#SocialHousing

Hey #Hashtaggers, hope all dozen of you loyal readers are doing well! I've got another hopefully spicy #CentristDataMining post for you guys. This one's a bit different since it's not just me analyzing someone else's policy, but it's me actually coming up with my own idea, which hopefully you all will find worth talking to your city councilmembers about. With that intro, let's dive in! 

This idea came when Matt Bruenig, a progressive policy researcher, was hyping up a policy paper his think tank authored which got the new "PHIMBY" (public housing in my back yard (wordy – I know)) activists really excited. You can read more details in his 40 page paper [PDF warning]. The tl;dr of the paper is that local governments should take out municipal bonds to build housing, and pay back the interest of those bonds from the rents tenants pay on the new projects. But if you read the paper, their analysis stops short of any actual pro-formas or analysis on the feasibility of the proposal in a real-world type scenario. So, I decided to extend the idea. 

What if the city instead buys apartment buildings on the open market, and rent them out to pay back the municipal bonds? 

First, we need to dive in the fun, wild world of Washington State municipal bonds. Old #Hashtag favorite MRSC comes to the rescue here. They describe the three kinds of bonds city governments can issue  - which you can read about in the link. But the one I want to turn your attention to is what's called a "revenue bond". MRSC describes revenue bonds as "bonds [that] may be issued to finance projects for any enterprise that is self-supporting". This is exactly what an apartment building is. And if you read the RCW statute 35.41.030, which authorizes local governments like cities to issue these bonds, you can see that acquiring buildings is one of the valid use cases of these bonds. The only restriction is that the project the bond funds "provide[s] funds for defraying all or a portion of the cost" of the bond itself – so keep that in mind for later. 

As the MRSC mentions, these revenue bonds are usually issued to fund things like sewer and electrical improvements. It turns out that the City of Seattle does issue these types of bonds on the open market, so we can tell how much interest these bonds generate to continue our analysis. The bonds that are on the market all have a 5% coupon, which means that a $1,000 requires annual payments of $50 a year. We'll have to make sure that the rents generated by the buildings the city is to purchase are high enough to pay back that 5% coupon. 

Now let's look at apartment buildings actually for sale today on the market! From some light googling, I found the website loopnet.com, which lists apartment buildings for sale (RIP your facebook targeted ads from now on), and you can see there are a bunch of buildings of all sizes and ages up on that website.  

Let's look at the traditional #CentristDataExcelSheet, which lays out all these numbers. 

Let me explain this a little bit: the number of units and the price of the building come from the apartment listing website. The Price Per Unit column is simply a division of the price by the number of units, just to give you a sense of how much these places are in comparison with one another. The "Average Rent Needed W/o Property Taxes" is a calculation of the average monthly rent each unit would have to pay to pay back 5% of the asking price every year (in reality, of course, this number would be proportional to the size of the unit, or some other fair calculation, but for simplicity's sake I'm skipping that), and the column "Average Rent Needed W/ Property Taxes" adds in the property taxes borne by the building since I'm guessing the city would still want to collect property taxes from these buildings. The next column "Average Rent + 10% Overhead" assumes the city would want a 10% buffer on the average rent to cover things like rent non-payment, routine maintenance and professional management, and the like. 10% is an arbitrary number I made up, so feel free to download the Excel sheet and play around with the numbers! Lastly, the two link columns at the end provide you a source for my data so you can fact check.

The last two rows which I titled Hypothetical (New) Quadplex and Apartment are performing the same analysis as buying a building but imagining that the city would build a new place with these revenue bonds. I made up the total cost of the quadplex, but the apartment is if the city built another El Centro De La Raza with the same cost as reported by its developers. 

Now, the city doesn't have to charge every tenant the same. It could set aside some units to pay way below market rate, and some above the average rent needed to pay back the bond. It just depends what needs the city wants to fill and how they want to achieve its goals. Just for this analysis, I'm ignoring anything like that, as well as any kind of analysis of the types of units each place has to figure out how much a 1 bedroom unit would have to pay, vs. a 2-, 3-, and 4-bedroom place.

But I hope I've shown you in this article that this plan is a) legal and b) feasible. If you think of any extensions to this scheme, please do let me know on twitter, or if you catch any errors.  

And if you do want to start pushing for this idea to happen in this city, give a guy a little credit! Tell your city councilmember to read the #HASHTAG! Let's start the #CENTRISTDATAMINERREVOLUTION! Until next time, #Hashtaggers!