UCLA Planning Professors Michael Manville, Michael Lens, and Paavo Monkkonen (MLM) have responded to my article “Hypocrisy in the local zoning debate,” posted by 48 hills on December 13. I appreciate the opportunity to engage them in a dialogue about their work on the relationship between local land use regulations and housing affordability.
MLM write: “Ms. Bronstein claims we wrote one paper saying that the connection between zoning and regulation was completely established and settled, but another saying there was actually no data to support that claim.”
Nowhere did I accuse MLM of “saying there was actually no data to support [the] claim” “that the connection between zoning and regulation was completely established and settled.” What I wrote: “they themselves had conceded that such evidence is partial, if not altogether lacking.”
In the article MLM prepared in collaboration with the Terner Center for Housing Innovation at UC Berkeley, “Built-Out Cities?,” published in February, the UCLA trio write that given the “variety and complexity” of local land use regulations in California cities, “there is virtually no way to capture regulatory stringency fully in a consistent quantitative measure” (6). In their reply to me, they acknowledge that “surveys [of planners] have errors, and biases, and anyone using them needs to account for those problems.” Yes, but my point was that they did not account for any errors or biases or the lack thereof in the survey they used, Terner California Residential Land Use Survey.
MLM also write: “one can’t say, as Ms. Bronstein does, that we rely exclusively on the Terner Survey.” They must be referring to this passage in my article: “…to discern how regulation does or does not influence housing supply, and thus price (4) MLM draw on just such responses [surveys of planners], without explaining how their source, the Terner California Residential Land Use Survey, got it right.” In retrospect, I should written “one of their sources.” MLM supplemented the data from the Terner survey with data from the Housing Elements in cities’ general plans. “[T]he main purpose of our [Terner] paper,” they write in their reply, “was to introduce a different metric of regulation—data from Housing Elements—and show how it overcame many of the problems that afflict surveys of planning staff.” So let’s take a closer look at their presentation of the Housing Element data.
MLM write: “Our interest in the HE process does not stem from its being a valid or objective planning tool. To the contrary, the HE process has numerous flaws (Monkkonen et al., 2019). One can argue that calculating a precise number of housing units each city ‘needs’ is largely impossible” (9).
So let’s get this straight: To correct unspecified problems that, they admit, “afflict” one source of data, the Terner survey of planners, MLM want to use data from another source that they deem flawed, cities’ Housing Elements.
Why do they bother with Housing Element data at all? The most important reason, they say, is that “the HE is useful precisely because it is, for many cities, a political exercise that can help the city deflect new housing” by for example, underreporting the amount of land they have that could be developed or “over-report[ing] their actual zoned capacity, satisfying the state without actually making room for new development” (10).
And now we come to their basic assumptions:
“No cities are in fact ‘built-out.’ The modern elevator was invented in 1903, and engineers have known for half a century to construct buildings over 100 stories tall. There are of course reasonable arguments against 100 story buildings, but most of urban California is between one and two stories. Build-out is a political judgment, and cities that determine themselves closer to build-out are cities that are politically more hostile to development” (10).
This is confusing. If there are reasonable arguments against 100-story buildings—and there are, though MLM don’t make them—why mention such behemoths in the first place?
Their belief that “no cities are in fact built out” is illuminated by the following passages, also in the Terner paper.
First: “Our goal in this paper is to look inside the black box of regulation, and discern how regulation does or does not influence housing supply, and thus price. We examine supply at the city level: our primary dependent variable is how much building occurs in a locality. Our focus is thus at the scale of the regulating authority (the city) rather than the scale of the market (the region). This allows us to assess the mechanism by which scholars theorize that regulation influences price” (4).
Maybe so, but where is the data that justifies that focus? How much latitude do cities have vis-à-vis regional economic pressures? That’s a fundamental question that, to my knowledge, MLM don’t raise. In the Bay Area, the answer is, not very much: The flood of highly paid tech workers has boosted the price of real estate to astronomical levels. Thanks to COVID, those levels are now falling for rentals (though not for single-family housing), but most apartment rents are still far beyond people of modest means. This is a case of the theoretical tail wagging the empirical dog.
Second illuminating passage: “When housing prices rise, developers should build more housing units. The extent and pace of this building is determined by the market’s price elasticity of supply. A city that constrains housing production is making the supply of housing less elastic. How so? Supply elasticities have multiple determinants—including the capacity of the housing development sector—but two that cities control are the complexity of the production process (what we characterize as process) and the availability of inputs (in this case, land zoned for development)” (8).
MLM note that supply has “multiple determinants,” but the only factors that interest them are “two that cities control.” They nod to “the capacity of the housing development sector” and then move on to municipal sovereignty. To justify this maneuver, they’d have to demonstrate that the impact of local regulations massively outweighs the impact of the housing industry’s capacity; and that absent regulatory constraints, developers would pile into a city. They do nothing of the sort. Their real target, then, is local political prerogative. Using the housing affordability crisis as a pretext, they seek to consolidate land use authority at the state level of governance so as to expedite endless growth.
Next in their reply to me, MLM bring up their exchange with Michael Storper and Andrés Rodríguez-Pose in Urban Studies. “To hear Ms. Bronstein tell it,” they write, “in that article we concealed from readers all the problems that we knew existed with land use surveys, even as, in the Terner paper, we admitted that those problems existed.”
Nope. All I said was that they “defended the conventional argument that regulation drives up prices.” What MLM actually wrote in their reply to Storper and Rodríguez-Pose about land use surveys is that the Wharton Index, “the best known” metric of regulatory stringency, is “imperfect” (5). But just as they never specify the problems with the Terner survey, they didn’t detail the imperfections of the Wharton Index. (In a letter they sent to Urban Studies editor Andrew Cumbers soon after their exchange with Storper and Rodríguez-Pose, MLM state that two other scholars “find no difference between Wharton and the Terner survey.”)
MLM exited their Urban Studies discussion of surveys with this statement: “let us concede in the broadest sense that some of [Storper and Rodríguez-Pose’s] criticisms are sensible: it would be advantageous to have measures of regulation with more observations, analyses that captured more complexity, and regressions with better identification strategies” (6). A concession made only “in the broadest sense” is an evasion.
I reported that a paper posted in July, “Superintending Local Constraints on Housing Development,” co-authored by Monkkonen and three other UC scholars, states: “HCD [California Department of Housing and Community Development] has no reliable data about the severity or prevalence of local governmental constraints” on housing. In their citation of my article, MLM omitted that passage. That omission lets Monkkonen off the hook of my hypocrisy charge. Putting him back on that hook, I ask: How do you reconcile the contradiction between your July paper’s categorical assertion about “no reliable data” and your present reply’s embrace of “uncertainty”? I invite Lens and Manville to answer that question as well.