The horse, as they say, is out of the barn. So are the pigs, and the chickens, and all the communists pushing for a tax on jobs in Seattle. The Seattle Times and Crosscut have posted stories about a very poorly argued post about a study by McKinsey Company about the City needing $400 million a year to solve homelessness. It is a study on confusing correlation with causation. This is a mortal sin for anyone who writes anything and cares about making good arguments. For McKinsey it is truly an intellectual embarrassment; blaring out the “findings” in a headline is too, and it is compounding the problem and confusion. Jobs don’t cause homelessness, bad housing policy does. This is my email to the writers and the Times and Crosscut. 

Hi David and Vianna,

The guys at McKinsey are smart.

They know the math.

All I did was study philosophy and religion. And you know what, I moved to American religion from Buddhism as a speciality in graduate school. You know why? Hindi and Sanskrit were going to be too hard.

The only “math” I took after high school was something called Logic and Language. I barely got out of there with a C. It hurt my head.

But even I know the difference between correlation and causation. The article you reference in your article is an embarrassment for the authors and by extension your work.

Their graph uses the word “correlation” when it looks at the rise in rents and homelessness, then in this paragraph jumps over another canyon of logic:

In King County, homelessness has risen in line with the fair-market rent (FMR), which has in turn increased in line with the county’s strong economic growth, propelled by the swelling ranks of high-income digital workers. On a single winter night in 2017, volunteers counted 11,643 homeless people, an annual average rise of 9.2 percent since 2014. Over the same period, the FMR has risen an average of 12.3 percent a year.

I love the “winter night” reference for added drama.

Homelessness has increased, they say, because of strong economic growth, which has created high paying jobs. Therefore all that suffering is because of “swelling ranks of high-income digital workers.”

Where is the causation here? How many “high paying jobs?” How many homeless people are homeless because of increases in “fair market rent,” a very specific definition used by HUD?

Then there is a non-sequtor: we need $400 million per year. It just isn’t clear what such a huge amount of money would get us. Is it even possible to build 7,800 units of subsidized housing in our City over 5 years? I’ve looked at annual production numbers of subsidized housing and that number in its peak year was only about 450 units of housing (see attached based on OFM data and numbers from the City).

Never in the last decade have non-profits produced that thousands of units of housing and it wasn’t for lack of funds as much as it is that it takes a long time to acquire land, complete financing, permitting, and construction. But costs are also rising. The average cost for non-profit subsidized housing rose more than 40 percent from about $224,000 per unit to about $324,000 between 2008 and 2016. Those costs are even higher today and rising.

As I have pointed out again and again and again (see one example below) all the money in the world will not overcome the scarcity of land and labor and the rising costs of capital construction. And the number of 244,000 “affordable homes” by 2040 is bizarre. There are, if my memory serves me, about 14,000 subsidized units in the city; that means producing over 10,000 a year over that time period. The entire housing market produced just 39,000 units since 2010 (again OFM data) about 5,500 units per year.

If we were simply to buy down all the cost burden in the city it would only be around $100 million by my rough estimates. Why are we building a few hundred units for lots of money that will be available years from now if people are living here now but a few hundred dollars short on rent?

We’re literally hours away from a Council vote on a disastrous measure being enacted; and it is a disaster not even just because of the loss of jobs it will create, but because it will deliver so little benefit.

We need our journalists to dig deeper into these numbers.

Your stories have handed a very big win to the pro side of this argument without looking more closely at these numbers and asking harder questions about whether or not Mckinsey can prove that jobs are the cause of homelessness.