In the press coverage of almost any public policy proposal there is a lot of debate about the numbers, especially how much new revenue might be raised and how it would be used. When it comes to Mandatory Inclusionary Zoning (MIZ) and Seattle’s version of it, Mandatory Housing Affordability (MHA) there seems to be very little curiosity about how the mandate to include affordable housing or pay a fee will turn out. How much money will be raised with the fee? How much subsidized housing will all that money pay for? These questions raise the stakes on figuring out how much non-profit subsidized housing really costs; if the fees generated by MHA are going to pay for units, shouldn’t we know how much those units would cost?

Armed with a calculator and plain reading of the online documents, I’ve come to the conclusion that, yet again, the MHA proposal simply doesn’t make sense on the spending side either; that is, in order to pay fees housing prices will go up but the fees generated don’t seem likely to cover the cost of very much housing.

Back in 2015, the MHA’s main assumption was announced by the Mayor: housing production for 10 years ahead would more or less match housing production for the previous 10 years. The idea was that MHA would be consistent with the Comprehensive Plan and previous real data. On page 2 of a document called Affordable Housing Production Model Summary, the City projects growth will occur “over the next 10 years, based on data from the last 10 years.”

That data in a report generated by the City’s Department of Construction and Inspection (formerly DPD), shows that about 47,000 units of housing were permitted between 2005 and 2015. Based on a conversation with a member of the Housing Affordability and Livability Agenda (HALA) Committee, the Mayor one day announced that the production of new housing would be 50,000 units over the next 10 years, 30,000 market rate and 20,000 affordable.

While these numbers align, it’s hard to find anything other than this guesswork in terms of what MHA assumes for production. It matters because if you’re going to charge a fee for something you ought to have of what it is supposed to pay for. In the first document I cited, and from what I remember, the number of subsidized units that would be created by MHA fees is pegged at 6,000 units. The number 50,000 appears in a later document, Implementing MHA Citywide (2017) again, with the Mayor’s numbers.

Let’s look at the numbers from the earliest documents I mention above. That model summary assumes that 50 percent of housing projects outside of Downtown and South Lake Union will include affordable units at somewhere around half of 6 percent inclusion of 30,000 units, or 900 units (30,000 * .06 =1,800/2 = 900). The rest of the units would be paid for with MHA fees. Let’s take a closer look at that.

The City’s model is that “new affordable housing funded by the Office of Housing (OH) is assumed to require a contribution of $80,000 per unit from OH (based on a model project leveraging four percent low-income housing tax credits and no additional public funds” (emphasis mine). Let’s take a look at an example, 12^th Avenue Arts.

Let’s go with their board of directors argument that the project really didn’t cost $47 million dollars, but about half that, at least for the housing portion, or about $23 million or about $260,000 per unit. Using the City’s model, that project would have been eligible for about $7,000,000 in MHA funding. Based on a ULI case study, the project received $5,600,000 in 4 percent tax credit equity. If it received no additional public funds, the project would have about $12,600,000, or about $10,000,000 less than it needed to complete the project.

Keep in mind that the 50 percent participation is aspirational, so the likelihood that the number of units from performance will be less and, many of the projects in low-rise zones are for sale projects that won’t include subsidized units. While it is true that as performance rates fall, fees go up to offset the increased number of MHA funded units to hit the 6,000 goal, it doesn’t look like the $80,000 per unit figure is realistic given a dramatically discounted per unit costs I used.

None of this means that projects like 12^th Avenue Arts won’t use other public sources of money, they would be allowed up to $1,500,000 in Housing Trust Funds (HTF), for example. Still, they’d be $8,500,000 short and that’s assuming a price tag weirdly and illogically discounted by 50 percent. The notion that $80,000 per unit is going to substantially offset the production costs of new non-profit subsidized housing strains credulity.

More importantly, if the fees suppress supply and eventually get added to the costs of new housing, those units will be less affordable and will add to the demand for non-profit subsidized units. If we believe the $427,000,000 figure for fees charged on roughly 30,000 units, that comes out to about $14,000 per unit — that’s about $1,400,000 for a hundred unit building, an astronomical increase in cost. Also keep in mind that the fees go as high as $35 per square foot, which means some projects, if we grant the City’s average estimate for size of housing, about 951 square feet, would be paying $3,300,000 for 100 unit building in a place like Capitol Hill.

The math gets even crazier if you consider the latest number is the most recent document of “20,000 affordable homes.” I did some math there and found that even backing out the 6,000 affordable units, and assuming the $80,000 allocation of MHA fees, the program would have to generate $1.2 billion in revenue, or about $33,000 per unit, to pay for those affordable units – and that’s assuming that those units can be produced with at least $2 in leverage from other funds ($160,000 per unit) and be completed at a cost of about $250,000 per unit.

I suppose someone else with time and a calculator could push these assumptions around, predict lower production costs, higher participation and market rate production, and get numbers that end up generating enough fees to cover the production costs of more non-profit subsidized housing. But my numbers are pretty conservative, that is, they are more or less based on the City’s model. Reality seems to be painting, and usually does, a different picture. Less participation means higher prices, and more demand for non-profit subsidized housing that gets more expensive as it seeks more and more sources of capital, even with MHA.

The bottom line here appears to be that the City’s projections are hopeful guess work and that there is no way to predict exactly how many market rate units will be produced over 6 years (an even shorter period of time than originally assumed) and how much demand there will be for fees to pay for non-profit subsidized housing. What’s most likely is that the massive fee hit many projects take won’t be enough, so non-profits will want even more. Yet when I read stories in the press they are all about the process and not the substance, NIMBYs and YIMBYs and no math, so I have yet another Mugatu moment: doesn’t anybody see this? I feel like I’m taking crazy pills!