I had to take a break from writing for a week. Illness. I’ll leave it at that. But I’ve been keeping up with things online, and decided I had to deviate from my usual topics to cover a pressing topic of the day – San Francisco Real Estate. I feel like I am about to put my hand into the hornet’s nest, but here goes.

The big theme online has been that San Francisco is being ruined by the tech companies who are driving up rents by paying their employees so well. I do not have to link to anything, you have seen the stories, or you can just Google “Google Bus” and you will get the idea.

I find the whole thing a bit baffling. While I understand that rising income inequality is a real problem in the US, I do not quite understand why Google, let alone individual Google employees are to blame. (And you can substitute any big tech company name for Google.)

Let’s start with a basic question. Would you rather live in a region where talented, driven people want to move or one where talented, driven people have all left long ago? I think most people would prefer to live in thriving, growing places, all else being equal.

And yes, the Google buses are too big for the streets they drive on, but I am pretty sure Google Inc. would be much happier if the US had a better public transit system.

The  real problem is housing. There is no argument that Bay Area real estate is expensive and has gotten much more expensive in recent years. I think it is impossible to buy anything in San Francisco for less than $500,000, which is something like 3.5x the national median. And rents are ridiculous too.

Can we really blame all the tech companies for that? True, they are hiring a lot of people and paying them really well, but above we agreed that is a good thing. So rather than look at the demand side of the housing equation, maybe we should look at the supply side. Prices go up when lots of people move in, and economics 101 tells us that will drive up short-term prices, but that eventually more supply comes on the market and prices subside. That has not happened.

I have a theory driven by personal experience to explain this.( I have data too, which I will get to in a minute. )

<rant>

I own a house in San Francisco. My house has an unfinished ground floor, with enough space next to the garage to fit out a one bedroom apartment. I could rent that space out for something like $3,000 a month, and would probably have a line out the door of people wanting to rent it. A lot of my neighbors have similar space. The problem is that I need to do some work on the house to fit that space out. I spoke to an architect recently and he told me it would probably cost $30,000 just to do the permitting. Did I mention that my house is now in an “Historic District”? Make that $50,000.  In almost any other city, I could remodel most of the house for that much money, but in San Francisco, I am just getting started.  That discourages small projects, and since my house is 100+ years old, I would have to do a bunch of other work just to fit out a garden apartment. Even at crazy San Francisco rents, it would probably take ten years to pay back the costs.

</rant>

Now for the data which I pulled from the Census Bureau

In the ten years to 2012, San Francisco permitted 1,120 new buildings or 19, 173 new units of housing. In 2012, the Country issued only 79 new building permits. New housing starts are not the best metric for measuring urban housing, but I think it should be clear that the remodeling market is not any more robust. From 1994 to 2004, the Country averaged 244 new buildings a year, but between 2002 and 2012 the average was only 119, less than half in a period that included a massive building boom everywhere else.

To be fair, I compared that to US and California data. And on these metrics, it turns out that San Francisco has actually been in line with or slightly ahead the State and the Country.  I compared new building and new unit starts from 2012 back to 1994. And over that period, San Francisco mirrored other places. And on new units, San Francisco actually greatly outpaced everywhere in new construction. So the City actually added a fair number of new units.This surprised me, because it really does not feel like there is much new housing on the market.

Then I found the data on construction costs. Full caveat, I still need to dig a bit to understand how the census bureau calculates this metric, but I used the same survey to compare San Francisco and California data. And the contrast is telling.

In 1994, San Francisco built 200 new buildings at a total cost of $122 million. In 2012, it built 79 new buildings for a cost of $918 million. The construction price per building went from $611,110 to $11,631,310.  By contrast, costs for the state of California went from $148,248 to $430,243. So costs grew at a compound annual growth rate of 18% for San Francisco, and 6% for the State.

Set aside growth rates for a moment, and just think about those figures in absolute dollars. On average, it costs over $11 million to build a building in San Francisco. Permits, labor, and materials all cost a lot more here. To get an 8% return on that construction, it would need to generate $930,505 a year in rent. For a ten unit building, that works out to $7,754 a month in rent.

Put simply, the only way to build a new building in San Francisco is to build expensive apartments. So expensive that in most other cities we would call them luxury apartments, but in San Francisco they are likely to be studio lofts or cookie cutter one-bedrooms at 900 square feet (~90 m^2).

So we can blame Google for high rents in San Francisco, or we can look for some way to lower construction and renovation costs. My guess is that changing those things in San Francisco is so hard it may just be simpler to blame Google.