The ordinance passed. Before it passed there was a report, and this desk has read the report, which is the part nobody does.

It is ninety-one pages. It is good.

That is not a set-up. The methodology section is explicit about its sources and about their limits. The baseline is built from the right data and the report says where the data is thin. There is a five-year projection with a stated confidence interval, which is more than most such documents attempt, and a sensitivity analysis that runs the projection against interest rates and against construction costs at three levels each.

Those are the two variables the field agrees are the variables. The report ran them properly. Somebody spent real months on this and it went through committee and the committee asked reasonable questions and got answers.

Wage growth is held at trend for the projection period. This is the standard convention. Nearly every study of this kind does it and it is defensible, because the alternative is to assert a wage path, and asserting a wage path is how studies get discredited.

This desk searched the document for a term and did not find it.

Not once in ninety-one pages. Not in the methodology, not in the limitations section, which is two full pages and unusually candid, not in the appendices, not in the footnotes, not in the list of factors considered and set aside.

That last one is the finding, and this desk wants to be exact about it, because the obvious reading is wrong.

A report that considered something and rejected it says so. That is what a limitations section is for. This report's limitations section is careful and it names several things it could not model — it names them by name, explains why, and moves on. Whoever wrote it was not hiding from difficulty.

So the term is not absent because it was weighed and dismissed. There is no record of a weighing. It is absent because it was never a candidate for inclusion, and a thing that was never a candidate leaves no trace anywhere in a document, including in the part of the document that exists to record what was left out.

We say again that nobody did anything wrong. There is no scandal here. Nobody suppressed a section. The convention was followed and the convention has no slot for this.

One further note and then this desk will stop.

There is a published model, with a DOI, holding that every ten percentage points of aggregate wage compression corresponds to five to six additional percentage points of severely cost-burdened renters, against a baseline of about a quarter of them.

This desk does not assert that the model is right. It has not been through the process that would make anyone confident in it, and it may be wrong by a lot.

What this desk observes is narrower. The report holds wage growth at trend for five years. If wages do something other than trend, for the reason that is not in the report, the report does not become wrong — it has no mechanism for becoming wrong about that, because the variable is not in it. The projection would simply be about a city that did not turn up.

The ordinance is in effect. The report is on file. It is ninety-one pages and it is good and this desk recommends reading it.

Never Learns covers assessment and reads every study the council commissions. It has not yet found one that contains the term.