Reducing the Complexity in California’s Affordable Housing Finance System

Authors:
Carolina Reid, Faculty Research Advisor
Tiffany Tran, MPP ’25, Graduate Student Researcher

Given the critical need to expand the supply of affordable housing in California, policymakers are increasingly looking for ways to reduce the costs of development. One way to do this is to reduce the complexity of the affordable housing finance system. Compiling sufficient public funding to build affordable housing is particularly onerous in California, requiring that developers navigate multiple agencies and departments at both the state and local level. This fragmented funding landscape can increase project complexity and extend the time it takes to build affordable housing. As we detail below, we find that across comparable projects, the inclusion of one additional public funding source adds, on average, four months to the timeline of being able to start construction, and is associated with an increase of approximately $20,460 in per-unit total development costs.^[1]

Reducing this complexity should be a policy priority, and there is now movement in that direction. Governor Newsom has put forward a proposal to consolidate housing planning and financing functions within a new housing and homelessness agency. In brief, the proposal reorganizes the Business, Consumer Services and Housing Agency (BCSH) into two agencies. One focuses on consumer protection and business regulation, and one—the California Housing and Homelessness Agency (CHHA)—consolidates most housing, homelessness, and civil rights functions within the Governor’s administration. CHHA’s mandate would be to “integrate housing programs, streamline policies, and simplify the administration of our state affordable housing programs.”

To help inform conversations on this proposal, we analyzed how new affordable housing in California is financed in order to better understand the current complexity and fragmentation, as well as their costs. The most important source of funding is the Low-Income Housing Tax Credit (LIHTC) program, which is included in almost all new affordable housing projects across the state. However, the subsidy that the LIHTC program provides is often insufficient on its own, especially for higher-cost regions of the state or when serving households with very low or extremely low incomes. Developers therefore often stack LIHTC with other funding sources to make their projects financially feasible. These sources include bank loans, as well as public funding sources such as local, regional, state, or federal grants and loans.

In this commentary, we present an analysis of LIHTC application data for new construction projects awarded tax credits in California between 2020 and 2023. We examine the number of different public funding sources needed to make projects feasible, and assess how this complexity increases the time it takes to get a new project off the ground. We also provide estimates for how this complexity is associated with higher housing development costs. A more detailed description of the methodology underlying the analysis is available in the Appendix below.

Our main findings include:

Between 2020 and 2023, 92 percent of new construction LIHTC projects relied on at least one public funding source, in addition to the tax credits, and 76 percent included two or more different public funding sources. Approximately a third of projects included at least two different state funding sources. Projects also often blend multiple locally administered funding sources, including funding from local bond measures, Community Development Block Grant, or HOME dollars.^[2]

Developments serving people with special needs, including supportive housing for people experiencing homelessness, tend to layer in more public funding sources than senior or family buildings, given the deeper subsidies needed to bring down rents to levels affordable for those populations (Figure 1). On average, special needs projects incorporate more than three other public funding sources, in addition to the LIHTC subsidy.

Figure 1: Average Number of Public Funding Sources by Project Type, LIHTC New Construction Awards, 2020 – 2023Source: Analysis of LIHTC project application data. Shazia Manji, Pratish Patel, Carolina Reid, and Quinn Underriner (2023).“California Low-Income Housing Tax Credit Database, 2020 – 2023,” Cal Poly, San Luis Obispo, and Terner Center for Housing Innovation, Berkeley, California.

Compiling additional public funding can add months or years to the development timeline. In general, new affordable housing developers must secure their local and state funding sources before they can apply for LIHTC. Developments that only need tax credits wait three months between their LIHTC application date and award date.^[3] However, when a project incorporates an additional public funding source, it will take, on average, over 10 months between the earliest recorded date associated with seeking that funding and their LIHTC awarded date. Projects with three to five additional public funding sources take nearly two years (on average) between the first funding application and their LIHTC award.

Longer timelines are the result of not only the need to apply for and combine different state funding sources, but also the gap between receiving a state funding award and applying for tax credits. Developers have to wait an average of 12 months between being awarded their last state funding source and their LIHTC award. In addition, some delays are due to limited tax credit availability: we find that approximately 27 percent of the new construction developments that were eventually awarded tax credits between 2020 and 2023 had to reapply at least once.^[4] We also find that the need to secure multiple sources of locally controlled funding adds significantly to the time before a developer can submit their tax credit application.

Figure 2: Average Number of Months between Earliest Recorded Public Funding Application Date and LIHTC Award, New Construction Awards, 2020 – 2023

Source: California Low-Income Housing Tax Credit Database, 2020 – 2023.

Of course, the number of public funding sources needed could just be a function of the cost or complexity of the project. For example, an infill, multi-story building dedicated to people experiencing homelessness will be more expensive to build and operate, and therefore will require more funding to be financially feasible. To account for these differences, we ran a series of regressions to help control for observable differences across LIHTC developments, including project type, region, infill location, number of units and stories, prevailing wage requirements, and structured parking (See Appendix Table A for full regression results).

We find that even after holding constant a variety of project characteristics, the process of applying for multiple different sources of public funding is still associated with a longer timeframe. We find that across comparable projects, the inclusion of one additional public funding source, on average, adds four months to the timeline of identifying, applying for, and waiting for an award.

This funding complexity—specifically, the time it takes to apply for and secure enough public funding from both local and state sources—is associated with higher development costs. Each additional public funding source is associated with $20,460 higher per-unit total development costs, on average.

Interviews with developers conducted as part of prior Terner Center research reveals why having to compile multiple sources of funding can increase costs. First, these higher costs are the result of direct mechanisms—such as higher staffing, legal, and compliance costs associated with managing multiple applications and contracts. Second, they also result from indirect costs associated with longer development timelines—such as higher predevelopment costs (including holding costs for private financing used during predevelopment), inflation and interest rate increases, and rising materials and/or labor costs.

Conclusion

This analysis reveals that fragmentation in California’s affordable housing financing system creates significant inefficiencies. The need to compile multiple sources of public funding—from different government agencies with different timelines, regulations, and reporting requirements—adds both time and costs. When we account for differences across LIHTC projects, we estimate that each additional public funding source adds an average of four months to the development timeline and $20,460 to per-unit total development costs.

While funding fragmentation is not the only, or even the largest, driver of the state’s high housing development costs, it is the one squarely within the ability of the public sector to address. The Governor’s recent proposal is a step towards reducing that fragmentation. But our analysis also suggests it doesn’t go far enough. The agencies that award tax credits (the California Tax Credit Allocation Committee (TCAC) and California Debt Limit Allocation Committee (CDLAC)) remain outside of the proposed Housing Development and Finance Committee, even though there is often a significant gap (an average of 12 months) between state funding and tax credit awards. Bringing LIHTC into the new structure would help to ensure that all of the state’s affordable housing resources and policy priorities are aligned. Multiple other states—including Illinois, Pennsylvania, and Minnesota—have this type of more consolidated structure, meaning that developers can apply for tax credits and other programs at the same time.

In addition to focusing on state funding governance, the new agency should work to minimize and/or better align state and local funding use (including city, county, and regional sources, as well as federal funding from local public housing authorities) in the context of specific projects.  For example, the agency should work to reduce the frequency with which developers must match local funds with non-LIHTC state funds. Existing scoring preferences in state housing programs currently reward applicants that leverage local funds, which may unnecessarily be contributing to increased fragmentation and cost.

The State could also accelerate efforts to develop standardized documents that would apply across both local and state funding programs. For example, Massachusetts has created a universal application, as well as universal closing documents, which are accepted by local, state, public, and private funders alike. They find that these standardized documents have cut costs, especially for projects with multiple funding sources.

The details of implementation also matter, including ensuring the capacity of the new agency to process awards effectively—such as reviewing, underwriting, and closing funding agreements in a timely manner—as well as to undertake asset management and compliance. Building in ongoing processes to get feedback from affordable housing providers and collecting better data to assess the impact of changes will be important for ensuring positive outcomes.

In conclusion, the analysis presented here underscores the ways in which funding complexity adds to time and costs, creating inefficiencies that work against the goals of containing costs and stretching subsidies further to house more people. Consolidating housing functions under a single state agency would help to address these issues and reduce unnecessary fragmentation and bureaucracy in the affordable housing finance system.

Methodology

This analysis relies on data scraped from 699 applications for new construction projects awarded LIHTC credits between 2020 and 2023. California’s Tax Credit Allocation Committee (TCAC) makes project application data available online. The Terner Center has been creating a database of these applications, extracting information on project characteristics, sources of funding, and cost-line items.^[5]

For this analysis, we focus on the total development costs per unit for LIHTC new construction awards. All dollar amounts have been adjusted for inflation to 2024 dollars. These data reflect the developer’s estimates of project costs at the time of application, not the final costs after the development is completed.

The funding timeline length was calculated in months using the Development Timetable. Within that table, the Other Loans and Grants section lists up to six public funding sources that are part of the project’s capital stack, including the application and award date for each of those sources. The number of months was calculated as the difference between the earliest recorded application date in the Other Loans and Grants section of the table and the date the project was awarded its LIHTC funding. LIHTC application award dates were matched to their application rounds and awarded dates published on the TCAC website. We use these reported dates to estimate how long projects spent compiling their public funding sources. Note that Project-Based Section 8 voucher allocations are often not listed as part of the permanent financing information on the application, meaning that the actual number of sources and coordination requirements are likely to be higher.

Analyzing the different public funding sources included in LIHTC applications is challenging, since developers do not enter funding source names consistently. The applications in the sample contained over 2,500 unique names in the Permanent Financing section of the application, as well as in the Other Loans and Grants section of the Development Timetable. We cleaned these data by reconciling different spellings of similar names (e.g., AHSC, Affordable Housing Sustainable Communities, HCD AHSC grant) or identifying through web searches other programs or sources of funding. We then categorized different funding sources into eight categories: federal funding programs, state funding programs, local funding sources, funding from public housing authorities, private funding, and “other.”

A final caveat is that applications also sometimes include errors or include information inconsistently. To ensure the highest-quality data possible, we fact-checked and entered missing data manually (for example, by cross-referencing TCAC staff reports), yet it is possible that there are still some errors or inconsistencies in the application data underlying this analysis.

​​Table A1: Regression Model Results, New Construction LIHTC Awards, 2020 – 2023

	Number of Months between Earliest Public Funding Source Date and Tax Credit Award		Per Unit Total Development Costs
Intercept	4.67		347,691	***
Number of Public Funding Sources	3.87	***	20,460	***
9 Percent Tax Credit	-0.99		-29,292	**
Number of Units	-0.01		-893	***
Infill Development	3.63	**	30,596	**
Structured Parking	2.14		63,272	***
Number of Stories (Reference: Single Story)
Two to Five Stories	0.22		47,202	**
More than 5 Stories	2.64		60,730	**
Prevailing Wage	4.47	**	108,484	***
Project Type (Reference: Non-Targeted)
Large Family	1.62		154,506	***
Seniors	3.32		29,981
Special Needs	1.05		64,615	***
Region (Reference: Rural)
Capital and Northern Region	-1.17		31,895
Central Coast Region	-7.02		137,512	***
Central Region	-1.65		-33,390
Inland Region	2.44		6,436
Los Angeles County	-3.67		96,360	***
North and East Bay Region	-1.1		185,135	***
Orange County	-7.23		38,892
San Diego	-4.5		38,802
San Francisco County	0.01		386,912	***
South and West Bay Region	-3.14		276,579	***
Year of Application (Reference: 2020)
2021	1.95		11,479
2022	0.98		40,902	***
2023	2.39		56,630	***
N	686		691
R2	0.2996		0.7037

Source: California Low-Income Housing Tax Credit Database, 2020 – 2023.

Notes: The number of months is calculated as the difference between the first recorded public funding application date and the date of the LIHTC award. The number of public funding sources is derived from the Other Loans and Grants section of the Development Timetable.

*** < .001, ** p < .01, * p< .05

Acknowledgments

We would like to thank Crankstart, the Conrad N. Hilton Foundation, and Yardi Systems for supporting the Terner Center’s research focused on increasing the supply of affordable housing in California.  We would like to thank Sarah Karlinsky, Ben Metcalf, and Ryan Finnigan, and several external reviewers for their comments on previous drafts. We also thank Colleen SchwartzCoffey and Quinn Underriner for their assistance in producing the report.

This research does not represent the institutional views of UC Berkeley or of Terner Center’s funders. Funders do not determine research findings or recommendations in Terner Center’s research and policy reports.

Endnotes

^[1] When we focus solely on the costs associated with state funding sources (rather than all public funding sources), the estimated increase in per unit development costs is $16,810. This is in line with previous research using earlier data that found that each additional state funding source added an estimated $15,800 in per unit development costs.

^[2] The number of funding sources presented in this commentary does not include some other critical public funding sources, such as project-based vouchers, nor does it include permanent bank loan financing.

^[3] This is the normal application/award cycle for tax credits.

^[4] For example, in 2023, only 37.5 percent of applications of 9 percent credits received an award. Enterprise Community Partners estimates that in 2024, there were 491 developments in the pipeline awaiting tax credits to start construction.  Projects that are unsuccessful in their first tax credit application wait even longer, an average of 17 months, before they have enough financing to start construction.

^[5] Manji, S. et al. (2023).“California Low-Income Housing Tax Credit Database, 2020 – 2023,” Cal Poly, San Luis Obispo, and Terner Center for Housing Innovation, Berkeley, California.