What's Missing in the Mamdani Housing Plan?
- Paul Francis with Joe Sheppard
- 1 day ago
- 35 min read
PDF Available:
Introduction
On May 29, 2026, New York City Mayor Zohran Mamdani and Deputy Mayor for Housing and Planning Leila Bozorg released a 112-page housing plan, titled Block by Block. The plan offers a wide-ranging strategy for housing policy, including zoning and administrative process changes to expand the siting opportunities for affordable housing, tenant protection, preservation of affordable housing, major renovation projects at New York City Housing Authority (NYCHA), and homelessness reduction efforts. But the core of the plan is the construction of 200,000 new units of affordable housing and the preservation of another 200,000 units of affordable housing over 10 years.
The reaction to the City’s housing plan was mixed. The New York Post and landlord groups focused critically on aspects of Block by Block that seem more ideologically based, such as a contemplated increase in court actions to transfer housing with persistent code violations from private owners to not-for-profit owners. But the plan was well received by housing advocates and other observers. A New York Daily News editorial said the “steps outlined in the plan are a blueprint to work off of.”
What is conspicuous by its absence from this blueprint, however, is an explanation of the plan to finance the construction of 200,000 new units of affordable housing and preserve another 200,000 units of affordable housing over 10 years. The financing question received relatively little scrutiny from the press. Crain’s was the exception, which noted in its first “key takeaway” about the plan: “The mayor is sticking with this housing target and adding a preservation goal even as funding remains a big question.” (Emphasis added)
That “big question” was left largely unanswered during the press conference announcing Block by Block. Mayor Mamdani said that the capital plan included with the FY 27 Executive Budget would provide $22 billion (including NYCHA) over five years (FY 2026-30) to support his housing plan, although only about $10 billion of that amount is dedicated to new construction of traditional affordable housing, “Special Needs” supportive housing, and senior housing.
When a reporter reminded the Mayor of his campaign pledge to borrow $70 billion in the municipal bond market to finance 200,000 new units of affordable housing and asked whether there were other funding sources beyond the $22 billion, the Mayor changed the subject to talk about finding efficiencies in capital projects related to the State’s mandated small class-size law, while Deputy Mayor Bozorg noted that, in addition to seeking ways to lower the cost of construction, the City would be looking at “innovative new financing tools” and cross-subsidization from market-rate housing.
Raising $70 billion in the municipal bond market to provide capital for the construction of 200,000 new affordable housing units was never realistic. This new debt was to be in addition to the $179 billion in municipal debt already projected by the New York City Comptroller to be outstanding by FY 35.[1] Adding anything close to that amount of additional debt would require workarounds from the City’s constitutional debt limit and amendments to its statutory debt limits.
Borrowing $70 billion would also increase total debt service by approximately $5 billion a year and increase debt service as a percentage of the City’s tax revenue from approximately 12% of the City’s tax revenue to approximately 18%, far above the 15% of City tax revenue limit that the City has observed for decades through policy guidance from the Comptroller’s office.[2] Adding that much debt and debt service would raise concerns among the City’s credit ratings and could trigger a downgrade that would increase all borrowing costs.
In fairness to Mayor Mamdani, both New York City and New York State housing plans advanced over the years have been prone to both ambiguity and exaggeration. Previous New York City housing plans lumped together new construction and preservation of affordable housing units. These plans sometimes (as was the case with Mayor Adams’s 500,000-unit goal over 10 years) included all housing units, not just affordable housing units. Gov. Kathy Hochul’s 2023 “$25 billion Housing Plan” over five years is mostly comprised of State operating support for housing programs of all types (including homeless shelters) and includes all federal low-income housing tax credits utilized in the State, including those allocated to New York City and other localities. Only $1 billion in the Hochul plan is dedicated to capital for new multifamily affordable housing.
Even what counts as “affordable housing” is a flexible concept. According to the New York Housing Conference’s (NYHC) 2026 NYC Housing Tracker Report, of the 13,605 units of “affordable housing” produced through new construction in 2025, 8,582 units were limited to those with household income of 80% of AMI ($122,150 for a family of three) or less, but 503 units were “moderate income” units available to those with household incomes between 81% and 120% of AMI ($183,240 for family of three), while 4,488 units were “middle-income” units available to those with household incomes between 121% and 165% of AMI ($251,955 for a family of three).[3] The NYHC Tracker is based on New York City’s Housing Preservation and Development agency’s (HPD) methodology as reported on New York City’s Open Data portal.[4]
New York City needs more housing at all income levels, including middle-income levels up to 165% of AMI, which includes approximately 70% of all New York City residents. However, we suspect that many New Yorkers would be surprised that units of this type are included in the count of “affordable housing,” particularly since housing at these income levels is not part of the typical description of affordable housing used by elected officials.
One reason elected officials get away with talking about their housing plans with so little precision – which may also explain why the lack of a financing strategy for Block by Block has not attracted more scrutiny – is that the press, public and most elected officials lack a nuanced understanding of the economics of affordable housing. Instead, they direct most of their focus to the issue of siting – especially controversial zoning changes. Siting opportunities are a necessary, but not sufficient, condition for the construction of affordable housing. The plan also needs to be financeable.
There may have been multiple motivations for not laying out a financing plan in Block by Block or the wider affordability mix of the plan. Highlighting higher levels of capital spending would force the city to reflect concomitant increases in debt service cost ratios, which could concern bond rating agencies. Moreover, the City is often opaque when it comes to the financing of affordable housing because they do not want developers to see the “best” deals Housing Preservation and Development (HPD) is providing to other developers. However, it is also the case that acknowledging that the financing necessary for realizing the 200,000-new-unit target would far exceed the amount of capital earmarked for new construction in the City’s capital plans might have been seen as a retreat from the Mayor’s campaign promise – an unattractive prospect for a Mayor who is frequently criticized for walking back his campaign promises.
A clear understanding of the City’s financing plan for Block by Block would help identify gaps so that actions to address them could be developed proactively. The assumptions on which the City’s plan was based could also be reviewed by the New York City Comptroller and the Independent Budget Office in the same way those offices currently comment on the City’s Executive Budget and other financial issues affecting the City.
In the absence of such an official financing plan, one of the purposes of this Issue Brief is to analyze the gap between the City’s financing capacity and the financial resources that appear to be necessary to reach the goal of constructing 200,000 new units of affordable housing over the next 10 years.
A second, but equally important purpose of this Issue Brief is to provide an overview of the economics of constructing affordable housing in New York City and a primer on the basis of the assumptions involved regarding the main drivers of cost and financing sources for developing affordable housing.
Our hope is that this Issue Brief will be useful to others trying to understand the economics of affordable housing in New York City. The paper is accessible to policy generalists, yet detailed enough for housing experts to evaluate the validity of our assumptions.
The Economics of Constructing Affordable Housing
Although the intricacies of affordable housing finance are technical and often confusing, the financing of affordable housing comes down to three basic factors: siting, cost of construction, and sources of financing.
Siting is largely a function of zoning, but also dependent on land acquisition costs. The cost of construction depends on the cost of materials and soft costs, but the largest expense category is labor expense, which in turn is heavily influenced by statutorily mandated labor costs, as well as regulatory factors such as building codes. Financing is a function of its own four drivers, which we discuss in more detail below: the level of rental revenue (which is largely dictated by the affordability mix of the project), the amount of debt that can be supported by the free cash flow of the project, the availability of low income housing tax credits or subsidized debt financing, equity supplied by the developer, and capital subsidies from a governmental entity.
In addition to affordable housing units financed with government subsidies, another major source of the supply of affordable housing consists of affordable units constructed and financed by developers of mixed-income projects without capital subsidies but rather through cross-subsidization from market-rate housing. Developers construct these affordable housing units to comply with statutory minimum set-asides for affordable housing, in order to achieve zoning flexibility, in consideration of land contributed by a governmental entity for the mixed-income project, and/or to receive tax abatements under programs such as 485-x in the case of affordable housing or 467-m in the case of office conversions to residential.
Although the overwhelming majority of affordable housing units in New York City receiving government subsidies received them from New York City, New York State’s housing entities – Homes and Community Renewal (HCR) and Housing Finance Agency (HFA) – also finance some units, principally in the form of supportive housing or housing for the frail elderly. The State’s various supportive housing plans contemplate approximately 1,000 new supportive housing units being developed in New York City on an annual basis, although not all of these units will necessarily involve new construction. Because these units do not require financial support from New York City or reduce the amount of LIHTCs available to the City, our analysis treats these State-financed units as functionally equivalent to affordable housing supplied by private developers.
In the absence of an official financing plan from the City for the construction of 200,000 units of new affordable housing, we have put together a pro forma model that includes the main drivers of the costs and financing of City-financed new affordable housing units to identify the main variables in the sources and uses of funds. The Base Case of this pro forma model includes assumptions that determine the cost of new affordable housing units and the availability and terms of financing sources.
Building on per-unit assumptions, the Base Case model shows how they apply in four scenarios: (i) a hypothetical single, 100% affordable project with 200 units and 4% LIHTCs, (ii) a hypothetical single, 100% affordable project with 119 units and 9% LIHTCs, (iii) 14,000 units of new affordable housing, which is the number of units the Block by Block plan indicates the City expects to be developed in FY 27, and (iv) 20,000 units of new affordable housing, intended to represent the average annual number of units developed over 10 years to reach the 200,000-unit goal.
In all four scenarios, we are only modeling the units that receive capital subsidies from HPD – which we call “City-financed” housing – and have the affordability mix described by the Block by Block plan for subsidized units. This affordability mix results in a maximum household income for these units of 80% of AMI and a blended average affordability standard of 59% of AMI – just below the 60% maximum weighted average AMI to be eligible for maximum use of LIHTCs.
The Base Case model makes some simplifying assumptions. Our Base Case model unit assumptions are based on estimated FY 26 costs and FY 26 terms and availability of various financing sources. The model is highly sensitive to factors that are particularly difficult to forecast, especially the number of units that will be developed without City capital subsidies, the percentage of projects in which land will be contributed versus purchased, interest rates, and the mix of labor costs in the projects constructed.
We estimate that the total cost of development of an affordable housing unit built on land contributed at no cost to the developer by the City, with labor at CJA rates (i.e., $40 per hour in wages and benefits), is approximately $671,000. An affordable housing unit built on purchased land with labor rates reflecting premium costs for certain projects under the 485-x tax abatement law and the premium costs of union labor has an estimated cost of $806,750. Since we assume that 50% of the City-financed units were built on purchased land and 80% of City-financed units will be built with labor at CJA rates, the average cost per unit of City-financed units can be estimated at about $714,000.
We know these costs will escalate due to inflation over the next 10 years. At the same time, cost efficiencies may be realized (as the Mamdani administration is working towards), and some sources of funds, such as rental revenue, will also increase to respond to inflation as the AMI rates on which rent is based increase. Moreover, other financing sources, such as the availability of LIHTCs, may become more plentiful as a result of policy decisions. Our model has the ability to choose whether to look at a 10-year plan in constant FY 26 dollars or to apply an inflation factor using the methodology described below.
Developer Supplied Units and State-Financed Units
Perhaps the biggest unknown factor in the City’s plan for new construction of affordable units is the extent to which sufficient incentives exist for developers to build and finance affordable housing units without City subsidies.
These incentives were significantly reduced in the State Real Property Tax Law section 485-x property tax abatement provisions, which scaled back the more generous provisions in a predecessor statute, section 421-a. Under 421-a, in cases in which City subsidies were not required, private developers generally could receive tax abatement on market-rate units provided that 30% of the project's units were rented at affordability levels of 130% of AMI. Moreover, 421-a, with limited exceptions, had no construction wage requirements.
By contrast, for most projects, 485-x reduces the number of affordable units that can be offered at rents offered to those with more than 80% of AMI and imposes materially higher wage requirements for projects of 100 units or more. Largely as a result of these reduced incentives, the number of mixed-income projects for which developers have filed applications in the two years since 485-x's passage has declined, with nearly all of those applications submitted for projects with fewer than 100 units.
In addition, the City only has the capacity to generate low-income housing tax credits for approximately 5,000 units of housing, assuming each unit takes the maximum advantage of LIHTCs allowed by law. Presumably, the City will reserve that federal tax credit capacity for City-financed units, making them less available for mixed-income projects with cross-subsidized affordable housing units without City financing than has been the case in previous years.
The Block by Block plan states that with $2.5 billion in capital subsidies from the City, the City will be able to support the construction of 8,000 subsidized units annually in FY 27 and FY 28, which implies that private developers will be responsible for financing another 6,000 units annually. The lack of a financing plan in Block by Block makes it impossible to know how many affordable units the City expects private developers will finance if and when the annual rate of production of affordable housing reaches 20,000 units annually or more, which is what would be required to meet Mamdani’s goal.
Historically, based on these programs, between 25% and 50% of affordable housing units are supplied by developers without City capital subsidies. For FY 27, the Block by Block plan implies that of the 14,000 new units of affordable housing to be “created” in FY 27 (i.e., the time at which HPD enters into a financing agreement with a developer), approximately 8,000 will be subsidized units financed in part with capital subsidies from New York City’s Housing and Preservation Department (HPD), while approximately 6,000 will be supplied by developers of mixed-income projects without City capital subsidies and likely with higher maximum income thresholds than the 80% of AMI threshold for City-financed units.
For purposes of the Base Case model, we assume that 40% of the “20,000 new affordable housing units annually” scenario will be supplied by developers or through State financing, without the need for City capital subsidies or LIHTCs supported by City financing. This compares to 43% of the 14,000 units the City projects for FY 27 and is the only reference point for this mix in the Block by Block plan. At 40%, the model assumes that 80,000 units of the 200,000-unit goal over 10 years will be built without the benefit of City capital subsidies.
Obviously, the number of developer-supplied affordable housing units is a significant variable in the model which affects the total amount of City capital needed to realize the 200,000-total-unit plan. Perhaps 1,000 units annually would be developed with capital subsidies from New York State. That would still leave 7000 units annually to be developed and financed by private developers in connection with mixed-income projects.
Given that 485-x only requires developers to include 20% affordable units in a mixed-income project, the assumption of 7,000 developer-supplied units a year (and 70,000 over 10 years) implies construction of an average of 35,000 mixed-income units a year for ten years — well above the rate of construction of mixed-income rental housing in New York City in recent decades.
Cost of Development of Affordable Housing Units
The assumptions we use for each of the variables in the model are inputs that can be adjusted to evaluate alternative scenarios from the Base Case. The Base Case model can be downloaded here.
The following is an explanation of the main assumptions in our Base Case model.
The four main categories of the cost of development of affordable housing units are: hard costs, comprised of the cost of land, non-labor hard costs such as materials costs, and labor hard costs; soft costs (including financing costs during construction); and the developer’s fee.
Hard Costs
The Base Case model assumes that the weighted average Hard Costs are $478,710 per unit based on FY 26 costs.
Siting and Land Costs
The model assumes that sufficient land is available to enable the construction of the affordable housing in our Scenarios. For purposes of the Base Case model, the primary question is whether the land is free (i.e., contributed by the City or another governmental entity) or must be purchased by the developer.
The percentage of affordable housing units developed on land contributed by New York City or another governmental entity in recent years is not readily available. Our Base Case model assumes that 50% of city-financed affordable housing units will be built on land contributed by the City or other governmental entities. Based on feedback we have received from affordable housing developers, experts, and publicly available information, we are assuming that when land must be purchased by the developer, the average cost of land is $50,000 per unit, meaning that the weighted average land costs for City-financed units is $25,000 per unit.
Non-Labor Hard Costs
Non-Labor Hard Costs are principally comprised of core construction, building systems such as mechanical, electrical and plumbing, site work, and construction contingency costs. Based on feedback from affordable housing developers and publicly available information, our Base Case model assumption is that the non-labor hard costs will be approximately $222,750 per unit, which represents approximately 25% of total project costs.
During the announcement of Block by Block, both the Mayor and Deputy Mayor Bozorg said that part of their overall strategy was to reduce the cost of construction of affordable housing. A section of the Block by Block plan addresses reforming building, construction, and housing codes to lower costs and improve accessibility,[5] and suggestions were also offered in the Preliminary Report of the Mamdani administration’s Commission on Government Efficiency. These proposed reforms include changes to requirements for infrastructure such as elevators and plumbing to make them more cost-efficient, as well as to streamline the building approval process. These efficiencies, if achieved, would reduce the cost of both non-labor and labor hard costs.
The Block by Block plan also describes the potential for “industrialized construction” (popularly thought of as modular housing) to reduce costs and accelerate construction time.[6] Facilitating industrialized construction could promote lower costs, but the way in which the Block by Block plan carefully navigates the impact modular housing could have on the construction workforce suggests the political challenge such an initiative would face.
The Base Case model does not assume a reduction of non-labor hard costs for these proposed efficiency initiatives, but these reforms could at least partially neutralize the impacts of inflation over the life of the plan.
Labor Costs
The labor component of Hard Costs is significantly affected by governmental policies that require minimum wage and benefit rates when public funding or tax abatements are involved. There has long been a tension between the competing policy goals of maximizing the amount of affordable housing that can be built and increasing workers’ wages. An increase in minimum labor rates required to receive the benefit of tax abatements was one of the main sticking points in negotiations over the replacement of the section 421-a property tax abatement law, which expired in 2022, and its replacement with the new 485-x law in 2024.
Historically, most affordable housing projects have been developed with non-union labor. The recently enacted New York City local law called the Construction Justice Act requires a minimum wage and benefit package of $40 per hour for any project receiving City capital subsidies. Section 485-x imposes a $40 minimum wage and benefit package for all projects with 100 units or more and imposes substantially higher minimums ($72.45 and $63 per hour) for larger projects in the more affluent neighborhoods of Manhattan, Brooklyn and Queens, known as Zone A and B.
The Real Estate Board of New York (REBNY) testimony to the City Council in mid-2025 stated that “the current 485-x program design is leading to significantly less unit production than seen under 421-a.” Moreover, it has been widely reported that developers are seeking to circumvent the higher wage requirements under 485-x by packaging larger projects as a series of 99-unit projects to avoid the 100-unit minimum that triggers the 485-x wage requirements. If this loophole is closed by subsequent legislation, as has been discussed, it would reduce the incentive for developers to develop mixed-income projects with affordable housing.
During his mayoral campaign, Zohran Mamdani’s housing platform stated that all 200,000 units would be constructed with union labor, which would require payment of the prevailing wage. However, during the announcement of Block by Block, Mayor Mamdani said that this was no longer his intention. Instead, he said that any City-financed affordable housing units would be subject to New York City’s Local Law referred to as the Construction Justice Act (CJA), which requires a minimum blended wage and benefit package for construction workers of $40 an hour. Although this wage floor doesn’t apply to projects of fewer than 100 units or office conversions, it applies to all projects receiving capital subsidies for New York City, so our Base Case model makes CJA the lowest labor rate for the City-financed portion of affordable housing.
While acknowledging that he was abandoning the 100% union labor position, the mayor emphasized that inter-agency working groups would also explore project labor agreements for unionized labor on affordable housing projects where possible, which would significantly increase the costs of development. Our Base Case model assumes that 20% of City-financed units will be developed at the higher labor rates of 485-x A/B or prevailing wage.
Because there is no official cost of non-union open shop labor or for prevailing wage in fully unionized projects, estimates can differ about the premiums to open shop labor costs in the CJA rate and the higher rates imposed by 485-x A/B and the “prevailing wage” rates paid in fully unionized projects or those with project labor agreements. Based on our conversations with developers and housing experts, as well as our review of the considerable literature on the subject, we assume that total Hard Costs will be approximately 14% higher with CJA labor rates than open shop rates and 35% higher with a blended average of 485-x A/B or prevailing wage labor. This implies that the labor cost premium compared to open shop of CJA labor rates and 485-xA/B-prevailing wage is 31% and 78%, respectively.
The table below sets forth the assumed labor cost under each of these scenarios and the percentage of units our Base Case model projects will be constructed under each of these categories.
Construction Labor Wage Impact on Total Hard Cost
Average Number of Bedrooms Per Unit
The cost of construction is also impacted by the bedroom mix of the project. Policymakers and planning documents use “units” of housing as the relevant metric for new development, but of course, not all units are the same. It costs much less to build a studio apartment than it does to build a three-bedroom apartment. Identifying the number of affordable housing bedrooms constructed would be a good metric for the City to report on, while maintaining the metric of units constructed. The focus on units instead of the number of people housed also creates perverse incentives for developers, since costs are based on square footage, not units.
For purposes of determining rent, federal regulations interpolate household size based on the number of bedrooms in the unit and the number of individuals regulations stipulate will be counted for household income purposes for a unit of that size.[7] These regulations identify the following average number of individuals living in various bedroom sizes:
New York City’s Housing Preservation and Development (HPD) agency term sheets suggest that developers limit studios to 25% or less of total units and build at least 30% of units as two-bedrooms or larger. Our Base Case model assumes a bedroom configuration of 20% studios, 40% one-bedroom apartments, 30% two-bedrooms, and 10% three-bedrooms. With this configuration, a 100-unit affordable housing project would house as many as 278 people. The bedroom mix of the affordable housing in mixed-income projects may be different because those developers are not subject to the HPD term sheet. At least anecdotally, mixed-income projects are more heavily weighted towards smaller units.
The unit costs in the Base Case model reflect this HPD-model bedroom mix in affordable housing projects. The minimum square footage by bedroom type is mandated by HPD, which produces a weighted average of 581 usable (or “net”) square feet per unit. Construction costs, however, are based on gross square feet. The model accounts for a 15% loss factor from gross square feet per unit to the net square feet per unit size specified in the HPD term sheet, which results in a weighted average unit size of 684 gross square feet.
However, in practice, the average unit size of affordable housing tends to be larger than the HPD-specified minimum. Based on feedback from affordable housing developers, we are assuming a weighted average unit size of 900 gross square feet.
Soft Costs
Soft Costs are principally comprised of architectural and engineering fees, legal and accounting costs, financing and lender fees during the construction phase, taxes during the construction phase, insurance, permits and approvals, marketing and leasing costs, reserves, and a soft-cost contingency factor. Based on feedback from affordable housing developers and publicly available information, our Base Case model assumption is that Soft Costs will be approximately $175,000 per unit. This excludes developer fees, which are forecasted separately.
Developer Fee
HPD’s term sheet specifies the maximum upfront cash developer fee on a sliding scale based on the number of units, with an absolute cap equal to 15% of improvement costs with certain exclusions and 10% of land acquisition costs. The portion of the developer fee that is not paid in cash upfront is paid on a deferred basis out of future project cash flow. The Base Case assumes an upfront cash developer fee of $35,000, which reflects a 200-unit project size. The Base Case model shows the gross developer fee including the deferred portion as a project cost and the deferred developer fee as a source of funds.
The table below provides a summary of the total uses of funds in development based on a blended average cost per unit:
Blended Average Cost of Development
Per-unit figures assume 200 units.
Financing: Sources of Funds for the Construction of Affordable Housing Units
We define “City-financed” affordable housing as units that receive a capital subsidy from New York City. In addition, such projects typically receive first mortgage senior debt issued by New York City’s Housing Development Corporation (HDC) and may receive subordinated second mortgage debt issued by HDC at a 1% interest rate as a subsidy and LIHTCs either directly allocated by the City or made possible through the provision of private activity bonds loaned by HDC. Other sources of financing and the permanent capital structure may include an equity investment from the developer sponsor and a deferred developer fee. As discussed below, the City has capacity limits that affect the availability of LIHTCs and HDC subordinated loans.
Amount of First Mortgage Senior Debt
The amount of first mortgage senior debt available in an affordable housing project is a function of the debt service coverage ratio (DSCR), which our Base Case model assumes must be 1.15 times the Net Operating Income (NOI) of the project. NOI is based on total rental income (adjusted for certain factors such as vacancy) minus operating expenses. Based on the assumptions below – which generate a NOI of $14,548 per unit – and an assumed interest rate of 6.25% on first mortgage senior debt with a 30-year term, the amount of senior debt per unit that can be supported is $171,213.[8]
Mayor Mamdani’s positions on rent increases for rent-regulated housing are likely to reduce the amount of private senior debt that affordable housing units can support. All affordable units built under the 485-x program are considered rent-stabilized. In view of his campaign promise to freeze the rent on rent regulated apartments for four years (and the decision by the Rent Guidelines Board to freeze such rent for at least the next two years), underwriters of debt for affordable housing are requiring higher debt service coverage ratios and/or higher reserves, which has the effect of reducing the amount of private sector debt affordable housing projects can support.
Rental Income
Total rental income is comprised of residential rental income and ancillary income. Residential rental income generally is calculated based on rent levels established to reflect 30% of that tenant’s household income on a sliding scale based on the tenant’s household income as a percentage of the Average Median Income (AMI) for the tenant’s household size.
The federal government establishes AMI each year for the New York City metropolitan region for various household sizes. As noted above, for purposes of determining rent, regulations interpolate household size based on the number of bedrooms in the unit and the number of individuals regulations stipulate will be counted for household income purposes for a unit of that size. Although 30% of household income is the usual standard for rent, the City announced as part of the Block by Block plan that it would generally only charge 25% of household income to tenants with household income of 30% of AMI or less.
Although rental revenue generally is limited by the AMI affordability levels, rental assistance programs such as Section 8, the CityFHEPS rental voucher program, and the State’s ESSHI program for supportive housing can play a significant role in financing affordable housing by enabling late charges equivalent to the “contract” rental rate (essentially a market rent). The tenant pays 30% of their income under these arrangements, while the funder of the rental assistance program pays the difference up to the contract rent.
Because we expect that the vast majority of City-financed affordable housing units will be built in less affluent neighborhoods, we expect the contract rent to still be below median rents for new apartments in New York City. Our Base Case model assumes that the contract rent will be $3,000 per month, which would still be more than three times the amount of revenue that would be generated by a tenant with a household income of 30% of AMI or less (the income band which most rental subsidies would be replacing). We assume that 10% of all units will generate rental income based on the rental assistance amount (i.e., the contract rate). This rental income enhancement increases NOI and thus expands the amount of senior debt a project can support.
The affordability mix of a project (along with its bedroom mix and the availability of rental assistance, if any) dictates its residential rental income. HPD’s term sheets largely dictate the standard bedroom mix, as described above. The Block by Block plan specifies that City-financed projects will have an affordability mix of 20% of units with rents based on household incomes of 30% of AMI or less, 20% of units with rents based on household incomes of 30%-50%, another 50% with rents based on household incomes of 80% of AMI or less, and 10% at the rental assistance rate. This mix produces a blended average AMI of 59% – just below the 60% requirement for the project to be eligible for LIHTCs.[9] The average gross rental revenue per unit is $2341. The Base Case model applies a 10% discount to this gross amount to account for vacancy, collection loss (which has been growing since Covid), and utility allowances.
In addition, a typical project will also have rental income from commercial space and perhaps parking, but also reflects a discount for nonpayment of rent and utility costs. The Base Case model assumes that this ancillary income increases total rental income by 5%, while rental income is decreased by 10% for non-collection in similar factors. Using our assumptions, this makes the total effective rental income (sometimes called effective gross income or EGI) $26,548 per unit annually in the Base Case.
Operating Expenses
Operating Expenses include cash operating expenses and exclude debt service. Operating expenses include utilities, water & sewer fees, annual insurance, repairs & maintenance, payroll & benefits, a management fee, and administrative expenses. Based on feedback from affordable housing developers and publicly available information, our Base Case estimate is that average per-unit Operating Expenses, for purposes of the NOI calculation, will be $12,000 annually based on FY 26 costs.
Net Operating Income
Net Operating Income (NOI) is equal to effective gross income minus operating expenses. The amount of Senior Debt per unit is determined based on the DSCR and the Base Case model estimate of Total Revenue and Operating Expenses.
Second Mortgage Debt Subsidy
HDC issues second mortgage debt with a 1% interest rate as a subsidy in the amount of up to $65,000 per unit.[10] The second mortgage debt subsidy is funded from HDC’s internally generated reserves. Because HDC is an authority (technically a public benefit corporation) as opposed to a City agency, the second mortgage debt subsidy is considered “off budget” and is not reflected in New York City’s operating or capital budget. Based on the HDC 2025 Annual Report, we believe that HDC issued approximately $188 million of these second mortgage subsidy loans in 2025.
Because HDC primarily must generate this capital internally (it also receives transfers from the Battery Park City Authority), we believe there is a limit on the number of units that it could extend this subsidy to as the number of City-financed units expands. We estimate that HDC has the capacity to support approximately $250 million annually in subsidized 1% loan units at a level of approximately $65,000 per unit, which would cover 3846 total units annually. When this HDC subsidized credit is not available, the difference needs to be made up by increased capital subsidies from the City.
Low Income Housing Tax Credits
Low Income Housing Tax Credits (LIHTC or “tax credits”) are the federal government’s main development subsidy for affordable housing. There are two types of LIHTCs: 9% LIHTCs, which are tightly limited and allocated to states on a per capita basis and allocated within states through a competitive process managed by New York State’s Housing Finance Agency; and 4% LIHTCs, which are as-of-right, but as a practical matter are subject to limitations under the private activity bond volume cap, as explained below. As explained below, a “private activity bond” is a tax-exempt bond under which a substantial share of the proceeds benefits a private use, such as housing, as opposed to a purely governmental purpose, such as transportation infrastructure.
In the case of both 9% and 4% tax credits, the amount of the LIHTC is based on the eligible cost basis of the portion of the project that includes affordable housing (i.e., 100% in the case of a 100% affordable housing project). The eligible cost basis is limited to depreciable costs, so land costs and some soft costs are excluded.
In the case of 9% tax credits, the developer will receive an amount equal to approximately 9% of the eligible cost basis of the affordable housing units in the project for 10 years. In the case of 4% tax credits, the developer will receive an amount equal to approximately 4% of the eligible cost basis of the affordable housing units in the project for 10 years. The current market price for 9% tax credits is 83 cents on the dollar, and the current market price for 4% tax credits is 83 cents on the dollar, although 4% tax credits often trade at a slightly lower price than 9% tax credits.[11] These discounts reflect the time value of money and other market factors. A rule of thumb is that when they are available, 4% credits generally cover about one-third of the cost of development.
LIHTCs are purchased by banks and other large corporations with predictable income streams they are looking to shelter. The market price of tax credits has declined in recent years for a number of reasons, including higher interest rates and competing supply of tax credits as other programs, such as renewable energy tax credits, have expanded. As a result, pricing has generally fallen from around 90-95 cents per dollar of tax credit in stronger markets to lower market prices today. The expanded availability of LIHTCs resulting from changes in the One Big Beautiful Bill Act of 2025 (HR 1 budget reconciliation) may put further pressure on the market price of tax credits.
The eligible basis on which the LIHTC amount is calculated can receive a 30% “boost” if the project is in certain Qualified Census Tracts (QCT) or Difficult-to-Develop (DTD) areas, which is the case for most affordable housing projects developed in New York City. The Base Case assumes that 85% of affordable housing projects will receive the 30% boost.
Private Activity Bond Volume Cap
There are limiting factors on the amount of LIHTCs that can be issued, which significantly affects a program of the scale proposed in Block by Block. The federal government tightly limits the amount of 9% tax credits. New York City typically receives an annual allocation of approximately $12 million-$14 million in 9% LIHTCs, but the amount of 9% LIHTCs was increased by 10% in the HR 1 budget reconciliation act, so we are assuming that $14,500,000 in 9% LIHTCs will be available, which translates to approximately $120 million in equity (assuming the credits are purchased in the market at $0.83 on the dollar).
In the case of 4% LIHTCs, a minimum share of the affordable housing units in a project must be financed with the proceeds of tax-exempt private activity bonds (PAB) to be eligible for the tax credit. That minimum share historically was 50% of the aggregate basis (i.e., essentially the eligible basis plus land costs), but the HR 1 budget reconciliation act in July 2025 (in what may have been that law’s only salutary provision) reduced that minimum share to 25% of the aggregate basis – a change that roughly doubled New York City’s capacity for drawing down 4% LIHTCs.
Although 4% LIHTCs theoretically are as-of-right, developers’ practical ability to access them is limited by the federal government’s volume cap on the amount of PABs it can issue. States are issued PAB volume cap on a per capita basis and New York State’s volume cap is approximately $2.5 billion. Under New York's Private Activity Bond Allocation Act, the statewide ceiling is divided as one-third for State agencies, one-third for local agencies such as HDC and local industrial development agencies (IDAs), and one-third for a statewide bond reserve for use by both the State and local agencies.
According to the Private Activity Bond Final Allocation report issued by the New York State Division of the Budget (DOB) on an annual basis, in 2024 it appears that the allocation of volume cap for New York City generally and HDC specifically was approximately $800 million. Our understanding is that New York City dedicates almost all of its volume to HDC to support New York City affordable housing.
Our Base Case model assumes that HDC has $800 million of PAB volume cap available to it. t this volume and given our other assumptions, the Base Case model indicates the City would not have sufficient room under its PAB volume cap to generate 4% LIHTCs for more than 4,379 units per year.
In the case of 9% LIHTCs, we assume for purposes of analysis that such credits are allocated in an equity amount of approximately $195,000 per unit. On this basis, 9% LIHTCs would cover 619 units, making the total number of units receiving LIHTCs 4,998 on an annual basis. In the case of units beyond that ceiling, the funding shortfall would need to be made up through increased capital subsidies from New York City.
Under H.R. 1's 25% financed-by test, a project qualifies for 4% LIHTCs if at least 25% of its aggregate basis (which, unlike eligible basis, includes land costs) is financed with PABs subject to the volume cap. Because of the consequences of not achieving the 25% level, developers often provide a margin for error, so we are assuming that the actual amount of PAB loans will be 28% of the aggregate basis.[12] An $800 million cap therefore supports a portfolio with an aggregate basis of $800 million divided by 28%, or approximately $2.9 billion. The amount of LIHTC capital that is available, assuming an $800 million PAB volume cap, is approximately $807 billion after excluding land and non-depreciable costs, which is approximately $1.01 billion based on our assumption that the eligible basis receives a 30% “boost” in the case of 85% of all City-financed units because of the locations of these units.
The Deferred Developer Fee and Sponsor Equity
HPD limits the amount of the developer fee that can be received in cash. The deferred amount, which in effect becomes an additional capital reserve, is treated as a source of funds. We assume that half of the total developer fee will be deferred and that this accounts for approximately 3% of the capital structure per unit in a typical project.
Affordable housing developers (also known as “sponsors”) seek to limit the amount of equity they invest in their projects. For purposes of conservatism, we have assumed that affordable housing sponsors would make an average equity investment equivalent to 2.5% of the capital structure.
Capital Subsidies
The pro forma Base Case model makes capital subsidies the “plug” number – i.e., the amount of financing required to cover the total cost of construction of new affordable housing units after taking into account all other available sources of funding. The model assumes that all of this capital is provided by New York City’s HPD. Affordable housing units financed by New York State’s Housing Authority are treated separately. That said, receiving more PAB volume cap and/or capital subsidies from New York State is a likely avenue for the Mamdani administration to pursue in making up for the shortfall in available LIHTCs and City capital to reach the 200,000-new-unit affordable housing goal.
The amount of capital subsidy required is affected significantly by whether LIHTCs are available to support the project. In our hypothetical single project scenario, the amount of capital subsidy required from the City is $253,942 if 4% LIHTCs are available, which covers 29.6% of the total development cost. However, if 4% LIHTCs are not available because of PAB volume cap limitations (and the project does not receive 9% LIHTCs), the amount of capital subsidy required from the City roughly doubles to $485,188.
Given our assumptions about the cost of development and limits on the availability of LIHTCs, we estimate that the amount of capital subsidies needed to develop 8,000 City-financed units annually would be approximately $2.7 billion, which is close to the amount of capital the City includes in its capital budget for new construction of affordable housing in FY 27. However, because of our assumptions that the City has only enough PAB volume cap and 9% LIHTC allocation to support LIHTCs for 4998 units, every 1,000 City-financed units above 8,000 units requires approximately $485 million in additional capital subsidies.
The split between City-financed affordable housing units (which the Block by Block plan describes as subsidized units) and developer-supplied affordable housing units has very significant implications for the level of affordability of these units and the amount of New York City capital subsidies required to meet the 200,000-unit goal. The Block by Block plan states that the City’s increased capital budgets for FY 27 and FY 28 (i.e., approximately $2.5 billion in each year for new construction and Special Needs housing) are sufficient to develop 8,000 subsidized, City-financed units for tenants with a household income of 80% of AMI or less.
This suggests that even if the Block by Block plan assumption that $2.5 billion annually in capital subsidies will produce 8,000 units of subsidized City-financed housing, as the number of City-financed units increases toward our estimate of 12,000 units annually, New York City will need materially more capital than it currently has budgeted. It is possible, of course, that a higher percentage of units than the historical average will come from developer-supplied production, but we consider that unlikely because of the increased headwinds on financing affordable housing that we have described above.
This capital shortfall becomes more acute when inflation is considered. Our inflation estimate seeks to take into account both inflation in costs and increases in revenues over time. Construction costs in New York City grew by approximately 4.4% last year and often exceed the rate of general inflation, particularly given political pressure to increase minimum wages and benefits for construction labor.
Revenue growth is primarily a function of growth in rental revenue. Because all new affordable housing units are subject to New York City’s rent regulation laws, it’s unclear whether rents will increase even if AMI rises if the Rent Guidelines Board imposes a lower rate of increase on rent-regulated units.
Balancing these factors, our model applied a 3% net inflation factor to determine the impact on the amount of capital subsidies required to develop 120,000 units of City-financed affordable housing over 10 years. Applying this inflation factor increases the total amount of City capital subsidies required from approximately $47 billion to approximately $54 billion.
The total estimated sources of funding in FY 26 constant dollars and with a 3% inflation factor are shown in the tables below:
Total Capital Structure: 12,000 Units (Annual) and 120,000 Units (Over 10 Years)
(Not Inflation Adjusted)
Capital Structure for 120,000 City-Financed Units Over 10 Years
(3% annual inflation-adjusted, FY 26 constant dollars)
Four Base Case Model Scenarios
The Base Case model includes four scenarios: 1) a hypothetical single 100% affordable project with 200 units partially financed with 4% LIHTCs; 2) ) a hypothetical single 100% affordable project with 200 units partially financed with 9% LIHTCs ; 3) 14,000 units annually, which reflects the expectation set forth in the Block by Block plan that there will be 14,000 starts of new affordable housing in FY 27 – of which 6,000 will be supplied by developers and 8,000 will be financed by New York City; and 4) 20,000 units annually, with an assumption that 8,000 will be supplied by developers or financed by New York State and 12,000 will be financed in part with capital subsidies from New York City.
These Scenarios assume that New York State will finance 1,000 units of supportive housing annually, with no contribution required from New York City.
The percentage of affordable units that developers are required to supply ranges by program and through negotiation with the City. Under 485-x, developers of mixed-income projects must include 20% affordable units to be eligible for tax abatement. The assumption that 7,000 units annually, or 35% of all affordable housing units, will be fully financed by cross-subsidization from market-rate apartments implies that 350,000 market-rate apartments would need to be developed to support the supply of 70,000 affordable units without City capital subsidies over 10 years. This rate of market-rate development far exceeds recent construction levels. Moreover, as we have described above, recent developments may make it more difficult to finance affordable units within mixed-income projects. As a result, even the assumption that 35% of the units will come from private developers in mixed-income projects may be optimistic.
The Base Case model is available for downloading here. The model includes granular assumptions that are based on numerous conversations with housing developers and housing experts. We believe that, on balance and in their totality, our assumptions are reasonable, although we are certain that any single assumption is less certain than our confidence in the whole. Since our main point is that the City should put out a financing plan for Block by Block, if our assumptions are wrong, they could be corrected by an official plan.
Based on the assumptions described above and the current availability for capital subsidies, the Base Case model suggests that New York City would require approximately an additional $25 billion-$35 billion (depending on inflation) in capital subsidies over and above the capital already dedicated to the construction of new affordable housing and Special Needs housing. This assumes that the approximately $10 billion in such capital subsidies in the Mamdani administration’s current capital strategy covering years FY 26-FY 30 would be matched in years FY 31-FY 35.
Conclusion
There is a wide consensus that development of additional housing – and especially more affordable housing – is the most pressing social and economic challenge facing New York City. The Block by Block plan outlines numerous strategies to address this challenge, but it does not provide a viable financing strategy for its core promise to construct 200,000 units of new affordable housing or to additionally preserve 200,000 units of affordable housing.
Our analysis using the Base Case model indicates that there will be a significant shortfall in both the availability of LIHTCs—because of the current amount of PAB volume cap directed to New York City—and the amount of HPD capital required to achieve the 200,000-unit target, given the City’s current capital budgets and plans.
The assumptions used in the Base Case model reflect our best efforts to identify the expected costs of affordable housing and the availability of sources of funds at various scales of annual development. This analysis is intended to begin a conversation, not meant to end one. Other observers and housing experts with more experience than we have in this area may suggest adjustments to our assumptions that would result in a larger or smaller shortfall of LIHTCs or HPD capital. And, as noted earlier, it would be particularly helpful if the Mamdani administration released its own analysis of the economics of implementing its plan to develop 200,000 new units of affordable housing over the next 10 years so that the policy debate can begin about what measures can be taken to remedy the situation.
Endnotes
[1] Annual Report on Capital Debt and Obligations, Fiscal Year 2026. New York City Comptroller.
[2] Debt Management Policy: New York City General Obligation and New York City Transitional Finance Authority. City of New York. September 2023. p. 9.
[3] New York Housing Conference 2026 NYC Housing Tracker Report. p. 3.
[4] The Block by Block plan says that it “use[s] the term ‘affordable housing’ to describe housing that is regulated by the government and which residents qualify for based on their income.” (Emphasis added.) The Block by Block plan seems to differentiate between subsidized affordable housing units with New York City capital subsidies, which are limited to those with household incomes of 80% of AMI or less, and other types of housing that still meets the broader definition of “affordable housing.”
[5] Section 7.3
[6] Block by Block. p. 99.
[7] Under Section 42 of the Internal Revenue Code, the maximum rent for a tax credit unit is not derived from the actual size or income of the household occupying it. Rent is instead capped at 30 percent of an imputed income limitation, and that limitation is selected by reference to a hypothetical household size determined solely by the unit’s bedroom count. IRC § 42(g)(2)(C) fixes that hypothetical size at one individual for a unit without a separate bedroom, and at 1.5 individuals for each separate bedroom in any larger unit. [Note: footnote written by Claude.]
[8] The maximum amount of senior debt is determined in two steps. First, the project's annual Net Operating Income (NOI) is divided by the required Debt Service Coverage Ratio (DSCR) of 1.15 to determine the maximum annual debt service (principal and interest payments) that a lender will permit. For example, with an NOI of $14,558 per unit, the maximum annual debt service is $14,548 ÷ 1.15 = $12,650. This amount is then converted into a loan principal using a standard mortgage amortization calculation assuming monthly payments, a 6% annual interest rate, and a 30-year term. Under those assumptions, an annual debt service of approximately $12,650 supports a first mortgage loan of approximately $171,213 per unit.
[9] Because the tenant in a rental assistance program pays only 30% of household income, this is treated as a rental at 30% AMI for purposes of determining LIHTC eligibility.
[10] The characteristics of our Base Case and Scenarios – i.e., less than 60% AMI and a high percentage of affordable units – imposes the restriction for HDC’s issuance of second mortgage debt under the “Extremely Low & Low-Income Affordability” (ELLA) program, which imposes a maximum per project of $15 million. This $15 million ceiling comes into effect after 237 units. Since our Base Case model assumes an average of 200 units per project, this project cap does not factor into our aggregate estimates in the 14,000 unit or the 20,000-unit annual production scenario.
[11] LIHTC Equity Pricing Trends: Equity Price Per Credit Averages History, January 2016 through March 2026. Novogradac.
[12] Because the amount of senior debt a typical affordable housing project can support is less than 25% of the aggregate basis, HDC must issue temporary debt to reach the 25% minimum. The amount of temporary PAB debt issued by HDC in the construction phase is often 50% of total development cost because of the absence of more attractive construction finance resources. However, we assume that HDC will use “recycled bonds” for PAB loans above our assumed 28% target. Although this temporary debt consumes volume cap while outstanding and reduces the cap available for other projects, federal law permits the issuer, within a limited time frame, to re-issue bonds in the same amount as the temporary debt that has been repaid without drawing on a fresh allocation of volume cap. Recycled bonds provide tax-exempt financing but do not themselves generate 4% credit eligibility; only bonds issued against fresh volume cap satisfy the 25% financed-by test.