5 ms·
I'm not sure if I can agree. For example, an analysis that puts a lower bound on the magnitude of rent decreases is here: https://www.federalreserve.gov/econre
by epistasis 8y ago
I'm not sure if I can agree. For example, an analysis that puts a lower bound on the magnitude of rent decreases is here:
https://www.federalreserve.gov/econres/feds/files/2018035pap.pdf https://www.federalreserve.gov/econres/feds/files/2018035pap...
They perform a simulation where everybody from Monterey to Marin to Stockton is equally happy living anywhere, and only cares about a handful of neighborhood characteristics. They then add a small amount of housing for individual desirable areas, holding everything else constant, and let people move whereever they will; i.e. if a person in Monterey gets to save $5 on rent for a place in the Sunset that's otherwise identical in "desirability" to their Monterey place, they're likely to move.
Even in this outlandish situation, rents decrease quite a bit. For a 5% increase in supply, there's a 0.5% decrease in rents. A 20% increase in supply, there's a 1.8% decrease in rents.
This study is often cited to make the opposite point: that it's impossible build to decrease rent. However, even small, focused amounts of building result in decreases in rents in this model.
Increasing density in the Sunset with lots of apartments where there are single family homes would have a huge impact, that would be potentially a 100% or 200% increase in supply. That would do a ton for the crisis in San Francisco.
- mjevans 8y agoI read the findings you're summarizing differently. Insane prices already indicate that a market is completely defective. Correcting that condition requires, as noted in the figures, ORDERS OF MAGNITUDE difference in either the demand or supply. When rents are that crazy there will be an already displaced mass that isn't seen in the focus area of study; the market needs radical, organized, and planned (at least in the broad strokes) re-classification and re-organization of zones. As a very rough back of the napkin approach, which an actual expert would probably have better initial numbers for, examine the quantity of jobs in the area. To approximate for youth, students, retirees, and the unemployed; double that figure. (Est) NUM_RESIDENTS ~= 2 * JOBS Then assume by the time you actually get things fixed the problem will have gotten even worse. So target the overshoot at an additional 25 to 50 percent. Also, require distribution of housing VERY ROUGHLY according to actual census data of demographics. Just based on the cost of rent I'd say NO PLACE builds enough housing for singles nor enough for 'large families' (not in houses).
- epistasis 8y agoYes, this model is not super informative, but I think it does put a lower bound. Also, I'm not sure which figures you're refereeing to for "order of magnitude", but recall that their "percentage of supply" change is a percentage of a PUMA, a mere 100,000 person district, not of the whole area. So even holding everything constant in the Bay, and assuming people will move places they would actually never move, a mere 5,000 housing units in one PUMA still makes a difference. Totally agreed about the need to balance zoning more.
- dragonwriter 8y ago> For example, an analysis that puts a lower bound on the magnitude of rent decreases is here: It uses a “the greater bay area is a closed universe” assumption, which means it necessarily excludes the effect of any supply increase on increasing people moving into the area, mitigating price decreases. So, it cannot reasonably put a lower bound on rent decreases. At best, it provides and upper bound on such decreases.