Chris Elmendorf Profile picture
Dec 30, 2025 22 tweets 4 min read Read on X
"For a typical mid-rise apartment in San José, construction costs can exceed $700k–$900k per unit."

I 💯% agree w/ @MattMahanSJ that reducing construction costs should be a top priority for 2026 -- and that this is mainly a job for the state legislature.

🧵/22
Reason #1. CA's fiscal constitution + local political incentives push local govs to extract "value" from development w/ impact fees, IZ & transfer taxes.

This drives up the cost of building enormously.

/2
The state leg should preempt most such fees, IZ, & taxes, ***and create a substitute source of local revenue.***

My preferred alternative: a state parcel tax assessed on the "net potential square feet" or "net potential units" created by upzoning pursuant to state law.

/3
Cities would be given a local option to reduce but not increase the tax.

/4
This would be functionally similar to a land-value tax assessed on the increment of value creating by state law, but it would sidestep Prop. 13 (b/c it's not assessed on value) as well as referendum-vote requirements for local taxes (b/c it's a state tax).

/5
The leg would need to pass it with a 2/3 vote. (Presumably the leg would exempt owners of SFHs.)

The tax would answer groups that insist on "value capture," while giving cities the revenue they need for infrastructure.

/6
Reason #2: The state has passed or strengthened numerous laws in recent years that both create development potential AND make it costly to build (AB 2011, SB 79, SB 423, housing element law).

/7
The costs are imposed through requirements that developers pay way-above-market wages, and/or set aside a substantial share of the new units as money-losing and hard-to-manage "deed-redistricted affordable housing."

/8
Happily, the Leg in 2025 came up with a much more economically feasible labor compromise for the CEQA reforms.

In 2026, the Leg should roll that compromise into AB 2011 and SB 423.

/9
As for the BMR mandates in state law: the Leg should repeal them all, and instead use a portion of the new parcel tax revenue to purchase units or deed restrictions via a reverse Vickrey's auction.

/10
The state would establish a formula for the housing services a dwelling provides (taking into account location, size, etc.), and developers would compete through the auction to sell housing services to the state at the lowest cost.

/11
Or, in lieu of buying BMR units, the state could use its share of parcel tax revenue to supplement the federal Housing Choice Voucher program.

/12
The voucher approach is probably superior to buying BMR units (not dependent on a housing-services formula, better incentives for building maintenance), but it would be a bigger conceptual break with the "let's make more BMR units" tradition and so might be harder to enact.

/13
Reason #3: Construction costs are determined in significant part by state building codes.

CA passed a stopgap "no new code requirements for 5 years" measure in 2025.

But freezing the code doesn't do anything to fix existing rules that wouldn't pass a cost-ben test.

/14
The Leg should revisit how state code requirements (and local code amendments) are made, requiring cost-benefit analysis.

Then audit the existing code and strip away rules that don't pass the test.

/15
(In 2024, @CAgovernor directed @California_HCD to identify code reforms to lower cost of building infill housing. Has anything come of this? Not to my knowledge. I'm not sure HCD has the capacity to do it--no builders or economists on staff.)



/16
Reason #5: Private incentives to invest in construction R&D are increasing in firm/project size.

No one city can do much about this, but the state has leverage.

/17

nber.org/papers/w33188
E.g., by upzoning & requiring ministerial review, it reduces incentive for firms to invest in local politics rather than R&D.

And by establishing a state permitting regime for large projects, it could spur construction-tech agglomerations.

/18

latimes.com/opinion/story/…
The state could also offer monetary prizes for new production processes or technologies that demonstrably reduce the cost of construction.

/19
Reason #6: the "post-construction" costs of a housing project are determined in significant part through state law. E.g.,
- construction defect liability
- fire insurance
- energy costs

/20
And the big local drivers of post-construction cost & risk, like new tenant protections and transfer taxes, are things that the state could control and standardize.

(The politics of doing so are probably easier at the state than at the local level.)

/21
In sum, it's great that local leaders like @MattMahanSJ and @DanielLurie are trying to reduce the cost of of building, and there's much they can accomplish, but the biggest bangs will have to be delivered through state law.

/end

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Chris Elmendorf

Chris Elmendorf Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @CSElmendorf

Dec 31, 2025
New decision from CA Court of Appeal on the fee-shifting provisions of AB 1633 has big implications for NIMBYs' incentive to challenge housing approvals under CEQA & beyond.

This one belongs in a Law of Abundance casebook.

🧵/24
law.justia.com/cases/californ…Image
Context: As part of the 1970s revolution in admin law, states & the federal gov't actively encouraged self-appointed "private attorneys general" to sue, via attorneys' fee bounties.

/2
Asymmetric fee-shifting provisions were written into scores of public laws: If a plaintiff challenging a gov't decision wins, the gov't has to pay for the plaintiff's attorney; if the plaintiff loses, they don't have to pay for the gov's attorney.

/3 Image
Read 25 tweets
Dec 28, 2025
Could L.A. really land in the Builder's Remedy penalty box, just for f'ing around with a single low-income housing project which a nonprofit developer wants to build on city-owned land?

Yes.

A quick explainer🧵.
In October, @California_HCD sent L.A. a sharply worded letter, warning that the city's housing element had relied on the Venice Dell project both as a "pipeline project" and as part of the city's strategy to "affirmatively further fair housing."

/2

hcd.ca.gov/sites/default/…Image
Image
The HCD letter also flagged five "policies" and two "programs" in L.A.'s housing element that per HCD should "facilitate the project."

The city's course of action has been "inconsistent with these policies."

/3 Image
Image
Image
Read 11 tweets
Dec 20, 2025
Cooking in San Diego: A turquoise, 23-story test of the Permit Streamlining Act's new-and-improved "deemed approved" proviso.

This could turn into a big constitutional battle.

🧵/22 Image
Image
Enacted in 1977, the PSA put time limits on CEQA and other agency reviews of development proposals.

If an agency violated the time limits, the project was to be "deemed approved" by operation of law. Wow!

It proved wholly ineffectual.

/2
As @TDuncheon & I explained, courts first decided that the Leg couldn't possibly have meant for a project to be approved before enviro review was complete.

Ergo, CEQA review must be finalized before the deemed-approval clock starts ticking.

/3

papers.ssrn.com/sol3/papers.cf…
Read 23 tweets
Dec 5, 2025
In the topsy-turvy world of CA Density Bonus Law:

- San Francisco almost certainly must approve this 25-story project on a site zoned for 4 stories

- The city's new ordinance deregulating density in "well-resourced areas" will operate as de-facto downzoning of such sites

🧵 Image
This project's site is zoned for retail use and is currently occupied by the Marina Safeway.

The zoning classification also allows residential use at density of 1 unit per 600 sqft of lot area or density of nearest residential district, whichever is greater.

/2 Image
The nearest residential district, RM-4, allows density of 1 unit per 200 sqft of lot area.

That translates into 567 units on site.

Developer proposes to build 790 units, which requires a 39% density bonus (790/567 = 1.39).

/3 Image
Read 21 tweets
Nov 19, 2025
Bharat's substack response ⤵️ to my thread about his & @nealemahoney's op-ed has brightened my day.

So refreshing compared to the snarks (and vivid expressions of desire for my assassination) conveyed on this platform.

A few notes on possible paths forward.

🧵/13
By describing the credible commitment problem (the need to reassure developers of new housing or energy that their project won't face price controls for a very long time) I didn't mean to imply, as some critics on the right insist, that the problem is insurmountable.

/2 Image
I think the problem can be greatly mitigated:

1. By offering DC-style "certificates of assurance" to developers, i.e., recordable contracts for compensation if the project is subjected to price controls within a defined period of time.

/3

Read 14 tweets
Nov 16, 2025
I have great respect for @nealemahoney & @BharatRamamurti, but I just about pulled my hair out reading their op-ed this morning.

Price controls aren't going to be "a way out" unless their advocates can credibly commit not to apply them to today's projects tomorrow.

🧵/12 Image
The authors briefly acknowledge this concern at the end of their piece but offer nothing beyond a brief nod to sunset clauses and income targeting.

/2 Image
They fail to acknowledge that the NYC controls that Mamdani campaigned on strengthening (w/o income targeting...) have been in place for 50+ years; that popularity of rent controls surely depends on them *not* being income targeted;

/3
Read 13 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(