I am trying to wrap my head around Scott Wiener’s bill SB 828. There’s a lot in it I really like, and there’s things that confuse me… particularly how it relates to the existing RHNA priorities. For example, the law would have the Department of Housing and Community Development (HCD):
- What datasets will we use to identify median rents and median incomes? For smaller jurisdictions we will have to use the five year rolling average of the American Community Survey, meaning a hypothetical assessment done by HCD at present would require the department to use ACS data that is a combined sample of surveys taken between 2012 and 2016. Maybe for long-term regional planning this is actually appropriate… I don’t know. But congress is probably going to continue to weaken the ACS, making even the 5 year wave estimates problematic (what do you do when the margin of error on the rental estimate is so big there *could* be a housing a crisis and there *could* be a glut–there are answers to that question that involve some aggressive computational geography, but no precedent for using those tools in policy just yet).
- How do we relate median incomes to median home values? The classic ‘rule of thumb’ is that a unit is affordable (for potential owners) if the price is 3 times the potential owner’s annual income. How will HCD operationalize what the median income can afford in each locality?
The latter question matters because when I tested this against ACS data, I found most cities would only pass the (b)(1) high cost tests under this owner rule of thumb (annual income 3 times price), not under the rental median test. This is because median incomes are actually high enough to afford median rents in many cities. That may surprise you. If it does, the main issue here is the reliance on medians which do not really capture ‘affordability’ problems for most people below said median incomes. Another issue is that homeowners have higher incomes than renters, so when you compare the median rent to median income (as opposed to the median renter income), rents dont look too bad. I’m not sure if SB 828 as written would empower HCD to differentiate median incomes by tenure in determining who needs to be ‘rapidly increasing housing supply’ as the bill states.
I have mapped which cities pass the (b)(1) high cost tests (and thus are high cost). Below is the Bay Area Map.
There are a few caveats/things to note here:
- I guess the rental estimates were not robust enough in many Silicon Valley cities (the ones in the west, in blue on the map). These cities still count as passing this (b)(1) high cost test with respect to home ownership prices (this is a given). But there are apparently so few rentals in those cities captured in the ACS sample that the Census Bureau was not comfortable publishing them. (Maybe we say any city with more than 20,000 people that has too few rentals for the ACS to estimate median rents is automatically ‘too expensive’ by virtue of its exclusionary nature–that would apply Los Altos, CA ala the map above).
- There are very few communities that pass both of these high cost tests, and there are two clusters of them:
- Expensive coastal towns (Santa Cruz, Seaside) and Watsonville (coastal hyper-gentrification???)
- Disadvantaged inner-suburbs and southern suburbs (Florin, Parkway, Fruitridge, Pocket, Lemon Hill–of Sacramento that area not part of the city proper).
Here’s the Map for Southern California:
Here data issues are less of a problem. A few things to note:
- A few pockets where both high cost tests are applicable, mostly places where incomes may have crashed and not recovered from the recession like Adelanto and Victorville, Sun Village.
- There is one pocket of places where only the rental test applies: Winchester and Homeland in Riverside County.
- And low and behold our first set of cities which are not high cost under either test! Pala (San Diego County), Green Acres & Warm Springs (Riverside), Mojave CDP, Rosamond and California City (Kern County), and Barstow, Lentwood and Silver Lakes (San Bernardino County).
So to summarize, via cutting and pasting the language in SB 828: “particularly housing supply for moderate and above-moderate income households” “will be” “rapidly increased where median rents or home prices exceed median incomes” which is almost everywhere under traditional definitions of affordability for home ownership.
HCD and the MPOs/COGs will be interesting places to work at if this passes and the old ‘common sense’ approaches to measuring affordable home ownership are applied. I’ll get to (b)(2) of the law in my next post.