Chris Elmendorf—UC Davis law professor, prominent Yimby enabler, and de facto Chronicle staff columnist—is a scourge of economic illiteracy. Usually he trains his contempt on “folk economics” —what he and his colleagues call the economics of “a mass public befuddled by the relationship between housing supply and prices.”
In an October 30 op-ed for the Chronicle, Elmendorf cast a withering eye on a new target: city planners—specifically the staff of the San Francisco Planning Department. For evidence of their cluelessness, he cited the “Family Zoning Plan: Economic Impact Report” released on October 29 and authored by SF City Economist Ted Egan.
The report shows that San Francisco will not meet the state’s demand that the city zone to “produce”—both Egan and Elmendorf use that term—82,000 homes by 2031. Instead, Egan found that under the best-scenario/high-growth forecast, the upzoning mandated by Lurie’s proposed plan is likely to generate only 14,646 additional homes by 2045.
Elmendorf warned that by next February, the shortfall could trigger the dreaded Builder’s Remedy, which gives developers wide leeway to build whatever they want.
The basic problem, he argued, is that the models behind the Family Zoning Plan and the state’s own housing framework were devised by planners, which is to say, “crafted without economic expertise…. [T]here is not a single staff economist at the state’s housing agency. Nor does the state Legislature have economists vet housing bills.” The upshot: “the state tells cities to make realistic plans but doesn’t furnish reasonable modeling tools that they may use to evaluate their plans’ sufficiency.”
According to Elmendorf, this situation is old news. Linking to a 2017 article by LA Times reporter Liam Dillon, he wrote:
For more than 50 years, California has been telling cities to plan for regionally needed housing. Over the same period, California home prices and rents have gone from a little expensive to wildly unaffordable. The state’s planning framework was meant to fix this problem, but it hasn’t gotten the job done.
Now, “by reasonably projecting the likely impact of the Family Zoning plan, the city’s chief economist has cut through 50 years of fakery in state housing law.”
In an X thread posted on the same day as his op-ed, Elmendorf absolved the California Department of Housing and Community Development (HCD) from blame. “They’re doing the job the Legislature gave them, with the staff the Legislature gave them. (No economists…).”
Help us save local journalism!
Every tax-deductible donation helps us grow to cover the issues that mean the most to our community. Become a 48 Hills Hero and support the only daily progressive news source in the Bay Area.
He called on the state to provide
realistic, economically-informed targets. Even more important, they need realistic, economically-informed tools with which they can quickly & easily gauge a rezoning plan’s sufficiency….It makes no sense to tell city planners to devise their own models, as the planners have neither the technical knowledge nor the political incentives to come up with something reasonable. End the guessing games. Please.
The first step is to “take a deep breath, step back, and contemplate the craziness.”
The real craziness: Wiener’s SB 828
The first California law that required cities (and counties) “to produce prodigious reports to plan for housing” was not, as reporter Dillon wrote, a 1967 statute but rather AB 2853, which was signed into law by Governor Jerry Brown in 1980. The 1967 law—actually twinned bills SB 1401 and AB 1952—only directed local jurisdictions to “endeavor to make adequate provision for the housing needs of all economic segments of the community” (emphasis added). It said nothing about satisfying regional housing needs. The regional fair share concept was introduced in HCD’s 1976 draft revision of the state Housing Element Guidelines. In 1976, too, the California Supreme Court ruled that zoning ordinances need to accommodate regional housing needs.
The basic framework of the Regional Housing Needs Allocations was established by AB 2853. Contrary to Elmendorf’s claim, the state did not intend that framework “to fix” the housing affordability crisis. Nor did it penalize cities if the amount of housing built within their boundaries fell short of their Regional Housing Needs Allocation (RHNA—sounds like ree-nuh).
Indeed, AB 2853 stated:
It is recognized that the total housing needs…may exceed available resources and the community’s ability to satisfy this need…. Under these circumstances, the quantified objectives need not be identical to the identified existing housing needs, but should establish the maximum number of housing units that can be constructed, rehabilitated, and conserved over a five-year time frame. [California Government Code, Section 65583(b)(2)]
This was a major concession to both home rule and reality. It acknowledged that planning for housing and producing it are different things. Accordingly, the state qualified its expectation that housing production would equal each jurisdiction’s RHNA.
That qualification was eliminated in 2018 by Wiener’s SB 828. Besides absurdly inflating the RHNAs (for a rundown of Wiener’s legislative antics, see Michael Barnes’ primer), SB 828 erased the distinction between planning for housing and producing it, by amending the passage cited above so that it reads:
It is the intent of the Legislature that cities, counties, and cities and counties should undertake all necessary actions to encourage, promote, and facilitate the development of housing to accommodate the entire regional housing need, and reasonable actions should be taken by local and regional governments to ensure that future housing production meet, at a minimum, the regional housing need established for planning purposes. [California Government Code, Section 65584(a)(2)]
The decisive term is “meet.” To “encourage, promote, and facilitate the development of housing to accommodate the entire regional need” allowed for some wiggle room between a city’s RHNA and the actual amount of development. To “ensure that future housing production meet, at a minimum the regional housing need established for planning purposes” removed that latitude. SB 828 encoded in California the assumption that planning determines production.
This legalistic approach may appear to be at odds with Yimbyism, given its core demand to remove legal barriers to development. In fact, Yimbys love regulation, as long as it preempts local land use agency on behalf of the private real estate industry. Witness the contents of the dense web of recent Yimby-endorsed California housing laws. SB 828 is a mainstay of that web. So is the 2017 bill that underlies Elmendorf’s plea for “realistic tools”: AB 1397.
Urban planning as real estate speculation: AB 1397
The requirement that the land that cities designate to meet their RHNAs must “have realistic and demonstrated potential for redevelopment during the planning period to meet the locality’s housing need for a designated income level” was introduced into California law by Evan Low’s AB 1397.
HCD’s guidance on the “analysis of sites and zoning” under AB 1397 says:
When establishing realistic unit capacity calculations, the jurisdiction must consider existing development trends of existing or approved residential developments at a similar affordability level in that jurisdiction, as well as the cumulative impact of standards such as maximum lot coverage, height, open space, parking, and FARs. The capacity methodology must be adjusted to account for any limitation as a result of availability and accessibility of sufficient water, sewer, and dry utilities. For non-residential zoned sites (i.e. mixed-use areas or commercial sites that allow residential development), the capacity methodology must account for the likelihood of residential development on these sites.
In short, AB 1397 mandated that California’s fair share law be driven by real estate speculation.
According to Elmendorf, the trouble with the model that San Francisco’s planners used to analyze Lurie’s Family Zoning scheme is that it doesn’t upzone enough sites that would realistically be developed for housing to meet the city’s RHNA. The city’s planners, he wrote on X, used
magical thinking about [the] “pipeline” of entitled but mostly stalled, b/c infeasible or needing massive infrastructure) projects to whittle the 82,000-unit target down to 36,0900 units that would be accommodated w/upzoning….Their target was suspect & definitely not reviewed by economists.
Worse yet, the planners had an economically realistic model that they could have used but didn’t.
Elmendorf’s op-ed links to an April 2024 Chronicle op-ed by Yimbys Salim Damerdji and David Broockman, self-described “housing wonks with graduate degrees in statistics and social science.” (Damerdji is an independent researcher; Broockman teaches political science at UC Berkeley.) The two authors tell how, using a model that the SF Board of Supervisors had approved in 2023, they determined that “the vast majority of addresses [rezoned] for taller buildings will never be they developed into new housing… simply because many existing owners won’t to develop their properties, often since it won’t make financial sense.” With “the city’s tools and with generous feedback from Planning Department staff,” Damerdji and Broockman “created a web app that “lets you build your own compliant rezoning plan for 36,000 new homes.”
You have to click on “About” to see their caveats:
[T]his model makes a number of key assumptions: that the economic environment looks like 2016, that statistical associations of the past will hold for hypothetical rezonings, and that projects will be built quite close to their maximum feasible capacity.
Mindful of the precariousness of these assumptions, Damerdji and Broockman recommend that “[r]ather than using this model’s output as a single source of truth,” people should “evaluat[e] a rezoning proposal against multiple models to ensure that the findings are robust to a range of different assumptions.” They do not suggest alternates to the assumptions that inform their app.
Elmendorf has no such qualms. As he tells it, the city’s planners “had a statistical model on hand that could do the job. It estimated sites’ probability of development based on characteristics such as parcel size and the type of existing structure.” Unfortunately, “when it came to assessing the Family Zoning plan, [they] left this model on the shelf.”
His Chron op-ed also links to a January 2025 paper he wrote for the Mercatus Center at George Mason University, “Considerations for State ‘Fair Share’ Housing Frameworks: Reorienting the planning process so that outcomes take the center stage.” This 21-page “special study” attacks subsidized, deed-restricted affordable housing, backs the expansion of market-rate housing, and touts builder’s remedy and other sanctions on “NIMBY jurisdictions” that fail to comply with fair-share laws.
I focus on the paper’s elaboration of the twofold argument that Elmendorf made in the op-ed and X threads cited above:
- housing policy must be judged by “outcomes”—that is, production
- to produce ample housing for households at all income levels, planning must be guided by “economic knowledge.”
From the paper’s abstract: “While fair-share frameworks hold the potential to mitigate housing shortages, they have historically placed too much emphasis on planning and not enough on performance”—that is, “expected yield in new units.” It follows that the “first order of business for lawmakers considering a fair-share law is to set the targets, i.e., the outcomes that local governments’ housing plans are supposed to achieve.”
Elmendorf commended San Francisco’s 2023 “expected-yield” approach but said “[t]here is no right formula for setting housing targets…[I]f state lawmakers want their fair-share programs to increase the supply of housing in high-demand, supply-constrained markets,” they need to devise “a formula that yields targets that represent a substantial increase in production over the status quo” and are still politically feasible.
The “best approach,” he suggested,
would be for states to base targets on a committee of economists’ rough judgment of economic feasibility. The feasibility determination could be grounded on rates of housing-stock growth achieved by fast-growing metros elsewhere in the nation, or estimates of the number of units that would have been built in the absence of local land-use restrictions, or estimates of the number of units that would be feasible to build today in the absence of local land-use restrictions, given prices, and construction costs.
In any case, “one should not expect actual housing production under a housing plan that’s been evaluated for expected yield to equal the expected yield.”
That proviso didn’t make it into his critique of the model that San Francisco planners used to assess Lurie’s Family Zoning scheme, presumably because their numbers were so far off the city’s 82,000 RHNA target. “The point of the expected-yield approach,” the Mercatus paper explains, “is not to hit the bullseye every time, just to hit it on average.” Citing economists Edward Glaeser and Joseph Gyourko, Elmendorf attributed the unavoidable misses to the cyclical nature of housing production.
In planning periods that coincide with economic booms, actual production should exceed a plan’s expected yield under normal conditions, whereas in planning periods that coincide with a recession or an interest-rate spike, actual production will typically fall short.
Precision matters less than the targets’ disciplinary effects. “[E]ven if the expected-yield estimates aren’t quite right on average, they at least put pressure on cities to acknowledge and wrestle with the various ways in which local rules, regulations, and existing uses together constrain development.”
What is housing supply?
Elmendorf’s repeated allusions to “economists” suggest that professionals in the field are of one mind, and that mind accords with his own. That is debatable.
Consider, for example, how economists Tim Helm and Cameron Murray parsed “housing supply” in an October 2024 article published in Housing Policy Debates.
They wrote: “When, why, and how is housing developed in response to growing demand, and how do policies such as land-use regulations affect that? These are questions to which housing researchers may be surprised to find economists can offer no agreed answer.” Having “left a gaping hole where theory about the determinants of supply should be,” economists have opened the way to confusion about “what drives property owners to invest in housing, and whether, when, and how policy settings, particularly tax and land-use regulation, influence the absorption rate” (how fast homes are selling).
To encourage “informed policy discussion,” Helm and Murray stipulated that “in economics, the word supply refers to willingness to sell, expressed as a relationship between price and quantities (graphically, a supply curve)…” Timing is decisive, because
[h]omebuilding is not a short-run production decision. It is an investment decision that responds positively to rising prices. In the ordinary course of events, rising demand and prices trigger new construction: higher prices and higher quantities are seen going hand in hand. In economic terms, of course, this is an increase not in supply but of the quantity supplied in response to an increase in demand.
Accordingly,“[t]he premise that policy changes have effects on markets akin to building additional housing independently of demand is wrong.”
Helm and Murray contrasted the demand-driven meaning of supply with its usage in the “everyday language” of “folk economics,” where “the word ‘supply’ often refers to observed market quantities such as the total number of dwellings, the rental stock, or the rate of construction of new dwelling.”
I asked Murray: “Is willingness to sell equivalent to readiness to build?” He replied:
To be clear, demand and supply have meanings only in relation to trading. Production is a secondary consideration.
What’s notable, then, is that Elmendorf uses “supply” in the way that Helm and Murray associate with folk economics: He equates the term with “expected yield,” that is, with numerical targets, above all with the construction of new homes. Demand has a crucial role in the Helm-Murray account. In Elmendorf’s setup, it’s part of the scenery. Production is primary, and the major factor in determining how much housing gets produced is the strength of local land-use restrictions—“impact fees, inclusionary housing mandates, environmental-study requirements, aesthetic standards, discretionary neighborhood reviews, limits on redevelopment of rent-controlled or tenant-occupied properties, local building code amendments, historic preservation ordinances, and more”—and that’s just a partial list, which he conditions on “given prices and construction costs.” The closest he comes to addressing demand is when he writes about the effects of booms and recessions on production. He repeatedly refers to “high-demand” areas but never delves into the character of the demand in such locales. Compared with the specificity of his regulatory roster, the abstractness is striking.
Why does Elmendorf downplay demand?
Analyzing demand means dealing with class and inequality—an assignment Elmendorf eschews. He shares that inclination with other followers of Harvard economist Edward Glaeser. Commenting in 48 hills on a 2020 article of mine, a regrettably anonymous reader laid out the political implications of the Glaeserites’ treatment of demand:
A great deal of the impetus to building deregulation in the last twenty years came from a flood of papers by Edward Glaeser and Joseph Gyourko. Nearly every pro-building position paper by a governmental agency (including Obama’s White House) quoted them.
There is much obviously sloppy work in their papers, but the most important one to me is a passage from the The Impact of Building Restrictions on Affordability (FRBNY Economic Policy Review, June 2003, p. 28:
As noted, we have decided to ignore the housing demand component of the housing prices. Two reasons underpin this decision. First, housing demand has been studied much more extensively than housing supply. A distinguished literature, including Alonso (1964), Muth (1969), Rosen (1979), and Roback (1982), has considered the determinants of housing demand. Labor market demand and consumption amenities, such as weather and schools, are both important causes of particular demand for some areas. We have little to add to these findings. Second, policy responses to housing prices are unlikely to change housing demand. Increasing supply is a much more natural policy response to high housing prices than is reducing demand.
In other words, they are knowingly ignoring the effect that richer housing seekers have on housing prices, because they think that “reducing demand” (or, more accurately, shifting down the demand curve) is not something that policies could achieve. This would be like investigating the causes of lung cancer but purposefully ignoring the effect of tobacco, because it’s unlikely that people could be convinced to smoke less; or modeling global warming and leaving carbon emissions out of the model, because it would be impractical to get them reduced anyway.
In fact, as you, I, Tim [Redmond], and many others know, a shifting up of the demand curve—that is, more rich people competing for housing—is the one largest reason for the housing crisis. No one is going to price their buildings or rentals for people earning $60,000, if there are enough people earning $200,000 who will outbid them. As long as tech companies pay thousands of people six-figure salaries to come here, it is these people the private market housing prices will accommodate.
Given that the fair-share concept is at bottom about affordability, this is a devastating critique of the Glaeserite line on housing.
Elmendorf’s study for the Mercatus Center doesn’t cite Glaeser and Gyourko’s 2003 paper but the passage excerpted above appears in a 2002 working version of that paper that is referenced in a 2018 essay he wrote with Darien Shanske.
The Glaeser-Gyourko paper that Elmendorf’s Mercatus study cites, “The Economic Implications of Housing Supply,” published in 2018, offers a counterpoint to the authors’ 2003 proviso about demand. It reprises their view of onerous local regulations as the major source of higher housing prices and takes economic inequality as a given, but it also comments at length on the relationship between demand and prices:
In lightly regulated housing markets with growing population and economies, like Atlanta, the supply curve for housing is relatively flat. Thus, as demand for housing expands over time, the result is that competition in the home building industry holds the price of housing reasonably close to its minimum production cost. In heavily regulated housing markets with growing economies, like the San Francisco Bay area, the supply curve for housing slopes up. As a result, additional demand for housing translates into prices that are substantially above the minimum profitable production cost, with rising land values driving up total costs. Finally, in a housing market like Detroit where the demand for housing declined sharply over time, the supply curve for housing has a kink at the existing level of housing because housing is durable and does not diminish quickly when demand falls. As a result, a reduction in demand leads to lower prices for housing and minimal new construction. (4)
Referring again to the upward sloping supply curve in San Francisco, Glaeser and Gyourko wrote: “Thus, shifts in the demand for housing affect price more than quantity…[W]hat makes San Francisco housing so expensive is the bidding up of land values.” (16,18)
Granted, Glaeser and Gyourko’s usage of supply occasionally wavers. In the passages just cited, supply functions in the manner that Helm and Murray associate with economics proper—which is to say, in relation to trade. Elsewhere the term operates à la folk economics—that is, with respect to the amount of construction. For example, “[t]here is no doubt,” they write, “that binding density restrictions affect supply….Minimum lot size is strongly negatively correlated with new building supply across communities in greater Boston.” (6)
But the important point is that Elmendorf appears to lack even Glaeser and Gyourko’s grasp of supply as a willingness to sell. Coupled with the Glaeserite reliance on policy to “produce” housing, that deficiency is fatal to his approach, as evidenced by the fact that the hundreds of Yimby housing laws enacted in California since 2015 have failed to spawn a boom in construction.
Chris Elmendorf, what do you say?
