“A problem well posed is half solved.” – John Dewey.

The converse is also true, a problem poorly defined will never be solved. In my 15 years doing public policy analysis I’ve encountered few problems as poorly defined as the “affordable housing” problem.

Here in Santa Cruz County, depending on who you talk to, we either have an affordable housing problem, an extreme affordable housing shortage, or a housing crisis.

The claim that I will base this and future posts on is the following: I don’t think anyone has a really well-developed idea of what defines our housing ‘crisis.’ To be clear, everybody can offer you some definition:

• people with good jobs can’t find housing
• people who have lived here their whole lives are being ‘priced out’ of their homes
• median home price/rental rate is too big a fraction of median income

These are certainly symptoms of a problem…but they aren’t particularly strong metrics upon which to base public policy strategies that might address our problem/shortage/crisis.

Resources

Although I will cite them throughout the following discussion when they are relevant, here are the main documents I will be drawing from:

The 2015 Santa Cruz County Housing Element

The Housing Element was specifically referenced in our mayor’s, “State of the City” address.

The Housing Element draws in a few important places from:

An Economic Vitality Study

and

Regional Growth Forecasts made by the Association of Monterey Bay Area Governments.

The City of Santa Cruz 2016-2017 Action Plan. A document by the Santa Cruz Housing and Community Development Division and The Santa Cruz Economic Development Department.

A UCSC led research project called No Place Like Home.

Yes, these are long. Yes, I read all of them. No, you don’t need to read them. That’s what I’m here for.

Defining the Issue

Claim 1: The current discussion about the affordable housing ‘crisis’ in Santa Cruz County lacks a precise definition of the problem.

Based on my extensive reading of reports to the County Board of Supervisors, attendance at various public forums (including the mayor’s recent State of the City gathering), and participation in the Monterey Bay Regional Economic Partnership Planning Committee I have determined that 3 documents appear central to the current discussion of affordable housing in Santa Cruz:

The 2015 Santa Cruz County Housing Element

The 2014 Santa Cruz County Economic Vitality Report

The City of Santa Cruz 2016-2017 Action Plan.

These documents when read together give a very detailed overview of the County’s current strategy for increased provision of affordable housing. They are unamimous in their recognition that we (Santa Cruz County) are indeed in the throws of a ‘housing crisis’. These documents are extensive, they are delightfully data driven in a way that municipal governance documents generally are not, and they reflect a deep commitment on the part of the community and city and regional governments to address the issue of availability of affordable housing. HAZAH, Good On-Ya.

However, a thorough reading of them reveals that nowhere in any of the planning/strategy/action documents does the city/county/whoever actually state the conditions under which we will deescalate from housing ‘crisis’ (presumbly to run-of-the-mill ‘housing problem’, ‘housing crunch’, or some other thing less scary than ‘crisis’). I find this very problematic. Let me offer an example:

Pg. 4-113 of the Santa Cruz County Housing Element lays out the following 6 primary housing goals:

Goal 1: Ensure land is available to accommodate an increased range of housing choices, particularly for multi-family units and smaller-sized units

Goal 2: Encourage and Assist in the Development of Housing

Goal 3: Remove Unnecessary Governmental Constraints to Housing

Goal 4: Preserve and Improve Existing Housing Stock and Expand and Preserve the Continued Availability of the County’s Existing Affordable Housing.

Goal 5: Promote Equal Opportunity and Production of Special Needs Housing Units

Goal 6: Promote energy efficiency in existing and new residential structures

It is worth pointing out that none of these are actually goals in the sense of being specific, measurable aims. These are actions. “Ensure land is available to accommodate an increased range of housing choices, particularly for multi-family units and smaller-sized units” is not a goal, it is a recommendation. It is not defined in such as way as to ever know if you’ve accomplished it or not.

Everyone seems to agree that we are in an affordable housing ‘crisis’. I’m not debating whether we are or we aren’t. What I am saying is this: currently in Santa Cruz we have an extensive, well documented, and generally well-reasoned strategy for addressing the housing ‘crisis’. There are multiple agencies working across multiple juristictions to develop specific actions to decrease housing insecurity for low income, special needs, and other vulnerable populations.

What we don’t have is a way of knowing when/if this combination of strategies, policies, and actions has worked. That is,

If we are in a ‘housing crisis’ when will we know that the crisis has been abolished?

A Few Examples

Example 1: Cops and Nurses

The statistics that people generally use to convey a housing problem are the ratio of home prices to median incomes and ratio of rents to median incomes. In my experience, people often prefer to put a face on the problem. In some circles there is a belief that the people in service professions (police officers, fire men, nurses, school teachers) should be able to afford to live in the communities they serve. If this agrees with your notion of affordable housing then I would propose the following specific, measureable goals associated with this notion of affordable housing:

Achieve a 30% homeownership rate across the following ‘preferred professions’: police, fire, public school teachers, nurses.
Achieve a median monthly rent that is no more than 30% of the median monthly pay in the ‘preferred professions.’

In the interest of developing a real success metric here we probably want to go even a step further and say, “Achieve a 30% homeownership rate across the preferred professions by 2020”..or 2025…or 2035, whatever.

Here it is worth noting that the median salaries for cops (65k USD) and elementary school teachers (62k USD) in Santa Cruz County are already consistent with the above 30% rule (65k/yr is about 5,416/month which is about 3.01 times the median rent of about 1,800/mo). The homeownership metric might be a little harder nut to crack.

Example 2: Everybody

Example 1 was consistent with a pretty conservative definition of affordable housing. It basically said, as long as cops and nurses can get housing then we’re all good. On the opposite end of the spectrum we could go ultra liberal and say that we won’t consider Santa Cruz County affordable until every single person who wants to live here can find housing that fits their budget and lifestyle. This goal would be both specific and measurable since meeting this goal would necessarily mean observing exactly 0 growth in rental rates and 0 growth in home prices.

This is a patently ridiculous definition but it serves as a good juxtoposition to Example 1. If we are going to say that we are in a housing crisis because rent is too damn high, then we need to know what rent SHOULD BE in order for us to declare that we are no longer in a crisis. People often don’t like to frame the issue in these terms because:

For good reasons there are some negative connotations associated with meddling in real estate markets in order to manipulate prices, and as soon as somebody commits to a number (say we determine that achieving median rent of $1300/mo means we are no longer in a ‘crisis’) some stakeholder will immediately get pissed that his/her economic circumstances aren’t covered by the new definition of affordable.

In order to avoid the complications that come with taking about hard numbers, people often like to define ‘affordable’ using anecdotes about people who can’t find adequate housing.

We might all generally agree that Example 1 is too rigid. Most of us probably have some vague sense that other ‘good jobs’ (bank teller, car mechanic, etc) should also be able to afford to live in Santa Cruz. And we probably all agree that Example 2 is too vague. We can’t have a housing policy that literally says that anyone on earth who wants to live in Santa Cruz should be able to. So in the middle is where the real action is.

Affordable housing policy is all about who should be able to afford to live where. Nobody wants to frame the issue that way because it’s pretty dickish….but it’s the only way that will ever lead to a definition of the problem which allows us to derive the conditions under which we determine that the problem has been solved.

Example 3: Median Income/Median Rent Ratio

These metrics were mentioned above but they deserve a more focused treatment. From what I have observed the affordable housing ‘crisis’ in Santa Cruz County is often described using these metrics: The median home price (~850,000) is about 12 times the median household income (65k/yr) and the median family rent (2,154) is about 40% of median monthly household income. People in Santa Cruz generally have no qualms about telling you that median home price of 850k is ‘crazy’, ‘insane’, or ‘ridiculous’.

But if median home price of 12 X median household income is ‘crisis’ and median home price of 3 X (or less) gross annual income is affordable, where is line between ‘unaffordable’ and ‘crisis?’ When does unaffordable become a full blown ‘crisis’? 4 times median household income? 5 times median household income? 8 times?

Quick Aside Here: Yes, I know HUD and others have defined what ‘affordable housing’ means. It’s an income based formula that defines rental rates that are within reach of above moderate, moderate, low, and very low income households. The household income levels for Santa Cruz can be found here. To get the ‘affordable’ rent for these groups divide the gross annual income by 12 and take 30% of that number. I’m not talking about defining what ‘affordable housing’ is. I’m talking about how we know when we’ve provided enough of it. If we are in a ‘crisis’ now because X number of resident in our county can’t find affordable housing how much do we need to reduce X by in order to get out of ‘crisis’ mode?

One reason we might not want an affordable housing policy that is exclusively based on the ratio of median rent to median income is that the easiest way to get median rent to be a digestable fraction of median income is to kick all the poor people out of the county. If every low income resident of Santa Cruz County moved to Seaside (23 miles away in Monterey County) our median income would go up, ratio of housing units to bodies would go up which would put downward pressure on rents and, Voila! The median rent:median income balance would improve dramatically.

I agree that this would be a pretty shitty way of doing things, so maybe median rent to median income isn’t the only metric we want to use. But, in my experience, too many people have gone all the way to the other end of the spectrum and said, “Well it’s a complicated problem with many facets. We want to provide housing that:

• retains socio-economic diversity
• is consistent with sustainable land-use and preservation of amenity values
• encourages use of public transportation
• etc, etc.

That’s all well and good and those are great guiding principals but somebody has to translate them into actual housing units that we want to supply to low income households. Just because the problem is complicated doesn’t absolve planners of the responsibility of proposing specific goals. Just because the choices are hard doesn’t mean they aren’t choices. Just because one metric is over simplified doesn’t mean the problem can’t be quantified.

Example 4: Section 8 Waiting List Decrease

Another subjective but not inconsequential indicator of our affordable housing problem is the extreme excess demand we have for housing assistance. The Section 8 Housing Voucher Program is a low income housing assistance program that works by granting low income households a voucher. They use the voucher to get rental housing on the open market and pay ⅓ of thier monthly income to the landlord in rent and the Housing Authority reimburses the landlord for the different between what the household pays in rent and the market rate (there are caps on how much the HA will reimburse).

In Santa Cruz County the waiting list for a Section 8 Housing Voucher is currently closed and there are around 5,000 households on the waiting list.

Another possible example of a specific, measureable goal of housing policy might be:

Reduce the Section 8 Housing Voucher Wait List by 2,500 people.

We could, of course make the goal to supply all 5,000 households currently on the Section 8 waiting list with affordable housing. But again, eliminating affordable housing need altogether seems like a pretty high bar just to move out of ‘crisis.’ What magnitude of reduction in the Section 8 waiting list would be consistent with just a regular old ‘affordable housing problem’ as opposed to a ‘crisis.’

Example 5: The Regional Housing Needs Assessment

Here I’m just going to launch a quick preemptive stike against those who might say that our housing needs have been quantified because we have a Regional Housing Needs Allocation Plan that does set specific targets for the number of additional housing units Monterey and Santa Cruz County should supply by 2023.

It is true that the RHNA set income specific housing targets, which are the number of units we need to add in each income category by 2023. If you refer to page 15 of the RHNA you will see the following recommendations for Santa Cruz County:

Geography	Total Allocation	Very Low Income	Low Income	Moderate Income	Above Moderate
Santa Cruz County	3,044	734	480	554	1,276
Capitola	143	34	23	26	60
Santa Cruz	747	180	118	136	313
Scotts Valley	140	34	22	26	58
Watsonville	700	169	110	127	294

However a thorough read of the RHNA makes it clear that these recommendations are what is needed in order to keep pace with projected population and employment growth. That is, if we are in an ‘affordable housing crisis’ now, meeting these targets will ensure that our crisis doesn’t deepen by 2023…but they are not recommendations for improving the state of our affordable housing affairs. Stated a different way, meeting these targets exactly will ensure that we have the same level of ‘affordable housing crisis’ in 2023 that we have today.

Why Does it Matter

In simple terms, I think precision matters for two reasons:

1. Responsible stewardship of public money, and
2. Related to responsible stewardship, when problems aren’t precisely defined mythology is allowed to propogate and eventually becomes accepted wisdom.

The first of these I will discuss here. Item 2 will be picked up in my next post.

Responsible use of public money: estimating relative impact

The biggest reason that it matters to me is that Santa Cruz County has at least 35 million USD of public money invested in various programs that address affordable housing need. I have no reason to believe that this public money is being irresponsibly spent. However, in the absence of specific, measurable goals for eliminating our affordable housing crisis it’s hard for me to see how we can even begin to evaluate whether the current strategy is the best way to spend tens of millions of dollars.

The city and county of Santa Cruz have allocated public money to lots of different projects, under lots of different authorizations, with different (but complimentary objectives). To their great credit they have a system for reporting progress on many of these various strategies/actions/projects which can be viewed here:

http://www.cityofsantacruz.com/home/showdocument?id=53604

You can refer specifically to page 21, Table AP 23 to see the total amount of public monies allocated to the goal of ‘increasing and preserving affordable housing.’ Progress is reported by several ‘Goal Outcome Indicators’ for example:

• Homeowner housing added
• Rental units constructed
• Tennant-based rental assistance/rapid rehousing
• Homelessness prevention

While I applaud the city for quantifying their progress, what I find missing from this is any discussion of what kind of impact they EXPECTED from money allocated to these initiatives. For instance, I can use this table to see that the needs addressed were:

Preserve existing rental housing and increase the amount and affordability of rental housing for the City’s lowest income renters.

And I can see that a total of 4.8 million was spent from different program pots (125k from the Community Development Block Grant Program, 125K from HUD’s HOME program, 1.3 million from tax exempt bond from the general fund, 2 million from Low Income Housing Tax Credits).

And I can see that for this 4.8 million there were:

• 50 household housing units constructed under the ‘rental units constructed’ metric,
• 30 households assisted under the ‘tennent-based rental assistance/rapid rehousing’ metric
• Some other stuff

What I can’t see is what they expected to get for this money. Is 50 household housing units a lot or a little? How big a dent in the ‘housing crisis’ do they think the addition of 50 low income rental units makes?

To be clear, I’m not saying it didn’t make a dent. I’m saying, numbers are great but these numbers lack context. They’re just arbitrary numbers. Without knowing how many units we were hoping to get for the money, it’s hard to know if the 4.8 million spent on this line item would have been better allocated to a different goal or would have been better spent on a different mix of projects.

Responsible use of public money: evaluating alternatives

Let’s go back to my list of 4 possible examples of how to define the affordable housing crisis and think about the appropriate way to address Example 1 vs. Example 4.

I would argue that if Example 1: Cops and Nurses strikes you as a good way to think about affordable housing then a possible solution might not be supply-side (adjusting the stock) related at all. If we really want cops and firemen and nurses to be homeowners in our communities we could think about setting up a home-purchase program for them (we could also just give them a raise but then they might take thier extra income, buy a cheap house in the next county over and our plan to have them living in the community would be defeated). Santa Cruz county currently has around 50,000 homeowners and a property tax rate of about 1.3%. If we increased everybody’s property tax around 100 bucks we’d have 500,000/year fund that could be used for downpayment assistance or some other incentive for cops, nurses, and firefighters to purchase houses with.

If Example 4: Section 8 Waiting List Decrease is more consistent with your notion of our affordable housing problem then a supply-side strategy might be the only one that makes sense. However, it should be realized that increasing the affordable/subsidized housing stock in order to move people off the Section 8 waitlist may incentivize low income residents from neighboring counties to apply for Section 8 vouchers in Santa Cruz (they may be subject to porting restrictions but those are generally 1 year waiting periods or less). So if we want one of our goals to be to reduce the size of the Section 8 waiting list, we need to at least consider that doing so may cause in-migration of low income families.

To be clear, the two things aren’t mutually exclusive. A good comprehensive housing policy might want to consider both goals of

• increasing homeownership rates among cops, nurses, and firefighters to 30% by 2035, and
• cutting the Section 8 voucher waiting list in half by 2020

which is exactly my point. If we articulate these two clear goals, then we can start talking about the best actions (in terms of bang-for-the-buck) to pursue these goals. If we just start talking about ‘creative solutions’ and ‘thinking outside the box’ before we have clear goals in mind, then it becomes almost impossible to evaluate any two competing uses of public money.

Summary

Wrapping Up

Discussion about affordable housing needs to be an honest discussion about tradeoffs. Everything has opporunity cost and most actions have some winners and some losers. Having winners and losers to public policy doesn’t mean we can’t make policy or strategy.

First, increasing the supply of affordable housing necessarily incentivizes in-migration of low income individuals. Where people choose to live is heavily influenced by labor mobility, the geophraphy of jobs, and access to transportation. But it’s also influenced by relative housing costs. If Santa Cruz County becomes more expensive relative to the neighboring counties of Santa Clara and Monterey we will see out-migration of people at the lower quartiles of the income distribution. And if Santa Cruz County becomes less expensive relative to neighboring counties we expect some in-migration of low income people.

Increasing affordable housing stock and adding low income residents puts downward pressure on existing housing values. Again, that’s not a reason not to do it. But we shouldn’t pretend that everybody wins. I value diversity and I’d rather live in a community with a mix of blue collar, grey collar, and white collar families but I’m not going to pretend that building such a community has 0 costs. If we’re not willing to articulate how much affordable housing we want to provide then there is no way to judge what we will be required to give up in order to provide it. If we’re not honest with members of our community about the magnitude of tradeoffs implicit in our housing policy, then we’re not offering everybody the opportunity to make informed choice.

Second, there is a necessary tension between increased development and preservation of greenbelt, coastal access, and other public goods. You can’t have more of both. To their great credit the city and county of Santa Cruz have embraced principals of smart growth and have invested considerable time and effort in growth strategies that minimize impacts to natural resources and amenity values. However, the basic question, How many total people can Santa Cruz County support? still needs to be answered. I realize questions of this flavor are unpopular because answering them necessarily means there are some people or groups of people for whom the answer means, “you don’t get to live here, sorry.”

I’ve been over the Housing Elements, Economic Vitality Report, Regional Housing Needs Allocation, and Santa Cruz Action Plan backwards and forwards multiple times. The words ‘urban infill’ and ‘densification’ are all over every one of those documents. But even after scrutanizing each one of them, you know what I still don’t know? Suppose we infill every square inch of Santa Cruz, how many people does that mean we house? AND:

• how does that number compare to projected growth?
• how does it compare to the number currently on the Section 8 voucher waitlist?

If we infill like a motherfucker, will our ‘housing crisis’ be over?

Looking Forward

The ‘War on Drugs’ that we have be waging for the past 30 years is another example of a high profile public policy problem that is incredibly ill-defined. To fight this war people have leveraged the language of ‘crisis’ to argue that we don’t have time to ‘be academic about things.’ ‘We have to take swift action on all fronts’, we’ve been told. We need an ‘all of the above’ strategy.

The problem with ‘all of the above’ strategies is that if you never articulate which parts of the strategy are supposed to attach which parts of the problem and evaluate what size impact you expect, you never know if you need to change things that aren’t working. When you pursue ‘all of the above’ without clear goals for each part of the strategy then the only recommendation for improvement you’ll ever get is ‘spend more money.’

In our War on Drugs we’ve:

• drastically increased spending on Border Patrol Resources
• militarized local police forces and supplied them with military grade weaponry
• used covert operations to facilitate ‘Regime Change’

And the only thing we know for sure that we’ve done is incarcerate an astronomical number of African-American males.

The language of crisis can be pretty powerful in getting one’s point across. It needs to be deployed responsibly.

Appendix: A Final Rant

Sociologists at the Center for Labor Studies at UCSC have been engaged in long-running research project called No Place Like Home: Affordable Housing in Crisis in Santa Cruz: http://www.noplacelikehomeucsc.org/en/. Much like the city, county, and regional planning/action documents I’ve cited here, this appears to be the work of very well intentioned professionals. It also, unfortunately, provides a great illustration of the seedy underbelly of the soft social sciences.

The project has clearly done some important things…like interviewed hundreds of low income residents to better understand the ways housing costs and housing insecurity impact their lives. As a resource for helping develop actual public policy to address affordable housing in Santa Cruz it appears to me to have very limited utility. This is mostly because the project is basically a sprawling expanse of web-links with infographics, interesting summary statistics, and powerful anecdotes. There is no unifying documentation that I can find anywhere (a Google Scholar search on the principal investigators turned up 0 scholarly publications related to the project, there is no .pdf on the website anywhere that has sections titled ‘Methods’ or ‘Study Design’). It’s a collection of interesting facts and perceptions with no clearly defined research questions or testable hypothesis.

I can forgive the lack of scientific integrity from city and county government reports that are generally prepared by snake-oil salesmen called consultants or severely overworked-underpaid municipal employees. It’s harder for me to stomach when it comes from people who should know better.

It’s not that I don’t think this study is valuable, quite the opposite, I think the information obtained could be really useful. The experiences and perceptions of the people surveyed probably could have provided answers to some important questions…but since the study designers appear not to have formulated any actual research questions, the results aren’t really answers. They’re just a semi-organized collection of interesting anecdotes.