages in the parentheses following each state as well as those in Figures 4.1.5(D) and (F) represent the incidence of properties that were less than 90% occupied in each of the three surveyed years, out of the total surveyed properties in each state that had at least three years of occupancy data.

Chronic Occupancy Underperformance by State (Below 90% Occupancy in All Three Years 2008â&#x20AC;&#x201C;2010)

Figure 4.1.5(d)

ME MT

ND MN

OR SD

IA NE NV

UT CA

NH VT MA RI CT

NJ DE MD DC

KY NC TN

OK NM

AR MS

LA FL

1.8% and below1.8% and below 1.9% to 3.9% 1.9% to 3.9% 4.0% to 8.0% 4.0% to 8.0%

8.1% to 19.0% 8.1% to 19.0%

19.1% andand above 19.1% above

Across the surveyed portfolio, nearly 14% of properties incurred operating deficits in all three years, 2008-2010. Multiyear persistent deficits suggest that the operational challenges of this subset of properties may be more difficult to correct. Not surprisingly, chronic financial DCR underperformance was most seen in Georgia (23.6%), Idaho (32.3%), Indiana (28.5%) and Michigan (30.7%).

52 | The Low-Income Housing Tax Credit Program

Chronic DCR Underperformance by State (Below 1.00 DCR in All Three Years 2008â&#x20AC;&#x201C;2010)

Figure 4.1.5(e)

ME MT

ND MN

OR SD

IA NE NV

UT CA

NH VT MA RI CT

NJ DE MD DC

KY NC TN

OK NM

AR MS

LA FL

6.0% and 6.0% below 6.1% to 10.0% and below 6.1% to 10.0%

10.1% to 18.0% 10.1% to 28.0%

18.1% 18.1%toto27.0% 34.0%

27.1% and above 27.1% and above

de minimus minimus desample size sample size

A CohnReznick Report | 53

Chronic Cash Flow Underperformance by State (Below $0 Per Unit Cash Flow in All Three Years 2008â&#x20AC;&#x201C;2010) Figure 4.1.5(F)

ME MT

ND MN

OR SD

IA NE NV

UT CA

NH VT MA RI CT

NJ DE MD DC

KY NC TN

OK NM

AR MS

LA FL

7.0% and 7.0% below 13.0%to 13.0% and below 7.1% to 7.1%

13.1% to 19.0% 13.1% to 19.0%

19.1% 19.1%to to27.0% 27.0%

27.1% and above 27.1% and above

4.2. Foreclosure The most significant investment risk for housing tax credit investors relates to foreclosure. If the owner of a qualifying housing tax credit project forfeits title to the property because of foreclosure or by tendering a deed in lieu of foreclosure, the transfer is treated as a sale of the property. As a technical matter, this transfer generates housing tax credit recapture. A recapture event prompted by foreclosure results in the loss of one-third of the housing credits previously claimed in addition to 100% of any future housing tax credits. Thus, while foreclosure of housing tax credit properties has been rare, the potential impact to investors can be financially significant. Historically, properties lost to foreclosure reported large and sustained cash flow deficits. The incidence of chronic deficits may be attributed to low occupancy levels, poor sponsorship and defective construction, among other issues. However, in large part because of the flexibility and variability with which affordable housing investments can be financially supported or restructured, a remarkably low number of properties are foreclosed in any given year. CohnReznick asked respondents to report the number of properties they have lost to foreclosure, including circumstances in which a deed may have been tendered in lieu of

54 | The Low-Income Housing Tax Credit Program

foreclosure. Respondents reported that out of 17,118 properties surveyed, a total of 98 properties were foreclosed and, of that number, almost half were foreclosed during the period 2008–2010. This number translates to an aggregate foreclosure rate of 0.57% calculated by number of properties. As previously noted, however, we believe the number of foreclosures may be understated because CohnReznick was unable to obtain data it might have obtained in previous years from syndication firms that have since left the business or become inactive. CohnReznick has reason to believe, strictly on an anecdotal basis, that the incidence of property foreclosure has been higher among these firms than has been the case for the rest of the industry. However, because we lack precise information concerning the size and number of foreclosures in such firms’ respective portfolios, any estimate we might make on the potential impact to the overall industry data would require speculation on our part. CohnReznick believes that the firms we surveyed represent the core of the housing tax credit industry and that their care in financing and asset managing their investments is an important part of why the foreclosure rate of housing tax credit properties continues to be so low. CohnReznick plotted the cumulative number of foreclosures on a yearly basis. The year in which foreclosures occurred was reported for 81 of the 98 foreclosed properties. To derive the yearly cumulative rate, CohnReznick divided the number of foreclosures through year end by the total number of properties placed in service on or before the corresponding year and distributed the “missing 17” properties evenly over the years.

Cumulative Foreclosure Rate by Year

Figure 4.2.1

Cumulative Foreclosure Rate

0.70% 0.60% 0.50% 0.40% 0.30% 0.20% 0.10% 0.00% 1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

While the increasing annual rate of foreclosure might be a cause for concern, the foreclosure rate needs to be presented in its proper context. • CohnReznick’s industry experience leads us to believe that the foreclosure rate reported in the early years of the program may have been artificially low. This is due, in part, to the fact that most foreclosures occur toward the latter years of the 15-year housing tax credit compliance period. The foreclosure rate in the early years of the program might have also been masked, to some degree, by the propensity of housing tax credit syndicators to support troubled properties until the end of the housing tax credit period or the point at which investors’ financial losses are minimal.

A CohnReznick Report | 55

• Some of the foreclosures reported during 2008-2010 appear to have had more to do with financially distressed developers than with fundamental weakness in the underlying properties in question. • The most recent increase in the incidence of foreclosures from 2008-2010 appears to have begun to dissipate. CohnReznick notes that, although corresponding information is not presented in Figure 4.2.1, survey respondents reported a total of five additional incidences of foreclosures from January 2011 through approximately the second quarter of 2011. The number of foreclosures in 2011 is not yet fully reported, but would appear to be substantially lower than the number of foreclosures in 2010. • Further, while foreclosure can be a catastrophic event for developers and lenders, the financial losses to investors tend to be much less significant than in a conventional foreclosure. Based on the data collected, tax credit properties were foreclosed, on average, in year 10.8 of the 15-year compliance period. While a foreclosure results in the loss of one-third of the housing credits previously claimed in addition to 100% of any future housing tax credits, tax credit recapture is not the end result, as the developer and guarantors are often obligated to make investors whole pursuant to a recapture guarantee. As a practical matter, the effectiveness of the guarantee will, in part, depend on the creditworthiness of the developer and guarantors. • Finally, CohnReznick notes that, at 0.57%, the rate of foreclosure in housing tax credit properties is much lower than is the case for any other real estate investment with which we are familiar.

4.3. Technical Underperformance In order for housing tax credit properties to maintain their status as qualified housing tax credit properties, they must be operated in conformity with a number of statutory provisions collectively referred to as the “compliance rules.” While there are numerous provisions of Internal Revenue Code §42 that govern how the program is regulated, the most significant are the rules around tenant qualification and rent limitations. While a full explanation of the aforementioned rules is not within the scope of this report, the following provisions are significant components of IRC §42: • In order for a prospective tenant to move into a housing tax credit property, the applicant must present written verification of household income and assets, and household income cannot exceed 60% of the area median income level for similarly sized households. • For tenants that meet the income qualification requirements, the property cannot charge rents that exceed 30% of the statutory maximum level of income based on household size. Pursuant to IRC §42, income limits and rent restrictions must be carefully monitored by the state housing credit agencies; consequently, any failure to comply with the program rules must, if not resolved in an appropriate period of time, be reported to the IRS. Survey participants reported that only 65 of the surveyed properties incurred a material level of noncompliance, yielding a nominal rate of 0.5%. However, we caution that, as is the case with the foreclosure rate analysis, the rate of noncompliance may be understated because of incomplete data.

56 | The Low-Income Housing Tax Credit Program

Chapter 5:

Fund Investment Performance 5.1. Introduction

n addition to operational data for property-level investments, survey respondents were asked to supply CohnReznick with performance data for every low-income housing tax credit fund that they had syndicated. The fund-level performance data analysis assesses the track record of funds comprised of low-income housing tax credit properties in terms of delivering the originally projected yield and credits to investors. Generally speaking, an investor contemplating investing in housing tax credits can choose from one of two investment approaches: a direct investment or a syndicated investment. The direct investment approach is typically feasible only for sophisticated investors with internal resources dedicated to the acquisition, underwriting and asset management of housing tax credit properties; therefore, this approach is favored by a minority of institutional investors. The syndicated investment approach enables investors to invest in a fund organized and managed by a third-party intermediary known as a syndicator. There are two primary investment options when working with a syndicator: proprietary fund and multi-investor fund investments. In both fund types, the syndicator originates potential property investments, performs underwriting and presents the potential investment to investors. Proprietary funds are typically sought out by a single investor with a desire for a higher level of control over the investment. The Community Reinvestment Act requires banks to make investments in areas in which they collect deposits, and they consequently received CRA “credit” for doing so. Therefore, one of the primary investment motivations for banks is to earn CRA “credit” for their investments. Proprietary funds are a common investment option for institutions that want to invest their capital in specific locations and negotiate higher yields from their investments. The principal advantage of a multi-investor fund is risk diversification. A multi-investor fund can be composed of 2 to 15 investors, all of whom share risk and rewards based upon their proportional equity contribution to the fund. Whether the fund has two or many investors, certain tax credit funds are credit-enhanced either by the syndicator or, more typically, by a third-party insurance company. An additional advantage of investing in a creditenhanced housing credit investment, other than a guaranteed return, is the availability of a method of accounting for the investment known as the effective yield method. However, the disadvantage of a guaranteed fund is that a substantial portion of the investor’s capital is used to finance the guarantee fee, resulting in a substantially lower investment yield. Traditionally, approximately 20% of all tax credit investments were structured with a minimum yield guarantee. More recently, minimum-yield guarantees have become relatively infrequent, largely because of the lack of creditworthy guarantors. Low-income housing tax credit investors are effectively purchasing a financial asset in the form of a stream of tax benefits (consisting of passive losses associated with depreciation and mortgage interest deductions and tax credits). Investors do not anticipate receiving

A CohnReznick Report | 57

cash flow distributions, because housing tax credit properties are generally underwritten to operate at or slightly above break-even and developers or syndicators are generally the recipients of cash flow surpluses. In general, housing tax credits are realized on a straight-line basis over a 10-year period. Tax credits not delivered to investors in the first year because of construction or lease-up delays are typically realized in the 11th year (unless there is a permanent tax credit shortfall, which is often covered by basis adjustors); the investor’s capital account is written off when the investor exits the partnership, making an investor’s loss neutral in nominal terms. The yield from housing credit investments is generally measured by the investment’s after-tax internal rate of return (IRR). The IRR is a function of the amount and timing of the projected housing credits and profit/loss versus the timing of the investor’s equity pay-in. Most housing credit investments are structured with one or more “true-up” provisions to assist with yield maintenance. For instance, a loss of the time value of credits can be compensated for by a so-called adjustor provision that reduces the investor’s remaining capital contributions to maintain the projected yield. It is important to note that, in addition to the timing of tax credit realization, the composition of the tax benefits (the relative proportion of tax credits to tax losses) is equally important to investors. Investors who are sensitive to the negative impact of losses on financial statement earnings are more inclined to invest in 9% tax credit properties with low debt leverage and less inclined to invest in 4% credit tax-exempt bond transactions that are highly leveraged. Seventeen survey respondents provided data for 902 low-income housing tax credit funds. For purposes of this analysis, we removed 62 funds that were closed in 1994 or earlier, as the property investments of these funds were already beyond their 15-year compliance periods as of the data collection period of this report. Figure 5.1 illustrates the remaining 840 funds that closed in or later than 1995 organized by fund type and segmented by gross equity and low-income housing tax credits. Note, the average age of the 840 funds is eight years as of the data collection period of this report.

Total Surveyed Gross Equity by Fund Type (Post 1994 Funds)

Figure 5.1

3.8%

37.0%

■ Multi-investor. . . . . . . . . . . . . . 59.2% ■ Proprietary.. . . . . . . . . . . . . . . . 37.0% ■ Guaranteed. . . . . . . . . . . . . . . . 3.8% 59.2%

58 | The Low-Income Housing Tax Credit Program

The 398 multi-investor funds in our sample account for 59.2% of the surveyed fund portfolio gross equity and had an average fund size of $73.2 million of gross equity. The 416 proprietary funds in the pool sample account for 37.0% of the total fund portfolio gross equity, with an average fund size of $44.1 million of gross equity. The difference in the average size of these funds is driven by the fact that multi-investor funds are typically larger to accommodate the investment appetite of multiple investors.

5.2. Fund Yields Figure 5.2(A) illustrates the historical relationship between tax credit pricing and fund investment yields. The chart clearly demonstrates an upward movement in tax credit pricing with corresponding decrease in yield. In addition, investment yields, as measured by their after-tax internal rates of return, steadily decreased from the early 1990s through 2006 with few exceptions. Figure 5.2(A)

14.0%

$1.20

12.0%

$1.00

10.0%

$0.80

8.0%

$0.60

6.0%

$0.40

4.0%

$0.20

2.0%

Gross Equity Price

fund Yield

Gross Equity Price vs. Fund Yield by Year

$0.00 1995

1997 1996

1999 1998

2001 2000

2003 2002

Annual Median Yield

2005 2004

2007 2006

2009 2008

2010

Gross Equity Price

Several major aftershocks in the housing credit market had the effect of significantly lowering demand for housing credit investments during 2007–2009. Dramatic financial and organizational changes within the two largest housing credit investors, Fannie Mae and Freddie Mac, in 2006 and 2007 occasioned their exit from the housing credit equity market in 2007 and 2008. In addition to the loss of these government-sponsored enterprises (GSEs) as investors, the devaluation of mortgage securities and subsequent collapse of financial markets severely decreased the demand for tax credit investment among the nation’s largest financial institutions. The cumulative effect of losing the GSEs as investors and losses in the banking sector resulted in a 50% cumulative drop in equity demand from the market highs observed in 2006–2009.

A CohnReznick Report | 59

Figure 5.2(B) illustrates the historical relationship between housing tax credit fund yields and 10-year Treasury security yields (adjusted for an after-tax rate equivalent of a 35% tax rate). The chart depicts the median originally projected housing tax credit yield by year and the annual trend in 10-year Treasury security yields. While 2006 and 2007 housing credit fund yields approached Treasury yields, they increased and subsequently diverted significantly over the next three years.

Surveyed Housing Tax Credit Fund Yield vs. 10-Year Treasury Security Rate (after tax equivalent)

Figure 5.2(B)

14.0%

Fund Yield

12.0% 10.0% 8.0% 6.0% 4.0% 2.0% 1995

1997 1996

1999 1998

2001 2000

2003 2002

A nnual Median Yield of Surveyed Housing Tax Credit Funds

2005 2004

2007 2006

2009 2008

2010

0-Year Treasury Yield 1 (after tax equivalent)

5.3. Yield Variance Analysis Important to consider is the performance of housing tax credit funds with respect to actual income tax benefits versus originally projected benefits. Investment performance is expressed in terms of yield (calculated based on a quarterly after-tax internal rate of return), overall tax credit delivery, and the initial years of tax credit delivery relative to originally projected amounts. Yield variance measures the difference between the originally projected yield at investment closing and the most current yield projection as of the survey date, generally as of December 31, 2010. Positive variances indicate greater than projected yield. On a weighted average basis (where yield variances for individual funds are aggregated and weighted by equity), survey respondents reported a positive 5.97% variance in meeting yield targets. We removed housing credit funds with credit enhancement (â&#x20AC;&#x153;guaranteed fundsâ&#x20AC;?) from this analysis, because guaranteed funds are structured with yield maintenance mechanisms that ensure a predictable yield to investors.

60 | The Low-Income Housing Tax Credit Program

While yield is a significant factor for housing credit investors, the individual components of yield computations have a major bearing on yield calculations. Yield can be maintained naturally or “manually” by pre-negotiated investment provisions in a number of ways: An investor can receive a more favorable yield as a result of an underperforming portfolio generating higher losses; equity pay-in schedules can postpone capital contributions from investors; and so-called adjustor provisions, under which remaining investor capital contributions are reduced to the extent necessary in order to re-establish the ratio of capital to benefits, work to ultimately restore the investment’s yield to projected levels. Figure 5.3 illustrates the yield variances in housing credit funds based on the year in which the funds were closed.

Fund Yield Variance by Year

Figure 5.3

20.0% 15.0%

Fund Yield Variance

10.0% 5.0% 0.0% -5.0% -10.0% -15.0% -20.0%

1995 11 Funds

1997 23 Funds

1996 18 Funds

1999 41 Funds

1998 25 Funds

2001 52 Funds

2000 44 Funds

2003 57 Funds

2002 48 Funds

2005 58 Funds

2004 54 Funds

2007 65 Funds

2006 52 Funds

2009 45 Funds

2008 47 Funds

2010 39 Funds

This graph illustrates the fact that 109 of the 680 funds CohnReznick studied reported negative yield variances. We totaled the negative yield variances relative to the overall number of funds closed in each year and found that the years in which funds with the highest incidence of negative variances were syndicated were 2002, 2004 and 2007.

5.4. Housing Credit Variance Analysis Consistent with CohnReznick’s industry experience, the survey data we examined demonstrate that the aggregate average variance in fund yields has been less than 0.50%.

A CohnReznick Report | 61

The average housing credit investment derives the majority of its benefits from housing credits with the balance from passive losses. Because housing tax credits are calculated based on qualified development costs, a property’s future delivery of tax credits is highly predictable. In this context, the timing of tax credit delivery is more likely to create variances, because delays in the construction and lease-up of housing credit properties may result in delayed delivery of housing credits. Our data suggest that such delays, not uncommon in the early years of the program, have become less frequent as the industry’s underwriting capability has become more efficient.

Housing Credit Delivery Variance by Investment Type Total Housing Credit Delivery Variance

1st Year Housing Credit Delivery Variance

2nd Year Housing Credit Delivery Variance

Figure 5.4.1 3rd Year Housing Credit Delivery Variance

Total

-0.4%

-16.3%

-13.9%

-8.5%

Proprietary

-0.1%

-7.8%

-12.8%

-6.8%

Multi-investor

-0.7%

-21.1%

-14.5%

-9.6%

Guaranteed

0.2%

-11.0%

-14.4%

-4.9%

Figure 5.4.2 graphically illustrates the trend in actual versus projected first-, second- and third- year housing tax credits.

Initial Years’ Housing Credit Delivery Variance by Year Fund Closed

Figure 5.4.2

Housing Tax Credit Delivery Variance

30% 15% 0% 15% 30% 45% 60%

1995

1996

1997

1998

1999

■ 1st Year LIHTC Variance

62 | The Low-Income Housing Tax Credit Program

2000

2001

2002

2003

2004

■ 2nd Year LIHTC Variance

2005

2006

2007

2008

■ 3rd Year LIHTC Variance

2009

As illustrated in Figure 5.4.2, survey respondents have historically projected higher than actual tax credits in the first few years; however, more recently, respondents have been more accurate at projecting the first few years’ credits. This trend was consistently observed across the pool of survey respondents. For instance, Enterprise reported that out of 191 housing tax credit properties it syndicated during 2007–2009, 34% were leased up on time, 42% were ahead of schedule and the remaining 24% were behind schedule, thereby incurring negative initial-year credit delivery variances.15 In addition to the fact that housing credits are a function of development costs, they are realized at a steady level once tax credit occupancy is achieved. To the extent that the original projection of total housing tax credits or the timing of their delivery falls short, investors are entitled to be “made whole” by adjustor provisions. Finally, properties that struggle to maintain break-even occupancy tend to generate higher passive losses than initially projected. Increased loss deductions have the effect of increasing yield, but the negative impact of losses on financial statements generally makes them unwelcome.

Source: Enterprise Community Investment, Inc. “Asset Management LIHTC Portfolio Trends Analysis – November 2010.”

A CohnReznick Report | 63

Chapter 6:

Portfolio Composition

his chapter summarizes the composition of the stabilized property sample consisting of 15,399 housing tax credit properties by various segmentation factors. Unless otherwise noted, percentages are expressed on the basis of stabilized net equity.

6.1. Portfolio Composition â&#x20AC;&#x201C; by Property Age The number of years that have transpired since a housing tax credit property was first and last placed in service is meaningful from an operating perspective. Older properties often have physical plant issues or face market competition from more recently developed properties. The composition of the housing tax credit properties surveyed is heavily weighted (48%) toward properties last placed in service within the past five years. This is a reflection, in part, on the programâ&#x20AC;&#x2122;s slow initial activity and the fact that, with exceptions in select years, 100% of the authorized national housing tax credit allocation has been used. Figure 6.1 displays the amount of net equity invested in surveyed properties according to the year that the corresponding property investments in the survey sample were placed in service.

Net Equity by Year Placed-in-Service

Figure 6.1

$8.0M

Housing Credit Net Equity

$7.0M $6.0M $5.0M $4.0M $3.0M $2.0M $1.0M $0.0 1990

1992 1991

1994 1993

1996 1995

1998 1997

2000 1999

2002 2001

2004 2003

2006 2005

2008 2007

2010 2009

The median age of stabilized properties in the CohnReznick database is nine years (placed in service in 2003). The following is a graph indicating the portfolio composition as a percentage of stabilized properties by age group.

64 | The Low-Income Housing Tax Credit Program

Percent Net Equity by Property Age (Years since Placed-in-Service, as of 12/31/2010)

Figure 6.1(A)

3.5% ■ 5 years or less.. . . . . . . . . . . . . 52.2%

13.5%

52.2%

■ 6–10 years. . . . . . . . . . . . . . . . . 30.8% ■ 11–15 years. . . . . . . . . . . . . . . . 13.5% ■ 16 years or older. . . . . . . . . . . 3.5%

30.8%

6.2. Portfolio Composition – by Property Size According to our survey, the average stabilized housing tax credit property is made up of 72 units. The following figure illustrates the trend of average housing tax credit units per property by year placed in service.

Average Project Size by Year Placed in Service

Figure 6.2

90 Number of Units/Property

80 70 60 50 40 30 20 10 0

1990

1992 1991

1994 1993

1996 1995

1998 1997

2000 1999

2002 2001

2004 2003

2006 2005

2008 2007

2010 2009

CohnReznick surveyed 4% properties averaging 116 units per property, while 9% properties averaged 59 units per property. The following table illustrates the trend of the size of both 4% and 9% stabilized properties.

A CohnReznick Report | 65

Average Project Size by Net Equity, Credit Type and Year Placed-in-Service

Figure 6.2(A)

$8.0M

Net Equity (in Millions)

$7.0M $6.0M $5.0M $4.0M $3.0M $2.0M $1.0M $0.0 1990

1992 1991

1994 1993

1996 1995

1998 1997

2000 1999

■ Average 9% Property Size

66 | The Low-Income Housing Tax Credit Program

2002 2001

2004 2003

2006 2005

■ Average 4% Property Size

2008 2007

2009

6.3. Portfolio Composition – by Investment Type In the housing tax credit industry, credits have been syndicated through the sale of equity investments in public funds, direct investments, proprietary funds and multi-investor funds. “Public funds” refers to the publicly registered offerings that were the major source of equity financing in the early years of the housing tax credit program. Beginning in the early 1990s, institutional investors began to represent the dominant share of the housing tax credit equity market, making public funds increasingly rare and no longer used to raise capital for this sector. “Direct investments” refers to investments made by a single investor directly into a project partnership as opposed to investing through a fund managed by a third party. Direct investments make up a smaller portion of the market because they require the use of internal resources to monitor real estate operations and compliance with housing tax credit program rules. Currently, most equity investments are made through third-party intermediaries or syndicators who raise investor capital, acquire equity investments in housing tax credit projects and provide long-term asset management services.

Percent Net Equity by Investment Type

6.3%

■ Direct. . . . . . . . . . . . . . . . . . . . . . . 6.3%

5.2%

54.5%

Figure 6.3

■ Multi-investor. . . . . . . . . . . . . . 54.5% 0.7%

■ Proprietary.. . . . . . . . . . . . . . . . 33.3% ■ Public. . . . . . . . . . . . . . . . . . . . . . . 0.7%

33.3%

■ Not Specified. . . . . . . . . . . . . . . 5.2%

CohnReznick notes that property investments made by multi-investor funds constitute the majority of the properties surveyed. Survey respondents indicated that multi-investor funds represented 54.6% of the total equity financing on a stabilized net equity basis. Proprietary fund investments account for the second highest market share, with 33.3% of net equity. In recent years, as the housing credit equity market declined sharply, investors still active in the housing tax credit market placed a disproportionate amount of their capital through proprietary funds. CohnReznick notes that this trend began to reverse in 2010 with the recovery of the equity market.

A CohnReznick Report | 67

6.4. Portfolio Composition – by Credit Type The housing tax credit statute provides for two types of housing tax credits – 9% and 4% housing tax credits. Projects that are conventionally financed and are awarded housing tax credit allocations are eligible for 9% credits. Generally speaking, an owner of a housing tax credit property may claim housing tax credits equal to 9% of the project’s qualified costs each year for 10 years. Conversely, properties that are financed in whole or in part by the issuance of tax-exempt bonds may claim a 4% tax credit for 10 years based, again, on qualified housing expenditures. As a general matter, 9% projects are heavily financed with investor equity and thus have a modest level of hard debt financing to service. Tax-exempt bond projects that qualify for 4% credits generate significantly lower levels of tax credit equity and require higher debt levels (albeit at lower tax-exempt interest rates). Figure 6.4 displays the inventory of housing tax credit properties surveyed and divides them into 9% and 4% housing credit types. Participants did not distinguish between credit types for approximately 2,000 out of the 17,118 properties surveyed. As a result, CohnReznick excluded the 2,000 unidentified properties from this analysis. As shown below, 9% properties account for 71.3% of the net equity surveyed, with the remaining 28.7% invested in 4% properties.

Percent Net Equity by Credit Type

71.3%

28.7%

Figure 6.4

■ 9% Housing Tax Credits ■ 4% Housing Tax Credits

6.5. Portfolio Composition – By Development Type CohnReznick requested that respondents specify whether their property investments represented new construction or the rehabilitation of older, existing properties. Newly constructed properties accounted for 68.2% of the net equity surveyed, and rehabilitated properties accounted for 27.9% of net equity surveyed, with the remaining 4% being properties that represent the rehabilitation of historic structures or properties involving mixed development types. With respect to financing, our data reflect that the average new construction development was financed with $3.7 million of net equity, and the average rehabilitation property required $2.8 million of net equity.

68 | The Low-Income Housing Tax Credit Program

Percent Net Equity by Development Type

Figure 6.5

2.4%

1.5%

■ New Construction. . . . . . . . . 68.2% ■ Rehab. . . . . . . . . . . . . . . . . . . . . 27.9%

68.2%

■ Historic Rehab. . . . . . . . . . . . . . 2.4% 27.9%

■ Mixed. . . . . . . . . . . . . . . . . . . . . . . 1.5%

6.6. Portfolio Composition – By Tenancy Type “Special needs” displayed in Figure 6.6 refers to properties that are set aside for unique tenancy groups. This determination is based on the state’s assessment of its most critical housing needs—principally tenants with significant housing challenges, such as the homeless or tenants with physical handicaps. The data show that family properties account for 73.3% of all properties surveyed; senior properties account for 21.5%; special needs properties account for another 2.5%; and the remaining 2.7% either have mixed tenancies or respondents did not specify tenancy type in their responses. The data suggested that there has been no meaningful variance in terms of investment size among the various tenancy types.

Percent Net Equity by Tenancy Type

Figure 6.6

2.7%

2.5% 73.3%

■ Family.. . . . . . . . . . . . . . . . . . . . . 73.3% ■ Senior. . . . . . . . . . . . . . . . . . . . . . 21.5%

21.5%

■ Special Needs. . . . . . . . . . . . . . 2.5% ■ Mixed or Not Specified. . . . . 2.7%

A CohnReznick Report | 69

6.7. Portfolio Composition â&#x20AC;&#x201C; by Region The data reflected in Figure 6.7 summarizes the housing credit net equity by 12 CohnReznick-defined regions of the country. The regions in Figure 6.7 are arranged in descending order by greatest to least in total housing credit net equity. CohnReznick bundled the 50 states, the District of Columbia, Guam, the U.S. Virgin Islands and Puerto Rico into 12 regions consisting of similar geographic composition that most ideally grouped areas of the country. The regions are composed of the following:

Portfolio Composition by Region Number of Properties Region Number

Survey Total

Figure 6.7

Number of Units

Stabilizied Properties

Survey Total

Stabilizied Properties

Housing Credit Net Equity Survey Total

Stabilizied Properties

Total Housing Credits Survey Total

Stabilizied Properties

Region 1

2,437

2,215

187,862

170,013 $ 12,418,490,575 $ 10,734,283,353 $ 12,983,060,878 $ 11,112,035,053

Region 2

4,469

2,622 $ 222,981,536 $ 139,716,992 $ 311,641,942 $ 167,453,573

Region 3

229

207

9,194

8,308 $ 501,254,208 $ 429,595,451 $ 594,558,443 $ 500,530,163

Region 4

788

716

57,883

52,515 $ 2,814,098,250 $ 2,371,709,285 $ 3,325,049,639 $ 2,808,278,287

Region 5

588

529

28,021

23,912 $ 1,345,586,660 $ 1,112,170,776 $ 1,608,104,263 $ 1,343,679,514

Region 6

1,219

1,081

63,372

53,513 $ 2,938,908,795 $ 2,246,460,634 $ 3,249,299,591 $ 2,433,270,659

Region 7

2,717

2,447

191,677

170,755 $ 8,947,816,963 $ 7,347,578,906 $ 10,873,359,834 $ 8,761,727,408

Region 8

1,338

1,226

123,421

107,997 $ 4,281,728,653 $ 3,565,145,154 $ 5,050,959,624 $ 4,296,055,907

Region 9

1,828

1,622

181,509

157,474 $ 7,995,325,242 $ 6,493,879,063 $ 9,394,008,999 $ 7,630,196,093

Region 10

1,897

1,712

119,966

106,790 $ 4,516,058,259 $ 3,861,332,569 $ 5,331,090,420 $ 4,552,484,846

Region 11

3,818

3,437

282,094

248,108 $ 15,515,157,446 $ 12,747,593,847 $ 19,128,578,259 $ 15,714,700,408

Region 12

116

101

9,831

8,241 $ 640,389,440 $ 484,449,104 $ 1,006,361,855 $ 711,930,795

Region 1: CA, OR, WA; Region 2: AK, HI; Region 3: ID, MT, WY; Region 4: AZ, CO, NM, NV, UT; Region 5: MN, ND, SD; Region 6: IA, KS, NE, MO; Region 7: IN, IL, MI, OH, WI; Region 8: AR, OK, TX; Region 9: AL, FL, GA, LA, MS; Region 10: KY, NC, SC, TN, VA, WV; Region 11: CT, DC, DE, MA, MD, ME, NH, NJ, NY, PA, RI, VT; Region 12: GU, PR, VI.

70 | The Low-Income Housing Tax Credit Program

Percent Net Equity by Region

Figure 6.7(A)

1%< 1% 21% 5% 25%

■ Region 3. . . . . . . . . . . . . . . . . . . . . . 1%

■ Region 4. . . . . . . . . . . . . . . . . . . . . . 5% ■ Region 5. . . . . . . . . . . . . . . . . . . . . . 2%

■ Region 6. . . . . . . . . . . . . . . . . . . . . . 4%

7% 14%

13%

■ Region 1. . . . . . . . . . . . . . . . . . . . . 21% ■ Region 2. . . . . . . . . . . . . . . . . . . . < 1%

■ Region 7. . . . . . . . . . . . . . . . . . . . . 14%

■ Region 8. . . . . . . . . . . . . . . . . . . . . . 7% ■ Region 9. . . . . . . . . . . . . . . . . . . . . 13% ■ Region 10.. . . . . . . . . . . . . . . . . . . . 8% ■ Region 11.. . . . . . . . . . . . . . . . . . . 25%

Region 11 represents the greatest amount of total net equity and accounts for nearly 25% of the entire portfolio. Region 1 encompasses the second highest amount of equity at approximately 20% of the portfolio, followed by Region 7, which equals roughly 15% of the net equity portfolio. Region 2 with the least amount of net equity in the portfolio, is made up of only 54 properties (44 of which are stabilized) and represents less than 1% of the total net equity portfolio.

Average Project Size by Region

Figure 6.7(B)

Average Number of Units/Property

120 100 80

88 77

20 0

Region 1

Region 3 Region 2

Region 5 Region 4

Region 7 Region 6

Region 9 Region 8

Region 11 Region 10

A CohnReznick Report | 71

6.8. Portfolio Composition – by State We segmented the portfolio property data by all 50 states, Puerto Rico, the U.S. Virgin Islands and Guam. The top five states, measured by total net equity, are California, New York, Texas, Florida and Illinois, collectively accounting for more than 42% of the overall portfolio net equity. Delaware, South Dakota, Hawaii, Guam and the U.S. Virgin Islands represent the bottom five in terms of overall net equity, each representing less than 0.2% of the overall portfolio.

Percent Net Equity by State

Figure 6.8

ME MT

ND MN

IA NE NV

UT CA

NH VT MA RI CT

NJ DE MD DC

KY NC TN

OK NM

AR MS

LA FL

1.3% and 1.3 below and below

1.4% to1.4% 4.0% to 4.0%

4.1% to 10.0% 4.1% to 10.0%

10.1% to 16.0% 16.0%

16.1% andabove above 16.1% and

6.9. Portfolio Composition – by MSA The question has been raised from time to time as to whether a disproportionate level of housing tax credits are being allocated to the nation’s largest cities. Figure 6.9 illustrates the capital concentration data for properties located in the top 10 metropolitan statistical areas. The general concept of a MSA is that of a large population center, together with adjacent communities that have a high degree of social and economic integration. MSAs may comprise one or more entire counties, except in New England, where cities and towns are the basic geographic units.

72 | The Low-Income Housing Tax Credit Program

As shown below, 26.3% of the total housing tax credit equity we surveyed was concentrated in properties located within the 10 MSAs. CohnReznick notes that the percentage of total housing tax credit equity closely correlates to the aggregate population of residents in the 10 MSAs versus the rest of the U.S. population.

Net Equity Concentration among Top 10 MSAâ&#x20AC;&#x2122;s $7B

Figure 6.9

$6.5

Housing Net Equity

$6B $5B $4B $2.7

$3B

$2.1

$1.9

$2B

$1.5

$1.3

$1B $0

NY-NJ-PA

San Francisco, CA

Los Angeles, CA

Philadelphia, PA

Chicago, IL

$1.0

Miami, FL

Washington, DC

$1.0

$0.8

Boston, MA Seattle, WA

Detroit, MI

Appendix F outlines the overall portfolio composition of all MSAs.

A CohnReznick Report | 73

Appendix A

Acknowledgments CohnReznick would like to thank the following organizations for contributing data and financial support for the study: • AEGON USA Realty Advisors

• Michel Associates

• Alliant Capital

• Midwest Housing Equity Group

• Bank of America

• Mountain Plains Equity Group

• Boston Capital

• National Development Council

• Boston Financial Investment Management

• National Equity Fund

• Centerline Capital Group

• Northern New England Housing Investment Fund

• Citibank

• Ohio Capital Corporation for Housing

• City Real Estate Advisors

• PNC Multifamily Capital

• Community Affordable Housing Equity Corporation

• Raymond James

• Enterprise Community Investment • First Sterling • Great Lakes Capital Fund

• Red Capital Group • RBC Capital Markets • Red Stone Equity Partners

• Housing Vermont

• The Richman Group Affordable Housing Corporation

• Hunt Capital Partners

• Stratford Capital

• Hudson Housing

• The Summit Group

• John Hancock

• SunTrust Community Development Corporation

• J.P. Morgan Chase • Massachusetts Housing Investment Corporation • Merritt Community Capital

• Union Bank of California • U.S. Bank • WNC Associates

* CohnReznick also would like to thank the National Association of State and Local Equity Funds for making a financial contribution on behalf of its member organizations.

74 | The Low-Income Housing Tax Credit Program

Appendix B

Survey Methodology

his report represents the second in a series of studies undertaken by the CohnReznick concerning the Low Income Housing Tax Credit program. In March 2011, CohnReznick transmitted data requests to 40 organizations, including all active housing credit syndicators known to the firm and a number of the nation’s largest housing credit investors. Investor respondents were asked to provide data limited to direct investments and fund-level performance to mitigate what would otherwise be a large overlap of properties’ data assembled from participating syndicators’ portfolios. CohnReznick’s first study was published in August 2011. A two-phase approach allowed CohnReznick to supply much needed current industry data while still operating within the timeframe necessary to perform an increasingly rigorous analysis of the data. While the first and second study results were released separately, data were requested and collected once. Thirty-two organizations chose to participate in the August 2011 study, and an additional six organizations participated in the current study, resulting in a 95% overall response rate. All outputs in the current study have been updated to include the additional participant’s data. CohnReznick believes that 17,118 properties, the sample size represented, is in excess of 70% of the housing tax credit properties placed in service since 1986 that are being actively assetmanaged by syndicators and/or investors. We suspect that the gap between Reznick’s database and 100% of all properties ever syndicated was largely a result of defunct syndicators, as well as properties placed in service in the earlier years of the housing tax credit program that have approached the expiration of their respective compliance periods and disposed of in ways that caused them to be “cycled out” of the program. Most important, we believe that the sample size represented in the study provides a statistically meaningful basis for our analysis and findings. To substantiate the estimate of our data coverage, we benchmarked our sample size against the housing tax credit property database maintained by the U.S. Department of Housing and Urban Development.

CohnReznick Survey vs. HUD Database (1995–2011) 1800

Number of Properties

1500 1200 900 600 300 0

1995

1997 1996

1999 1998

2001 2000

2003 2002

■ CohnReznick Survey

2005 2004

2007 2006

2009 2008

2011 2010

■ HUD Database

A CohnReznick Report | 75

Data for this report were compiled and analyzed with the support of Integratec, CohnReznick’s affiliated real estate services and software solutions company. Integratec’s involvement in the study was to provide data rollup, filtering, calculation and aggregation/ segmentation services per CohnReznick’s instructions. Integratec’s goal was to provide CohnReznick with high-quality output in a format that would facilitate further data analysis and presentation. The remainder of the appendix outlines the timeline and methodology we followed to carry out the study.

Data Collection A participant solicitation letter was mailed to 40 organizations on March 10, 2011, announcing and requesting support for a housing credit industry study to be undertaken by the CohnReznick with the assistance of Integratec. The solicitation letter was followed by a Microsoft Excel data collection template, along with a confidentiality and nondisclosure agreement and acknowledgement of participation. Respondents were initially requested to return the collection template no later than May 20, 2011. A few participating respondents who indicated that they lacked sufficient time to complete the survey properly were offered a deadline extension. All contacts, whether made by telephone, email or mail, were recorded in response contact logs. The six additional participants in this study submitted their data following publication of the August 2011 report. For participating organizations that are existing clients of Integratec, we requested that each respondent grant consent for use of portions of the dataset that Integratec manages on each respondent’s behalf. Upon receipt of the consent form, Integratec loaded requested data to our data collection template, then sent it back to respondent for review and completion of any remaining fields. Respondents that are not clients of Integratec were instructed to complete the data collection template directly. In either case, data were transmitted directly by the respondents to CohnReznick.

Questionnaire Design The following table shows the main data points requested from each participating investor and syndicator. Instructions were attached to each collection field to minimize interpretations. Contact information of CohnReznick professionals was supplied along with the collection template for questions related to the data request. CohnReznick believes that data fields included in the collection template have been carefully designed to allow the study to be informative, essential and influential. Where applicable, audited financial data were requested and were represented as having been furnished in that form. However, neither CohnReznick nor Integratec performed any independent validation of the data nor did we ascertain that the data were indeed audited.

76 | The Low-Income Housing Tax Credit Program

Data Fields

Definition/Explanation

Property Investment Identification Fund identification

Provide the name of the fund or a unique identification number from participant’s database which permits future identification.

Investment type

Choose one from the following categories that best describes the investment type of the fund: direct, proprietary, multi-investor, guaranteed or public.

Property identification

Provide the name of the property or a unique identification number from participant’s database which permits future identification.

Property address

Provide the property’s street address, city, state (in 2-letter abbreviation) and 5-digit zip code.

Location type

Choose one from the following categories that best describes the location characteristics of the property’s location: urban, suburban or rural.

Name of MSA

Provide the name of the Metropolitan or Micropolitan Statistical Area where the property is located.

Credit type

Choose one from the following categories: 4%, 9% or 4%/9%.

Property status

Choose one from the following categories that best describes the current status of the property: preconstruction, construction, lease-up (completed construction but hasn’t achieved 100% qualified occupancy), prestabilization (achieved 100% qualified occupancy but not yet reached stabilization benchmarks), stabilization or disposition.

Development type

Choose one from the following categories that best describes the type of the development: new construction, rehabilitation, acquisition/rehabilitation, historic rehabilitation, or other.

Tenancy type

Choose one from the following categories that best describes the population the property has committed to serve: family, senior, special needs, family with special needs set-aside, senior with special needs set-aside, or other.

Number of units

Provide the total number of housing units the property consists of.

Number of housing credit units

Provide the total number of units that are eligible for federal low-income housing tax credits.

Project-based rental assistance

Indicate whether the property benefits from project-based rental assistance.

Type of project-based rental assistance

If the property benefits from project-based rental assistance, choose one from the following categories that best describes the type of assistance: Section 8, Rural Development, ACC, or other. If more than one type of rental assistance program is involved, choose the primary one that covers the majority of the subsidized units.

Hard debt

Indicate whether the property has permanent debt that requires mandatory debt service payments regardless of available cash flow.

Hard debt ratio

Provide the percent of property total development costs financed with permanent hard debt. Leave the field blank if the property has no permanent hard debt.

Year placed in service

Provide the 4-digit year in which the property was or is projected to be placed in service.

First year of credit delivery

Provide the 4-dight year in which credit delivery first commenced (or is projected to commence).

Total housing credit net equity

Provide the amount of total net equity associated with federal low-income housing tax credits and (to be) contributed to the property investment.

Total housing credits to investor

Provide the 10-year total amount of federal low-income housing tax credits (projected to be) available to the investor limited partner.

A CohnReznick Report | 77

Data Fields

Definition/Explanation

Property Performance Data Physical occupancy for years 2008 to 2010

Provide the average physical occupancy (average occupancy over the period during which the property had stabilized operations) for each of the last three years.

Debt coverage ratio for years 2008 to 2010

Provide the year-end debt coverage ratio (net operating income minus required replacement reserve contributions then divided by the mandatory debt service payments) for each of the last three years in accordance with audited financial statements. Choose “NA” if the property does not have any hard debt. For properties with partial year stabilized operations during a certain year, provide the average over the stabilized period.

Per-unit cash flow for years 2008 to 2010

Provide the year-end per-unit cash flow (net operating income minus required replacement reserve contributions and mandatory debt service payments, if any, then divided by the total number of units) for each of the last three years in accordance with audited financial statements. For properties with partial-year stabilized operations during a certain year, provide the average over the stabilized period.

On AHIC watch list

If the watch list criteria published by the Affordable Housing Investor’s Council has been adopted, indicate whether the property is currently on the AHIC watch list.

On internal watch list

Indicate whether the property is currently on participant’s internal watch list.

Operating deficit funding source(s)

If the property incurred operating deficits during 2010, choose from the following funding sources (choose all that apply): investor capital call, upper-tier reserve or syndicator advance, lower-tier reserve or general partner advance, and debt restructuring.

Primary operating deficit funding source

If the property incurred operating deficits during 2010, choose the single largest funding source from the following: investor capital call, upper-tier reserve or syndicator advance, lower-tier reserve or general partner advance, and debt restructuring.

Foreclosure

Indicate whether the property has been foreclosed upon or a deed in lieu of foreclosure has been tendered.

Year of foreclosure

If the property has been foreclosed upon or a deed in lieu of foreclosure has been tendered, specify the year in which the foreclosure event occurred.

Noncompliance

Indicate whether the property has ever suffered from credit loss due to noncompliance issues arising from the IRS or state audit.

Fund Identification and Performance Data Fund identification

Provide the name of the fund or a unique identification number from participant’s database that permits future identification.

Investment type

Choose one from the following categories that best describes the investment type of the fund: direct, proprietary, multi-investor, guaranteed or public.

Year closed

Provide the 4-digit year in which the fund was closed.

Total gross equity

Provide the amount of total gross equity raised from tax credit equity investor.

Total housing credits to investor

Provide the 10-year total amount of federal low-income housing tax credits (projected to be) available to the investor limited partner.

Total historic tax credits to investor

Provide the total amount of federal historic rehabilitation tax credits (projected to be) available to the investor limited partner.

Total energy tax credits to investor

Provide the total amount of federal renewable energy tax credits (projected to be) available to the investor limited partner.

Total other tax credits to investor

Provide the total amount of any other federal or state tax credits (projected to be) available to the investor limited partner.

Original IRR

Provide the quarterly investor IRR projected at the time of fund closing, with necessary adjustment for property removals/additions

Current IRR

Provide the quarterly investor IRR according to the latest asset management or investor report

78 | The Low-Income Housing Tax Credit Program

Data Fields

Definition/Explanation

As of date for current IRR

Specify the as of date, in the form of xx/xx/xxxx, for the current IRR

Total projected housing credits

Provide the 10-year total amount of federal low-income housing tax credits projected to be available to the investor limited partner at fund closing.

Total actual housing credits

Provide the 10-year total amount of federal low-income housing tax credits delivered (or projected to be delivered based on the latest information such as 8609s) to the investor limited partner.

As of date for actual housing credits

Specify the as of date, in the form of xx/xx/xxxx, for the actual housing credits

Total projected initial years of housing credits

Provide the annual amount of federal low-income housing tax credits projected to be available to the investor limited partner at fund closing; specify the amount for each of the first three years.

Total actual initial years of housing credits

Provide the annual amount of federal low-income housing tax credits projected delivered (or projected to be delivered based on the latest information such as 8609s) to the investor limited partner; specify the amount for each of the first three years.

Current working capital reserve balance

Provide the current balance of working capital reserve established for the fund.

As of date for actual housing credits

Specify the as of date, in the form of xx/xx/xxxx, for the current working capital reserve balance.

Data Processing The receipt of a completed survey questionnaire and any relevant comments made by the respondents were recorded in the contact logs. All questionnaires were first reviewed for data completeness and systematic errors for reasons such as misinterpretation. If questionnaires were returned with incomplete data, respondents were contacted immediately to determine the possibility of providing missing data and, in limited circumstances, the consequences of participants being unable to accommodate the entire data request. Other follow-up activities were conducted to ensure data integrity. Upon completion of the first round processing, data were compiled, filtered and normalized. Each data element provided was then uploaded to a MS SQL Server database jointly maintained by Integratec and CohnReznick. The database was built in a completely confidential manner to ensure that no individual data points or groups of individual data points could be attributed to any data provider. The data were loaded into the MS SQL Server to ensure the consistency of field data types and to allow for flexible and repeatable calculation. Data entered into the database were checked for arithmetical errors, and flagged for any large discrepancies between the current and previous yearsâ&#x20AC;&#x2122; data for trend warnings. Based on industry standards and a lengthy, programmatic filtering system designed by CohnReznick, outliers that could skew the study results were screened and later removed from the affected calculations. Based on predefined data outputs and calculation definitions, Integratec developed SQL scripts to perform calculations and group datasets (e.g., linking zip codes to applicable MSAs) for segmentation analysis. Because of the lack of a median function in the MS SQL Server, Integratec created stored procedures to calculate median values. Median calculation accuracy was independently checked against Excel median function calculations. Finally, aggregated data and outputs were re-exported into an Excel template for further testing and quality control reviews.

A CohnReznick Report | 79

Appendix C

Glossary Credit type

There are two types of low-income housing tax credits under the Internal Revenue Code § 42: the 9% credits are available to support new construction or rehabilitation projects that are not considered federally subsidized; the 4% credits are available to support new construction or rehabilitation projects that are financed with tax-exempt bonds, or the acquisition costs of existing buildings. While the actual value varies based on a number of factors, the 9% and 4% credits are designed to subsidize 70% and 30% of the low-income unit costs in a project.

Debt coverage ratio

Net operating income (effective gross operating income minus operating expenses) minus required replacement reserve contributions, divided by mandatory debt service payments

Direct investment

Investors make equity investments directly into a property partnership as opposed to investing through a fund managed by a third-party intermediary.

Economic occupancy

Collected gross rental income divided by gross potential rental income

Foreclosure

The legal process by which a mortgagee or other lien holder obtains, either by court order or by operation of law, a termination of a mortgagor’s right to a property usually as a result of default

Guaranteed investment

Investors make equity investments to an investment fund (which, in turn, owns interest in multiple property partnerships) organized by a third-party intermediary. Under a guaranteed investment structure, the yield, as contractually agreed upon, is guaranteed by a creditworthy entity for a premium.

Metropolitan Statistical Areas (MSAs)

A geographical region with relatively high population density at its core and close economic ties throughout the area. MSAs are defined by the U.S. Office of Management and Budget, and used by the U.S. Census Bureau and other U.S. government agencies for statistical purposes.

Multi-investor investment

Multiple investors jointly make equity investments into an investment fund (which, in turn, owns interest in multiple property partnerships) organized by a third-party intermediary, and thus share investment benefits and risks.

Net equity

The amount of equity raised from “allocating” housing credits to investors. Net equity is distinguished from gross equity by excluding the “load” (i.e., fees charged by syndicators for underwriting and managing the investment fund) and the capital set aside for reserves.

80 | The Low-Income Housing Tax Credit Program

Physical occupancy

The number of occupied units divided by the total number of rentable units in a given property

Placed-in-service

The date when the property is ready for its intended use; a housing credit property can either claim credits beginning the year it is placed in service (provided that units are occupied by income qualified tenants) or defer the beginning of the credit period to the following year.

Proprietary investment

A single investor makes equity investments and assumes the limited partner role in an investment fund (which, in turn, owns interest in multiple property partnerships) organized by a thirdparty intermediary.

Public investment

Investment funds commonly seen in the early years (pre-early 1990s) of the housing credit program when investment capital was primarily derived from individual investors

Qualified occupancy

All of the housing credit units have been leased to tenants who have been income-certified and deemed eligible to occupy such units.

Recapture

Housing credit properties are subject to a 15-year compliance period, which extends five years beyond the credit period. Credits may be recaptured during the 15-year compliance period if the property ceases to qualify as a housing credit property or ceases to be occupied by qualified tenants. The amount of recapture will be calculated based on two-thirds of the previously claimed credits plus applicable interest charges.

Soft debt

Mortgage loans where payments are subject to available cash flow

Stabilized operations

Properties that have completed construction, achieved 100% qualified occupancy and closed on permanent financing

State allocating agencies

State or local agencies that have the authority to allocate federal low-income housing tax credits to a property

A CohnReznick Report | 81

Appendix D

Property Performance by State Operating Performance by State Median Physical Occupancy State

2008

2009

2010

Median Debt Coverage Ratio 2008

2009

Median Per Unit Cash Flow

2010

2008

2009

2010

96.0%

95.5%

96.9%

1.16

1.15

1.21

$505

$481

$642

95.0%

95.1%

95.5%

1.19

1.32

1.35

$201

$269

$297

96.0%

1.16

1.18

1.17

$179

$245

$250

95.5%

94.9%

95.8%

1.19

1.23

1.24

$227

$236

$304

97.9%

97.5%

97.9%

1.34

1.36

1.34

$753

$844

$795

96.5%

96.3%

97.2%

1.10

1.15

1.25

$352

$460

$691

96.6%

97.0%

1.09

1.18

1.19

$(87)

$233

$123

97.1%

96.9%

96.8%

1.10

1.28

1.25

$359

$705

$540

96.0%

97.0%

96.0%

1.20

1.16

1.19

$271

$240

$329 $329

95.0%

94.6%

95.0%

1.12

1.15

1.16

$231

$259

94.8%

95.3%

1.01

1.03

1.05

$18

$29

$67

94.4%

85.1%

89.4%

1.17

1.23

1.45

$548

$735

$1,482

99.0%

1.41

1.68

1.69

$1,586

$2,496

$2,422

94.4%

94.1%

95.0%

1.11

1.12

1.17

$165

$225

93.8%

94.4%

94.0%

0.95

0.99

1.04

$24

$(7)

$111

96.0%

96.7%

96.1%

1.11

1.22

1.27

$233

$385

$479

94.0%

94.4%

0.85

1.05

1.14

$(228)

$77

$212

96.0%

1.12

1.16

1.13

$234

$287

$216

96.5%

96.2%

96.1%

1.04

1.17

1.32

$230

$352

96.3%

96.0%

1.32

1.24

1.25

$389

$384

$503 $537

96.2%

96.7%

97.0%

1.17

1.27

$369

$421

96.8%

97.0%

1.20

1.22

1.26

$298

$387

$480

97.5%

97.4%

97.2%

1.29

1.38

1.40

$360

$432

$545

93.9%

94.0%

95.0%

1.01

1.07

1.11

$23

$154

$240

97.2%

97.0%

97.2%

1.26

1.31

1.36

$554

$688

$691

94.9%

95.3%

95.8%

1.16

1.14

1.21

$236

$224

$287

95.3%

95.0%

96.0%

1.25

1.13

1.32

$284

$207

$430

95.8%

94.4%

95.3%

1.08

1.18

1.27

$161

$391

$447

96.3%

95.1%

96.2%

1.21

1.18

1.17

$445

$274

$230

97.0%

97.2%

1.20

1.35

1.36

$259

$495

96.7%

97.4%

97.5%

1.08

1.19

1.25

$356

$411

$409

95.4%

95.8%

96.0%

1.15

1.24

$272

$216

$336

97.6%

97.0%

1.06

1.42

1.47

$129

$627

$749

97.4%

97.3%

97.6%

1.17

1.22

1.25

$333

$250

$517

95.8%

96.0%

96.6%

1.24

1.26

1.33

$495

$470

$598

96.5%

96.0%

95.3%

1.22

1.19

1.29

$265

$296

$465

97.3%

97.5%

1.26

1.45

1.52

$261

$528

$638 $200

96.2%

96.1%

0.99

1.05

1.13

$(53)

$62

95.7%

95.5%

96.0%

1.23

1.17

1.24

$288

$278

$277

96.8%

96.1%

96.4%

1.12

1.17

1.20

$268

$287

$372

82 | The Low-Income Housing Tax Credit Program

Median Physical Occupancy State

2008

2009

2010

Median Debt Coverage Ratio 2008

2009

Median Per Unit Cash Flow

2010

2008

2009

2010

97.0%

1.16

1.23

1.29

$84

$205

$256

99.9%

99.5%

99.7%

1.18

1.24

1.21

$379

$432

$494

97.0%

97.5%

1.18

1.22

1.20

$227

$433

$371

96.0%

96.8%

96.4%

1.13

1.17

1.24

$162

$289

$379

95.0%

95.5%

95.8%

1.17

1.20

1.30

$354

$258

$365

95.6%

94.0%

95.0%

1.01

1.08

1.12

$42

$169

$227

95.4%

95.1%

95.8%

1.14

1.19

1.23

$222

$323

$455

97.7%

97.0%

1.20

1.27

1.28

$494

$735

$571

96.3%

96.2%

97.0%

1.17

1.15

1.19

$319

$423

$416

99.0%

99.2%

2.39

2.20

2.09

$2,987

$2,539

$2,215

97.5%

97.1%

97.5%

1.13

1.29

1.26

$213

$611

$523

97.0%

96.2%

97.0%

1.24

1.25

1.26

$427

$454

$520

96.0%

95.9%

95.7%

1.03

1.14

1.17

$99

$352

$399

96.0%

95.6%

95.4%

1.18

1.22

1.14

$157

$233

$255

97.5%

96.0%

95.7%

1.15

1.14

$263

$270

$250

A CohnReznick Report | 83

Appendix E

Property Underperformance by State 2010 Operating and Chronic Underperformance by State

State

Period

Below 90% Physical Occupancy

Below 1.00 DCR

Below $0 Cash Flow

Equity %

2010

11.2%

39.7%

36.0%

Last 3 Years

N/A

34.5%

22.9%

2010

8.1%

20.2%

23.5%

Last 3 Years

5.1%

15.7%

16.0%

2010

18.5%

33.3%

31.6%

Last 3 Years

10.2%

21.2%

18.3%

2010

15.6%

32.4%

34.3%

Last 3 Years

6.7%

21.4%

23.9%

2010

4.7%

13.3%

14.0%

Last 3 Years

1.2%

5.8%

5.1%

2010

3.9%

18.6%

19.4%

Last 3 Years

0.2%

7.9%

8.3%

2010

5.0%

33.8%

37.1%

Last 3 Years

0.9%

15.2%

20.0%

2010

6.2%

21.9%

23.9%

Last 3 Years

0.8%

12.3%

11.1%

2010

4.8%

32.5%

31.4%

Last 3 Years

5.1%

16.6%

14.6%

2010

15.2%

28.6%

30.2%

Last 3 Years

10.1%

20.8%

18.7%

2010

17.9%

43.7%

42.6%

Last 3 Years

11.4%

24.6%

23.4%

66.0%

6.6%

5.9%

17.4%

32.0%

33.8%

2010

Last 3 Years

2010

Last 3 Years

8.2%

14.8%

15.7%

2010

24.4%

35.6%

34.9%

Last 3 Years

19.5%

37.6%

30.7%

2010

11.1%

30.1%

Last 3 Years

4.3%

12.9%

14.0%

2010

16.5%

41.5%

40.8%

Last 3 Years

9.3%

28.5%

27.4%

2010

17.0%

31.9%

32.4%

Last 3 Years

5.1%

19.6%

20.5%

2010

11.9%

30.1%

31.7%

Last 3 Years

4.4%

13.6%

19.7%

2010

8.3%

20.1%

18.5%

Last 3 Years

2.5%

8.6%

11.7%

2010

6.8%

27.4%

24.8%

Last 3 Years

2.4%

9.4%

8.8%

2010

7.2%

24.7%

22.8%

84 | The Low-Income Housing Tax Credit Program

State

Period

Below 90% Physical Occupancy

Below 1.00 DCR

Below $0 Cash Flow

Equity %

Last 3 Years

2.0%

9.3%

2010

8.8%

22.3%

20.3%

Last 3 Years

2.5%

8.0%

2.5%

2010

19.8%

37.8%

38.1%

Last 3 Years

12.6%

30.7%

31.6%

2010

7.8%

18.3%

19.8%

Last 3 Years

2.0%

15.1%

11.8%

2010

18.5%

27.7%

29.1%

Last 3 Years

9.7%

14.0%

14.3%

2010

13.6%

17.3%

20.3%

Last 3 Years

3.7%

15.7%

18.0%

2010

14.5%

24.2%

19.5%

Last 3 Years

4.0%

14.7%

11.7%

2010

6.3%

20.2%

18.4%

Last 3 Years

2.8%

12.4%

11.7%

2010

8.8%

17.9%

13.4%

Last 3 Years

3.2%

15.4%

11.9%

2010

16.1%

24.5%

30.4%

Last 3 Years

6.4%

10.0%

13.3%

2010

5.8%

9.4%

11.4%

Last 3 Years

N/A

5.2%

6.2%

2010

7.7%

28.3%

28.6%

Last 3 Years

3.0%

17.0%

16.2%

2010

11.3%

12.8%

11.5%

Last 3 Years

1.1%

7.2%

3.6%

2010

13.0%

23.7%

22.5%

Last 3 Years

2.6%

16.3%

16.1%

2010

4.1%

19.2%

19.8%

Last 3 Years

1.4%

9.0%

10.8%

2010

12.7%

36.7%

37.6%

Last 3 Years

5.8%

21.6%

23.2%

2010

13.6%

28.0%

25.0%

Last 3 Years

4.7%

7.1%

7.7%

2010

7.7%

19.8%

19.0%

Last 3 Years

2.1%

8.2%

2010

6.4%

25.3%

26.1%

Last 3 Years

2.0%

19.5%

14.7%

2010

2.5%

9.1%

8.2%

Last 3 Years

3.8%

2.5%

4.2%

2010

7.1%

30.0%

29.2%

Last 3 Years

N/A

14.5%

16.2%

2010

6.2%

28.1%

26.8%

Last 3 Years

2.5%

11.4%

14.9%

A CohnReznick Report | 85

State

Period

Below 90% Physical Occupancy

Below 1.00 DCR

Below $0 Cash Flow

Equity %

2010

13.6%

14.8%

Last 3 Years

2.7%

5.1%

4.2%

2010

18.4%

39.0%

36.8%

Last 3 Years

10.7%

32.5%

24.7%

2010

12.9%

29.1%

28.3%

Last 3 Years

5.5%

20.9%

17.2%

2010

5.7%

25.5%

24.7%

Last 3 Years

N/A

8.5%

7.8%

2010

11.2%

25.3%

24.9%

Last 3 Years

2.4%

17.3%

12.3%

2010

1.7%

18.6%

21.6%

Last 3 Years

N/A

2.3%

2010

5.3%

18.8%

19.5%

Last 3 Years

0.8%

9.5%

8.7%

2010

9.3%

31.0%

29.8%

Last 3 Years

3.3%

21.4%

15.4%

2010

21.0%

28.7%

24.9%

Last 3 Years

12.2%

18.0%

16.6%

2010

18.2%

38.4%

36.5%

Last 3 Years

2.0%

29.6%

28.8%

86 | The Low-Income Housing Tax Credit Program

15.4%

Appendix F

Property Performance by MSA Operating Performance by MSA Median Physical Occupancy State

MSA

2008

2009

2010

Median Debt Coverage Ratio 2008

2009

Median Per Unit Cash Flow

2010

2008

2009

2010

Alabama

BirminghamHoover, AL

97.8%

96.4%

95.8%

1.01

1.07

1.16

$42

$88

$200

Alabama

Daphne-FairhopeFoley, AL

88.3%

89.8%

94.4%

1.15

$238

Alabama

Florence-Muscle Shoals, AL

97.8%

96.6%

97.4%

1.77

1.71

1.43

$629

$602

$383

Alabama

Huntsville, AL

94.7%

94.6%

1.39

1.32

1.40

$406

$358

$337

Alabama

Mobile, AL

95.5%

97.8%

96.9%

1.08

1.32

1.28

$(63)

$574

$497 $222

Alabama

Montgomery, AL

95.0%

96.0%

1.06

1.23

1.11

$51

$283

Alabama

Tuscaloosa, AL

92.0%

95.6%

94.8%

0.72

1.22

1.35

$(256)

$177

$77

Alaska

Anchorage, AK

96.0%

95.5%

96.9%

1.16

1.15

1.21

$795

$364

$538

Arizona

Flagstaff, AZ

94.4%

96.9%

98.0%

1.21

1.41

1.37

$104

$340

$708

Arizona

Phoenix-MesaGlendale, AZ

93.0%

92.0%

92.1%

0.96

0.92

1.09

$(99)

$(24)

$(49)

Arizona

Prescott, AZ

95.4%

96.0%

97.7%

1.41

1.28

1.34

$330

$485

$345

Arizona

Show Low, AZ

97.3%

95.8%

95.5%

1.38

1.17

1.07

$590

$318

$190

Arizona

Sierra VistaDouglas, AZ

96.0%

98.5%

97.1%

1.45

1.27

1.26

$758

$293

$313 $322

Arizona

Tucson, AZ

95.7%

96.7%

96.0%

0.94

0.95

1.26

$152

$(118)

Arizona

Yuma, AZ

99.0%

97.5%

97.0%

1.76

1.55

1.42

$998

$789

$583

Arkansas

FayettevilleSpringdaleRogers, AR-MO

95.1%

97.0%

96.7%

1.34

1.43

1.38

$426

$508

$438

Arkansas

Fort Smith, AR-OK

91.7%

95.0%

92.3%

1.14

1.11

1.14

$155

$105

$144

Arkansas

Harrison, AR

97.8%

97.5%

98.3%

1.50

1.41

1.32

$333

$392

$399

Arkansas

Little Rock-North Little RockConway, AR

95.9%

95.0%

95.2%

1.04

1.08

1.02

$(47)

$63

$(7)

Arkansas

Mountain Home, AR

96.6%

97.2%

96.7%

1.19

1.33

1.23

$454

$444

$136

Arkansas

Texarkana, TXTexarkana, AR

94.9%

94.3%

92.5%

0.89

1.02

1.00

$530

$46

$(40)

California

BakersfieldDelano, CA

98.0%

98.8%

98.6%

1.34

1.27

1.48

$489

$531

$510

California

El Centro, CA

99.2%

99.3%

98.8%

1.39

1.22

1.25

$749

$517

$519

California

Eureka-ArcataFortuna, CA

95.1%

95.5%

97.8%

1.74

1.45

1.74

$1,016

$854

$1,025

California

Fresno, CA

96.3%

96.0%

97.5%

1.25

1.21

1.31

$690

$611

$647

California

HanfordCorcoran, CA

97.8%

97.4%

98.6%

1.29

1.28

1.38

$731

$987

$1,003

California

Los AngelesLong BeachSanta Ana, CA

98.0%

97.5%

97.9%

1.53

1.52

1.44

$982

$1,028

$1,042

California

MaderaChowchilla, CA

97.6%

97.0%

98.7%

1.39

1.93

1.59

$965

$807

$913

California

Merced, CA

98.2%

97.0%

1.49

1.27

1.35

$1,019

$397

$486

California

Modesto, CA

98.6%

98.0%

98.4%

1.29

1.09

1.40

$709

$187

$795

California

Oxnard-Thousand Oaks-Ventura, CA

99.0%

98.5%

98.9%

1.65

1.70

1.53

$2,078

$2,498

$2,002

A CohnReznick Report | 87

Median Physical Occupancy State

MSA

2008

2009

2010

Median Debt Coverage Ratio 2008

2009

Median Per Unit Cash Flow

2010

2008

2009

2010

California

Redding, CA

95.9%

97.8%

95.5%

1.18

1.40

1.06

$378

$559

$35

California

Riverside-San BernardinoOntario, CA

98.0%

1.31

1.44

1.38

$646

$894

$661

California

Sacramentoâ&#x20AC;&#x201D; Arden-Arcadeâ&#x20AC;&#x201D; Roseville, CA

96.8%

97.0%

1.17

1.18

$460

$456

$685

California

Salinas, CA

98.5%

98.7%

98.2%

1.23

1.34

$433

$828

$670

California

San DiegoCarlsbad-San Marcos, CA

97.9%

97.4%

97.9%

1.44

1.46

1.43

$918

$1,188

$1,079

California

San FranciscoOaklandFremont, CA

97.2%

97.4%

97.5%

1.19

1.23

1.21

$628

$706

$688

California

San JoseSunnyvaleSanta Clara, CA

97.8%

97.0%

97.5%

1.21

1.23

1.18

$594

$863

$545

California

San Luis ObispoPaso Robles, CA

100.0%

96.4%

99.7%

1.31

1.38

1.35

$879

$1,099

$1,061

California

Santa BarbaraSanta MariaGoleta, CA

99.6%

97.6%

99.0%

1.16

1.18

1.32

$677

$908

$1,266

California

Santa CruzWatsonville, CA

98.8%

98.0%

98.9%

1.71

1.82

$1,445

$1,505

$1,222

California

Santa RosaPetaluma, CA

98.0%

1.52

1.54

1.41

$879

$1,241

$974

California

Stockton, CA

95.9%

92.0%

93.8%

0.84

1.26

1.32

$255

$372

$391

California

Truckee-Grass Valley, CA

98.6%

98.9%

97.4%

1.35

1.16

1.39

$1,048

$398

$732

California

Vallejo-Fairfield, CA

95.5%

94.6%

95.7%

1.42

1.49

1.15

$845

$1,406

$739

California

VisaliaPorterville, CA

96.7%

96.5%

1.49

1.25

1.26

$717

$413

$486

Colorado

Boulder, CO

99.0%

97.2%

97.9%

1.30

1.31

1.40

$808

$980

$1,062

Colorado

Colorado Springs, CO

95.1%

94.6%

94.9%

1.07

1.11

1.14

$270

$445

$603

Colorado

Denver-AuroraBroomfield, CO

96.3%

96.0%

97.0%

1.07

1.10

1.18

$290

$344

$622

Colorado

Fort CollinsLoveland, CO

98.0%

95.8%

97.5%

1.33

1.35

1.42

$818

$783

$855

Colorado

Pueblo, CO

97.3%

97.8%

97.6%

2.36

1.66

1.64

$1,566

$2,137

$1,561

Connecticut

BridgeportStamfordNorwalk, CT

97.0%

96.8%

97.0%

1.10

1.26

1.12

$43

$74

$(84)

Connecticut

Hartford-West Hartford-East Hartford, CT

96.0%

96.1%

96.0%

1.04

1.14

1.05

$(143)

$233

$84

Connecticut

New HavenMilford, CT

97.0%

97.5%

96.0%

0.81

1.05

1.13

$(332)

$32

$(139)

Delaware

Dover, DE

96.0%

98.0%

97.3%

1.05

1.13

$101

$134

$206

Delaware

Seaford, DE

95.4%

96.0%

96.2%

1.45

1.27

1.28

$476

$348

$329

District of Columbia

WashingtonArlington-Alexandria, DC-VA-MD-WV

97.0%

96.5%

96.8%

1.17

1.24

1.25

$581

$736

$764

88 | The Low-Income Housing Tax Credit Program

Median Physical Occupancy State

MSA

2008

2009

2010

Median Debt Coverage Ratio 2008

2009

Median Per Unit Cash Flow

2010

2008

2009

2010

Florida

Cape Coral-Fort Myers, FL

95.3%

89.8%

93.5%

1.38

1.04

1.05

$497

$240

$(8)

Florida

Deltona-Daytona Beach-Ormond Beach, FL

96.1%

94.8%

95.7%

1.23

1.21

1.24

$207

$280

$218

$597

Florida

Gainesville, FL

95.0%

92.7%

94.4%

0.83

1.23

1.18

$322

$660

Florida

Jacksonville, FL

93.7%

95.7%

95.0%

1.02

1.05

$(177)

$70

$75

Florida

Lakeland-Winter Haven, FL

95.3%

95.0%

95.3%

0.89

0.90

1.14

$(222)

$(72)

$107

Florida

Miami-Fort LauderdalePompano Beach, FL

96.7%

95.9%

95.6%

1.17

1.15

1.18

$281

$303

$474

Florida

Naples-Marco Island, FL

87.3%

89.5%

93.4%

0.75

1.08

1.15

$(860)

$(658)

$329

Florida

North PortBradentonSarasota, FL

92.7%

91.0%

93.4%

0.94

0.99

1.12

$(116)

$(31)

$164

Florida

OrlandoKissimmeeSanford, FL

94.0%

93.0%

93.9%

1.27

1.19

1.20

$770

$627

$553

Florida

Pensacola-Ferry Pass-Brent, FL

92.1%

91.5%

93.9%

0.91

0.97

1.12

$(342)

$11

$(102)

Florida

Sebastian-Vero Beach, FL

91.7%

89.8%

93.2%

0.83

0.57

0.52

$(286)

$(972)

$(1,098)

Florida

Tampa-St. PetersburgClearwater, FL

94.4%

94.3%

1.18

1.27

1.24

$347

$464

$279

Georgia

Atlanta-Sandy Springs-Marietta, GA

94.1%

92.4%

93.5%

1.00

0.98

0.89

$42

$(38)

$(77)

Georgia

AugustaRichmond County, GA-SC

96.6%

97.6%

96.0%

0.97

1.15

1.18

$(44)

$144

$153

Georgia

Savannah, GA

91.0%

95.3%

96.0%

0.98

0.96

1.51

$(38)

$73

$898

Georgia

Valdosta, GA

95.7%

97.8%

98.2%

1.03

1.11

1.09

$(39)

$116

$104

Hawaii

Honolulu, HI

99.0%

1.36

1.64

1.54

$1,586

$2,022

$2,336

Idaho

Boise CityNampa, ID

95.1%

92.5%

95.8%

0.84

0.81

0.85

$(8)

$(383)

$(377)

Idaho

Coeur d'Alene, ID

95.5%

96.5%

94.9%

0.96

1.01

1.14

$(47)

$19

$258

Idaho

Lewiston, ID-WA

97.0%

97.8%

1.35

1.33

1.32

$761

$766

$617

Idaho

Twin Falls, ID

91.3%

95.6%

93.8%

0.90

1.14

1.41

$(263)

$190

$515

Illinois

Chicago-JolietNaperville, IL-IN-WI

96.4%

97.0%

96.5%

1.16

1.22

1.24

$294

$418

$531

Illinois

Ottawa-Streator, IL

99.6%

99.9%

95.8%

1.11

1.34

1.72

$506

$389

$910

Illinois

Peoria, IL

94.8%

98.0%

97.4%

0.98

1.06

1.60

$(42)

$242

$880

Illinois

Rockford, IL

94.4%

93.0%

92.9%

0.86

1.08

1.17

$(342)

$300

$670

Illinois

Springfield, IL

95.0%

92.5%

1.10

1.17

1.13

$215

$212

$386

Indiana

Evansville, IN-KY

96.0%

97.0%

95.0%

1.46

1.50

1.27

$652

$711

$322

Indiana

Fort Wayne, IN

92.0%

94.5%

92.0%

1.05

1.12

1.05

$231

$209

$119

Indiana

IndianapolisCarmel, IN

92.6%

93.2%

94.0%

0.77

0.99

1.17

$(482)

$49

$284

Indiana

Muncie, IN

94.0%

96.8%

97.9%

0.73

1.56

0.89

$(73)

$497

$72

A CohnReznick Report | 89

Median Physical Occupancy State Indiana

MSA South BendMishawaka, IN-MI

2008 91.5%

2009 93.0%

2010 93.5%

Median Debt Coverage Ratio 2008 0.81

2009 0.83

Median Per Unit Cash Flow

2010

2008

1.17

$(422)

2009

2010

$(185)

$159

Iowa

Cedar Rapids, IA

95.4%

90.9%

93.5%

1.33

1.03

1.06

$436

$48

$(31)

Iowa

DavenportMoline-Rock Island, IA-IL

96.4%

94.6%

95.8%

1.09

1.07

1.12

$204

$174

$134

Iowa

Des Moines-West Des Moines, IA

94.5%

95.5%

96.0%

1.14

1.09

1.14

$316

$423

$416

Iowa

Iowa City, IA

85.3%

88.8%

91.1%

0.69

1.08

1.40

$(535)

$77

$609

Iowa

Sioux City, IANE-SD

92.1%

95.3%

95.8%

1.11

1.17

1.22

$213

$331

$692

Iowa

Waterloo-Cedar Falls, IA

92.1%

97.6%

97.7%

1.28

1.27

1.55

$(241)

$95

$312

Kansas

Coffeyville, KS

95.0%

96.9%

96.8%

1.28

1.31

2.06

$286

$282

$897

Kansas

Hutchinson, KS

99.2%

98.6%

98.8%

1.23

1.09

1.08

$281

$124

$137

Kansas

Manhattan, KS

98.0%

96.0%

97.0%

1.30

1.10

$419

$571

$109

Kansas

Topeka, KS

93.8%

91.0%

92.9%

0.55

1.15

0.97

$(11)

$42

$(46)

Kansas

Wichita, KS

95.9%

96.1%

95.5%

1.27

1.14

1.07

$345

$382

$148

Kentucky

LexingtonFayette, KY

97.7%

96.8%

97.0%

0.96

1.27

1.40

$(306)

$639

$491

Kentucky

Louisville/Jefferson County, KY-IN

95.1%

95.5%

95.4%

1.17

1.19

1.27

$250

$261

$436

Louisiana

Alexandria, LA

97.3%

95.5%

96.8%

1.83

1.35

1.19

$1,284

$690

$792

Louisiana

Baton Rouge, LA

95.0%

94.8%

95.0%

1.14

1.23

1.14

$303

$344

Louisiana

Hammond, LA

98.8%

98.1%

98.7%

1.71

1.36

1.44

$434

$342

$428

Louisiana

Houma-Bayou Cane-Thibodaux, LA

97.0%

98.1%

98.9%

1.09

1.20

1.41

$(227)

$190

$385

Louisiana

Lafayette, LA

97.5%

95.0%

98.5%

1.48

1.22

1.36

$733

$330

$681

Louisiana

Lake Charles, LA

98.0%

96.1%

1.27

1.23

1.30

$719

$750

$702

Louisiana

Monroe, LA

96.0%

98.0%

1.21

1.33

$399

$464

$526

Louisiana

New OrleansMetairie-Kenner, LA

99.1%

96.0%

95.9%

1.20

1.14

1.07

$(114)

$(25)

$198

Louisiana

OpelousasEunice, LA

95.0%

94.6%

95.9%

1.42

1.56

1.37

$397

$813

$592

Louisiana

Ruston, LA

95.2%

95.8%

96.6%

1.36

1.46

$393

$710

$1,067

Louisiana

ShreveportBossier City, LA

95.3%

96.9%

97.0%

1.35

1.36

1.37

$767

$702

$976

Maine

Bangor, ME

98.0%

98.5%

97.8%

1.18

1.63

1.83

$290

$102

$490

Maine

Portland-South PortlandBiddeford, ME

97.6%

97.4%

97.0%

1.28

1.34

1.32

$800

$604

$750

Maryland

BaltimoreTowson, MD

96.4%

97.0%

1.20

1.17

1.28

$211

$352

$431

Maryland

Cambridge, MD

97.5%

96.9%

97.6%

1.14

1.79

1.36

$1,076

$580

$508

Maryland

HagerstownMartinsburg, MD-WV

97.1%

97.5%

98.6%

1.33

1.23

1.50

$372

$325

$981

Maryland

Salisbury, MD

97.0%

96.5%

96.0%

1.12

1.18

1.12

$161

$253

$181

Massachusetts

Barnstable Town, MA

98.0%

98.5%

99.0%

1.42

1.49

1.60

$800

$538

$506

90 | The Low-Income Housing Tax Credit Program

Median Physical Occupancy State

MSA

2008

2009

2010

Median Debt Coverage Ratio 2008

2009

Median Per Unit Cash Flow

2010

2008

2009

2010

Massachusetts

BostonCambridgeQuincy, MA-NH

96.5%

97.0%

97.2%

1.15

1.21

1.25

$378

$639

$723

Massachusetts

Pittsfield, MA

96.3%

96.7%

97.0%

1.29

1.15

0.80

$61

$252

$(153)

Massachusetts

Springfield, MA

96.5%

95.8%

96.9%

1.05

1.13

1.27

$61

$232

$231

Massachusetts

Worcester, MA

97.6%

97.3%

98.5%

1.24

1.18

1.19

$763

$566

$591

Michigan

Allegan, MI

93.5%

92.5%

95.0%

0.93

1.06

0.97

$(25)

$256

$(36)

Michigan

Ann Arbor, MI

97.0%

96.9%

96.0%

1.22

1.26

1.15

$368

$285

$518

Michigan

Detroit-WarrenLivonia, MI

94.0%

94.1%

95.0%

0.85

1.00

0.99

$(275)

$(54)

$(14)

Michigan

Flint, MI

92.0%

96.0%

95.8%

1.01

1.12

1.30

$144

$392

$546

Michigan

Grand RapidsWyoming, MI

96.0%

97.0%

1.02

1.08

1.09

$46

$220

$232

Michigan

Holland-Grand Haven, MI

87.0%

90.0%

92.0%

0.62

1.16

1.06

$(771)

$(778)

$248

Michigan

KalamazooPortage, MI

92.0%

94.9%

95.0%

0.76

1.01

1.07

$(403)

$49

$149

Michigan

Lansing-East Lansing, MI

92.0%

93.0%

0.85

0.99

1.07

$(251)

$15

$191 $(343)

Michigan

Midland, MI

96.0%

97.3%

96.0%

0.94

0.83

0.80

$(75)

$(198)

Michigan

Mount Pleasant, MI

96.0%

95.0%

96.5%

1.35

1.37

1.27

$738

$632

$634

Michigan

Muskegon-Norton Shores, MI

88.7%

93.0%

94.0%

1.01

0.98

1.04

$(58)

$(176)

$137

Michigan

Niles-Benton Harbor, MI

95.5%

95.0%

96.0%

1.21

0.97

1.02

$525

$50

$(31)

Michigan

Owosso, MI

91.5%

92.5%

97.0%

1.13

1.06

1.20

$377

$186

$477

Michigan

Saginaw-Saginaw Township North, MI

91.0%

92.0%

94.0%

1.01

1.07

1.18

$27

$233

$397 $822

Michigan

Traverse City, MI

92.0%

90.0%

89.0%

0.99

1.08

1.13

$143

$341

Minnesota

Bemidji, MN

98.6%

98.1%

99.5%

1.43

1.37

1.22

$201

$321

$120

Minnesota

Brainerd, MN

96.0%

98.0%

99.0%

1.64

1.50

1.63

$1,192

$701

$618

Minnesota

Duluth, MN-WI

96.0%

94.6%

94.8%

1.23

1.38

1.59

$648

$675

$1,110

Minnesota

Minneapolis-St. Paul-Bloomington, MN-WI

96.6%

96.4%

97.1%

1.15

1.19

1.25

$493

$695

$713

Minnesota

Rochester, MN

96.3%

97.3%

96.9%

1.65

1.52

$1,177

$989

$588

Minnesota

St. Cloud, MN

97.4%

96.9%

97.8%

1.30

1.24

$617

$691

$653

Mississippi

Gulfport-Biloxi, MS

96.0%

89.0%

90.0%

1.15

1.37

1.24

$410

$848

$336

Mississippi

Hattiesburg, MS

96.5%

96.3%

98.6%

1.40

1.31

1.63

$378

$318

$374

Mississippi

Jackson, MS

93.6%

94.0%

97.0%

1.07

1.11

1.28

$155

$56

$601

Missouri

Branson, MO

92.9%

91.3%

91.0%

1.16

0.99

1.00

$286

$(10)

$10

Missouri

Jefferson City, MO

87.1%

90.8%

92.9%

0.86

0.72

0.88

$(56)

$(227)

$(156) $528

Missouri

Joplin, MO

96.6%

95.8%

96.3%

1.32

1.41

1.48

$260

$395

Missouri

Kansas City, MO-KS

95.6%

95.7%

96.4%

1.06

1.09

1.14

$128

$187

$290

Missouri

Springfield, MO

95.3%

96.6%

95.7%

1.46

1.53

1.50

$345

$771

$579

Missouri

St. Louis, MO-IL

95.0%

94.0%

95.0%

1.10

1.12

1.19

$154

$315

$302

Montana

Bozeman, MT

94.8%

86.8%

96.1%

1.09

0.78

1.04

$349

$(563)

$161

Montana

Kalispell, MT

98.0%

97.5%

99.0%

1.41

1.43

1.33

$524

$472

$392

A CohnReznick Report | 91

Median Physical Occupancy State

MSA

2008

2009

2010

Median Debt Coverage Ratio 2008

2009

Median Per Unit Cash Flow

2010

2008

2009

2010

Montana

Missoula, MT

92.6%

94.4%

96.4%

1.21

1.56

2.08

$870

$1,357

$1,061

Nebraska

Grand Island, NE

95.9%

96.1%

96.6%

1.14

1.01

1.33

$246

$93

$507

Nebraska

Lincoln, NE

97.6%

97.9%

97.6%

1.20

1.44

1.39

$632

$892

$870

Nebraska

Omaha-Council Bluffs, NE-IA

94.1%

94.4%

95.2%

1.12

1.13

1.24

$158

$230

$292

Nevada

Las VegasParadise, NV

97.8%

97.0%

95.8%

1.44

1.26

1.31

$433

$487

$639

Nevada

Reno-Sparks, NV

97.1%

96.0%

95.2%

1.13

1.20

1.05

$112

$213

$120

New Hampshire

Concord, NH

96.3%

96.0%

99.6%

1.01

1.43

2.04

$37

$980

$1,656

New Hampshire

Keene, NH

97.6%

95.8%

96.0%

0.64

1.59

1.79

$(117)

$546

$466

New Hampshire

Lebanon, NH-VT

97.6%

97.5%

96.7%

1.04

1.28

1.24

$68

$535

$522

New Hampshire

ManchesterNashua, NH

97.0%

97.2%

96.3%

1.26

1.15

1.38

$463

$412

$609

New Jersey

Trenton-Ewing, NJ

98.0%

95.4%

96.2%

1.44

1.03

1.09

$43

$24

$130

New Mexico

Albuquerque, NM

94.5%

96.7%

96.2%

1.27

1.22

1.32

$525

$566

$713 $557

New Mexico

Las Cruces, NM

96.6%

96.0%

97.5%

1.17

1.21

1.39

$366

$426

New Mexico

Santa Fe, NM

94.3%

92.5%

95.9%

1.22

1.14

1.16

$809

$771

$860

New York

AlbanySchenectadyTroy, NY

98.4%

98.0%

97.9%

1.51

1.53

1.57

$922

$808

$990

New York

Binghamton, NY

97.9%

97.0%

1.08

1.58

1.34

$359

$208

$277

New York

Buffalo-Niagara Falls, NY

98.0%

97.3%

97.9%

1.10

1.22

1.16

$114

$357

$516

New York

Glens Falls, NY

98.0%

97.9%

98.2%

1.50

1.45

1.77

$613

$400

$732

New York

JamestownDunkirk-Fredonia, NY

95.6%

95.9%

95.0%

1.89

1.26

$(149)

$276

$(9)

New York

Kingston, NY

99.2%

99.0%

1.24

1.58

1.80

$426

$670

$674

New York

New York-Northern New Jersey-Long Island, NY-NJ-PA

97.4%

97.7%

97.8%

1.21

1.44

1.46

$251

$557

$675

New York

PoughkeepsieNewburghMiddletown, NY

97.1%

97.5%

98.1%

1.04

1.28

1.57

$20

$480

$797

$679

New York

Rochester, NY

96.0%

96.7%

96.4%

1.28

1.30

1.49

$329

$408

New York

Syracuse, NY

98.2%

96.5%

97.7%

1.61

1.68

1.76

$436

$619

New York

Utica-Rome, NY

96.1%

97.0%

1.91

1.54

1.71

$517

$609

$539

North Carolina

Asheville, NC

98.0%

98.1%

98.4%

1.15

1.35

1.50

$158

$357

$486

North Carolina

CharlotteGastonia-Rock Hill, NC-SC

96.2%

97.1%

97.0%

1.28

1.53

1.34

$488

$779

$598

North Carolina

DurhamChapel Hill, NC

96.5%

96.1%

97.9%

0.97

0.91

0.97

$(50)

$(243)

$(208)

North Carolina

Fayetteville, NC

95.2%

96.4%

96.9%

1.59

1.86

1.49

$669

$786

$531

North Carolina

GreensboroHigh Point, NC

95.5%

93.5%

97.0%

1.18

1.15

$93

$169

$210

North Carolina

Greenville, NC

91.7%

95.3%

95.0%

1.53

1.92

1.84

$374

$879

$883

92 | The Low-Income Housing Tax Credit Program

Median Physical Occupancy State

MSA

2008

2009

2010

Median Debt Coverage Ratio 2008

2009

Median Per Unit Cash Flow

2010

2008

2009

2010

North Carolina

Hickory-LenoirMorganton, NC

93.0%

93.3%

96.0%

1.19

1.13

1.35

$274

$180

$481

North Carolina

Jacksonville, NC

96.5%

96.2%

98.5%

1.04

1.45

1.24

$(10)

$527

$547

North Carolina

Lumberton, NC

96.9%

99.0%

97.5%

1.33

1.79

1.64

$131

$883

$684

North Carolina

Raleigh-Cary, NC

96.0%

97.0%

1.12

1.21

1.27

$223

$290

$501

North Carolina

Roanoke Rapids, NC

94.6%

99.8%

97.8%

1.15

1.65

1.51

$177

$862

$478

North Carolina

Rocky Mount, NC

99.0%

99.1%

1.73

1.89

2.24

$659

$996

$1,377

North Carolina

Shelby, NC

96.2%

96.8%

97.6%

0.91

1.01

0.96

$(223)

$29

$100

North Carolina

Wilmington, NC

98.1%

96.0%

95.5%

1.37

1.27

1.28

$422

$342

$535

North Carolina

Winston-Salem, NC

97.3%

96.7%

95.8%

1.10

1.11

1.05

$124

$197

$88

North Dakota

Bismarck, ND

94.5%

97.5%

98.3%

1.03

1.27

1.33

$87

$605

$728

North Dakota

Fargo, ND-MN

97.4%

98.3%

96.4%

1.12

1.08

1.14

$291

$252

$289

Ohio

Akron, OH

97.3%

97.6%

0.97

0.88

1.33

$(62)

$(208)

$342

Ohio

CantonMassillon, OH

96.5%

96.3%

97.8%

0.96

0.91

1.18

$(152)

$(161)

$340

Ohio

CincinnatiMiddletown, OH-KY-IN

94.6%

94.8%

94.0%

0.97

1.03

1.06

$(262)

$39

$42

Ohio

Cleveland-ElyriaMentor, OH

97.2%

97.3%

96.7%

1.00

1.12

1.13

$(85)

$116

$258

Ohio

Columbus, OH

96.9%

96.5%

97.2%

0.90

1.02

1.20

$(76)

$57

$389

Ohio

Dayton, OH

95.8%

95.0%

95.8%

1.10

0.94

1.09

$108

$91

$223

Ohio

Mansfield, OH

97.0%

96.0%

94.6%

1.06

1.19

1.05

$68

$377

$103

Ohio

Springfield, OH

95.4%

95.0%

95.8%

0.90

1.05

0.96

$(139)

$52

$(72)

Ohio

SteubenvilleWeirton, OH-WV

96.7%

95.8%

96.7%

0.95

0.97

1.12

$(94)

$(43)

$231

Ohio

Toledo, OH

98.9%

93.8%

95.4%

1.15

1.00

1.29

$290

$(51)

$387

Ohio

YoungstownWarrenBoardman, OH-PA

95.2%

94.8%

94.9%

0.85

0.86

1.05

$44

$(257)

$104

Oklahoma

Oklahoma City, OK

95.8%

95.5%

96.0%

1.30

1.29

1.25

$368

$465

$372

Oklahoma

Stillwater, OK

95.7%

92.0%

94.3%

1.40

1.08

1.34

$478

$273

$554

Oklahoma

Tulsa, OK

96.0%

96.3%

1.26

1.14

1.27

$269

$208

$268

Oregon

Bend, OR

95.7%

87.4%

95.8%

1.46

1.11

1.03

$948

$475

$149

Oregon

EugeneSpringfield, OR

97.9%

97.4%

98.7%

1.34

1.30

1.22

$349

$310

$335

Oregon

Medford, OR

97.1%

96.9%

97.2%

1.34

1.31

1.23

$986

$639

$550

Oregon

PendletonHermiston, OR

95.8%

94.0%

96.4%

1.09

1.27

1.26

$480

$212

$305

Oregon

PortlandVancouverHillsboro, OR-WA

96.7%

95.8%

96.0%

1.14

1.11

1.20

$342

$287

$497

A CohnReznick Report | 93

Median Physical Occupancy State

MSA

2008

2009

2010

Median Debt Coverage Ratio 2008

2009

Median Per Unit Cash Flow

2010

2008

2009

2010

Oregon

Salem, OR

97.4%

96.8%

96.2%

1.08

1.33

1.39

$76

$516

$452

Pennsylvania

AllentownBethlehemEaston, PA-NJ

96.0%

96.7%

97.6%

1.21

1.49

1.46

$647

$331

$275

Pennsylvania

HarrisburgCarlisle, PA

95.5%

92.4%

95.9%

1.30

1.36

1.44

$476

$497

$331

Pennsylvania

PhiladelphiaCamdenWilmington, PA-NJ-DE-MD

96.7%

97.0%

96.7%

1.09

1.12

1.16

$80

$120

$211

Pennsylvania

Pittsburgh, PA

97.0%

96.5%

97.0%

1.10

1.20

1.36

$(10)

$293

$208

Pennsylvania

Scranton— Wilkes-Barre, PA

99.0%

98.0%

1.43

1.72

1.30

$(180)

$(162)

$438

Pennsylvania

York-Hanover, PA

96.3%

98.4%

97.8%

1.70

1.30

1.49

$358

$395

$546

Puerto Rico

San Juan-CaguasGuaynabo, PR

99.9%

100.0%

1.09

1.22

1.23

$340

$388

$495

Rhode Island

Providence-New Bedford-Fall River, RI-MA

97.0%

96.8%

1.26

1.30

$266

$433

$371

South Carolina

Anderson, SC

96.2%

92.2%

94.6%

1.01

1.12

1.28

$23

$145

$463

South Carolina

Charleston-North CharlestonSummerville, SC

96.4%

97.1%

95.9%

0.98

1.06

1.19

$132

$301

South Carolina

Columbia, SC

96.7%

97.5%

97.0%

1.20

1.19

1.26

$530

$376

$379

South Carolina

Florence, SC

96.8%

96.3%

98.7%

1.40

1.31

1.51

$399

$260

$705

South Carolina

GreenvilleMauldin-Easley, SC

95.8%

95.0%

95.5%

1.12

1.16

1.20

$353

$329

$567

South Carolina

Hilton Head Island-Beaufort, SC

95.1%

95.4%

94.4%

1.16

1.14

$163

$372

$331

South Carolina

Myrtle Beach-North Myrtle BeachConway, SC

93.8%

96.2%

95.6%

1.29

1.17

1.31

$344

$201

$492

South Carolina

Spartanburg, SC

96.0%

96.5%

96.0%

0.87

1.05

0.87

$(411)

$(190)

$(165)

South Dakota

Rapid City, SD

95.3%

95.8%

96.4%

1.17

1.46

1.36

$355

$956

$727

South Dakota

Sioux Falls, SD

93.9%

92.1%

92.0%

1.18

1.20

1.06

$568

$447

$201

Tennessee

Chattanooga, TN-GA

97.5%

98.0%

97.4%

1.53

1.00

1.13

$179

$137

$149

Tennessee

Knoxville, TN

94.6%

94.0%

93.9%

0.92

1.08

1.14

$(436)

$171

$115

Tennessee

Memphis, TN-MS-AR

95.0%

93.9%

95.0%

1.10

1.15

1.22

$33

$324

$281

Tennessee

NashvilleDavidson— Murfreesboro— Franklin, TN

97.0%

93.1%

95.9%

1.15

1.09

1.13

$276

$257

$418

Texas

Austin-Round Rock-San Marcos, TX

94.9%

92.0%

94.6%

1.01

0.93

$45

$184

$(78)

Texas

Beaumont-Port Arthur, TX

96.6%

96.5%

96.6%

1.02

1.18

1.38

$56

$312

$466

94 | The Low-Income Housing Tax Credit Program

Median Physical Occupancy State

MSA

2008

2009

2010

Median Debt Coverage Ratio 2008

2009

Median Per Unit Cash Flow

2010

2008

2009

2010

Texas

BrownsvilleHarlingen, TX

96.4%

97.2%

96.4%

1.23

1.57

1.32

$599

$1,167

$620

Texas

College StationBryan, TX

93.7%

94.6%

94.8%

1.46

1.24

1.25

$510

$139

$480

Texas

Corpus Christi, TX

96.9%

97.0%

96.9%

1.24

1.12

1.23

$622

$258

$426

Texas

Dallas-Fort WorthArlington, TX

94.3%

94.0%

1.07

1.06

1.11

$158

$198

$266

Texas

El Paso, TX

97.0%

97.6%

97.9%

1.30

1.42

1.66

$489

$708

$993

Texas

Houston-Sugar Land-Baytown, TX

95.4%

94.8%

94.9%

1.02

1.09

1.16

$97

$262

$327

Texas

Lubbock, TX

90.0%

91.8%

94.5%

1.13

1.24

1.02

$(66)

$229

$73

Texas

McAllen-EdinburgMission, TX

96.6%

97.9%

98.3%

1.32

1.35

1.24

$708

$647

$642

Texas

San AntonioNew Braunfels, TX

96.2%

95.0%

96.0%

1.02

1.09

1.16

$92

$461

$609

Utah

OgdenClearfield, UT

96.0%

97.0%

97.5%

1.21

1.26

1.55

$510

$712

$1,512

Utah

Salt Lake City, UT

Utah

St. George, UT

97.8%

96.0%

96.3%

1.22

1.20

1.13

$521

$535

$330

100.0%

1.23

1.83

1.90

$689

$1,215

$1,261

Vermont

Barre, VT

96.9%

96.7%

94.8%

0.88

1.28

1.26

$(249)

$444

$307

Vermont

Burlington-South Burlington, VT

97.5%

97.9%

98.0%

1.21

1.25

1.32

$404

$666

$800

Virginia

Danville, VA

96.5%

95.1%

96.0%

1.04

0.99

1.11

$175

$469

$216

Virginia

Richmond, VA

94.1%

93.7%

95.0%

1.11

1.07

1.12

$153

$271

$336

Virginia

Roanoke, VA

96.8%

97.6%

1.20

1.03

$71

$171

$43

Virginia

Virginia BeachNorfolk-Newport News, VA-NC

97.5%

97.0%

97.3%

1.25

1.29

1.34

$491

$584

$745

Washington

Bellingham, WA

97.0%

97.3%

98.5%

1.14

1.30

1.47

$282

$256

$633

Washington

BremertonSilverdale, WA

95.6%

95.0%

96.4%

1.36

1.35

1.29

$667

$770

$641

Washington

Moses Lake, WA

97.0%

96.4%

95.9%

1.22

1.64

1.55

$480

$689

$669

Washington

Mount VernonAnacortes, WA

96.3%

96.1%

98.2%

1.07

0.98

0.91

$118

$268

$(139)

Washington

Seattle-TacomaBellevue, WA

97.1%

96.4%

97.0%

1.24

1.23

1.19

$441

$474

$395

Washington

Spokane, WA

97.0%

99.2%

95.0%

1.42

1.26

1.48

$415

$372

$535

Washington

Yakima, WA

95.0%

95.5%

97.0%

1.23

1.24

1.23

$272

$241

$402

West Virginia

Charleston, WV

98.3%

97.2%

96.5%

1.09

0.98

1.03

$98

$119

West Virginia

Huntington-Ashland, WV-KY-OH

95.0%

97.0%

98.4%

1.14

1.01

1.30

$161

$313

$355

West Virginia

ParkersburgMarietta-Vienna, WV-OH

98.5%

94.4%

97.0%

1.16

1.26

1.40

$139

$397

$511

West Virginia

Wheeling, WV-OH

96.2%

97.4%

98.9%

1.14

1.17

1.15

$241

$314

$389

Wisconsin

Janesville, WI

96.0%

96.5%

95.1%

1.10

1.09

1.04

$219

$233

$108

Wisconsin

Madison, WI

95.8%

97.7%

96.3%

1.13

1.15

1.20

$603

$577

$950

Wisconsin

MilwaukeeWaukesha-West Allis, WI

96.0%

94.6%

95.4%

0.98

1.08

1.10

$39

$218

$201

Wisconsin

Racine, WI

92.6%

90.6%

96.0%

1.15

1.35

1.27

$296

$1,039

$839

A CohnReznick Report | 95

About Us About the Tax Credit Investment Services Group The Tax Credit Investment Services (TCIS) group is a dedicated business unit within CohnReznick focused on evaluating and advising clients on tax-advantaged investments, including low-income housing, historic rehabilitation, new markets and renewable energy. As a group made up of experts with a fairly narrow industry focus, TCIS covers a variety of consulting areas, including investment due diligence, investment and business strategy, and industry benchmarking research for the benefit of investor and syndicator communities. The TCIS team is composed of a multidisciplinary group of professionals, including CPAs, attorneys, financial analysts and other professionals with experience as state housing finance agency and commercial real estate executives. CohnReznick’s TCIS team members have authored a number of affordable housing industry studies, speak regularly at industry conferences and have been widely quoted in the financial press concerning tax credit investments. In addition to the professional experience of TCIS team members, the group’s clients benefit from the knowledge and experience of hundreds of CohnReznick audit, tax and consulting professionals working on investment tax credit transactions on a daily basis. For more information about TCIS, please visit www.cohnreznick.com/tcis. To contact TCIS, please call 1.617.648.1400 or write to: CohnReznick – TCIS One Boston Place, Suite 500 Boston, MA 02108

About CohnReznick With origins dating back to 1919, CohnReznick LLP is currently the 11th largest accounting, tax and advisory firm in the United States, combining the resources and technical expertise of a national firm with the hands-on, entrepreneurial approach that today’s dynamic business environment demands. The firm was formed out of the combination of J.H. Cohn and Reznick Group in October 2012. CohnReznick serves a large number of diverse industries and offers specialized services to Fortune 1000 companies, owner-managed firms, international enterprises, government agencies, not-for-profit organizations, and other key market sectors. Headquartered in New York, NY, CohnReznick serves its clients with more than 280 partners, 2,000 employees and 25 offices nationwide. The firm is a member of Nexia International, a global network of independent accountancy, tax, and business advisors. For more information, visit www.cohnreznick.com.

96 | The Low-Income Housing Tax Credit Program

One Boston Place, Suite 500 Boston, MA 02108 617.648.1400 www.cohnreznick.com