How Do We Make Housing More Affordable in Urban San Diego?

Continuing on some recent posts, I wanted to explain a little more where I am coming from on the idea of non-profit infill development as a tactic to increase affordability.  This is a little wonky, so feel free to skip if you just like the more broad brush dreaming.  

Still here?  OK, a few givens that I won't discuss here:  

  1. Housing in San Diego is unaffordable and this is a bad thing.
  2. Housing near central public transit (aka "Transit Oriented Development") is better than suburban or rural development for many reasons, ranging from affordability, to health, to the environment.
  3. San Diego County needs about 390,000 new housing units in the next 35 years to keep up with population growth (let's call it 10,000 units/year).  This isn't happening.
  4. Related to point 2, transportation costs add a lot to the cost of living in San Diego, so housing with low transportation costs (i.e., in walkable areas close to transit) is more affordable.

So how do we create more affordable housing?  Housing prices are inelastic and difficult to comprehend, with both supply and demand effects.  Simply put, there is not much supply in San Diego and a lot of demand.  We don't want to mess with demand, so we have to mess with supply.  While seemingly difficult to precisely quantify, the greater the supply, the lower the price.  For example, if we could just create a million new housing units with the wave of a wand, housing prices in San Diego would plummet because the supply would far exceed the demand, until more people move to San Diego to take advantage of the low housing costs.  We don't want this to happen, but in the real world, it is safe to say that creating a new housing unit has a negative effect on price.  That is, building housing units lowers costs.  To keep costs down, we need to not only increase supply, but also keep pace with demand. This doesn't seem to be happening yet in San Diego, so in reality building a new unit just means prices will increase at a slower rate.  Bottom line, to increase affordability we need to build as many units as possible and as close as possible to transit. Some of this is happening with market rate development, but the market isn't keeping up with demand.

Speaking more specifically, there are ways to maximize affordability when building those units.  Broadly speaking, the cost of housing can be defined as: 

(Cost of vacant land) + (Construction Costs) + (Developer Profit Margin) = Cost

This gives us four variables to work with.  The current affordable housing scheme works largely only as to one variable, cost, by subsidizing the actual cost for residents, either by accepting below-market rent that doesn't cover costs or paying landlords on behalf of low-income tenants.  There is also some work in the other variables, but most of the focus is on cost.  The City can also increase affordability by raising income, like raising the minimum wage.  Other affordability policies, like density bonuses, create additional affordable units, but potentially at the cost of higher prices and less affordability for everyone else (If some units in a project are being rented/sold below cost to low income tenants, the other units have to increase in cost to maintain profitibility).  So the question is, how can we create more affordable housing by attacking the other three variables?

1. Cost of vacant land.  This is where the majority of the unaffordability comes from in San Diego.  Speaking broadly again, the cost of the land is driven in large part by the potential of that land to derive income (simply: farming; more complex: rents from development).  One way to "lower" the value of the land is to build denser development.  This actually makes the land more valuable in most cases, but it also allows a greater number of units to be built on the same land, leading to a lower land value per unit.  If we replace a single house with 10 units and everything else is constant, our land cost per unit decreases.  (example: a million dollar lot with one house leads to a land cost of a million dollars/unit.  Putting 10 units on the same million dollar lot leads to a land cost of $100,000/unit).  Increasing density as part of zoning changes, or ending parking minimums (parking takes up land and costs money to build) can lead to "lower" land costs, but that's a huge topic for another post.

2.  Construction Costs:  This is comprised of hard costs (lumber, windows, faucets, etc.), labor (construction worker wages), and soft costs, meaning architects, consultants, and permit fees. Here, affordable housing often costs more to build than market-rate housing, given the need to build to greater durability, include certain amenities, and pay prevailing wages for labor.  To the extent housing is built with loaned money, affordable housing does save on financing costs.  

So how do we save on construction costs?  Hard costs aren't going to change, in a basic sense, but simpler units could be built that are designed well but done without frills. Let's assume no change in permitting costs (again, another post).  The only other real opportunity here is to lower the costs of architects and other design and permitting consultants.  Large suburban developers do this by creating a few models and then reusing the model multiple times.  This doesn't happen as often in one-off projects.

3.  Developer Profit Margin:  There's no way to get around this with market-rate building, and in a sense, it is sometimes eliminated with affordable housing development.  Speaking simply, let's say this adds 5-10% to the housing cost.  A non-profit developer would forgo profit entirely or, depending on the sales strategy, partially (discussed below). 

Under this broad calculation, how can we build more affordable housing? (not using "affordable" in the technical sense, just meaning cheaper)  Summarizing the discussion above, there are a five key goals that a non-profit infill developer (NPID) could hit, some of which complement the current affordable housing development scheme and others that are in new territory:

  1. Generally, increase the number of units being built:  one more developer out there is better than none, regardless of whether profit is being earned.
  2. Build units close to transit and the urban center: the NPID would only build in proximity to transit and in the urban core.
  3. Build units as densely as possible:  target underutilized property and build as dense as possible given the zoning.  Use the density bonus to increase density and lower parking requirements if it results in lower overall costs.
  4. Lower design costs:  the NPID would start with case study developments to formulate the best architectural design and then iterate off that plan for future developments. Architectural fees can add up to 15% of the cost.  The plans would have inherent aesthetic flexibility to avoid monotony.  Once the first development is approved and built, the continuing costs of reusing that plan drop dramatically. As a value multiplier, these plans could then be sold to private landowners for a low cost to use on other projects.  If construction costs drop, more landowners may undertake their own redevelopment.  Similarly, by building nearly identical units in the same areas, the NPID will be more efficient at getting projects through the entitlement process.  As I discussed briefly in the last post, the NPID could also develop standardized ADU plans and provide them, along with permitting assistance, to homeowners.  Perhaps even financing assistance could be provided (we'll build the unit for you in exchange for percentage of rent until costs paid off).  
  5. Forgo developer profit:  This amounts to 5-10% of a project cost.  By being non-profit, there are no investors to please.

A non-profit infill developer could hit all of these goals, realistically building units coming in at 10-20% below current market prices.  Although this would still not reach the level of affordable housing, it would be cheaper housing.  As more units are built, prices would continue to fall (or, most likely, not rise as steeply). 

There are several selling strategies. The NPID could sell half the units are market price to increase funding for the enterprise and to cover for inflation, etc.  The NPID could sell the other units using certain (legal) criteria, like income and local relocation (only current SD residents can buy, or only those currently in the neighborhood but living in less dense housing can buy). If there was some limitation where only neighborhood locals can buy, some gentrification concerns could be alleviated. To avoid flippers, the sale could include a buy-back option for a set period of years.  With this development, the NPID could target middle-income people, not the low-income people targeted by current affordable housing programs.  Thus, it is a supplement to low-income housing, not a replacement.

Best of all, unlike affordable housing programs, the NPID would be self-sustaining.  By building and selling units at cost, the NPID wouldn't need constant influxes of charitable or governmental contributions to survive.  Once one project is complete (or more if funding permits) and sold, the NPID would then move on to the next project.  Any additional contributions could be used to build more units, not to subsidize costs.

The NPID would be lean and nimble, with a small core of employees as a design-build team to run the development for a fixed salary.  They wouldn't get rich, but they would earn a decent salary and be doing good  and fun design work.  For a relatively small donation (a couple million), a donor could help increase the supply of housing in San Diego.  As affordable housing works from the bottom up,  the NPID would work from the top down, working in smaller sites not hit by large affordable projects.  Who's in?