Public Comment

The quadplex upzoning item fast-tracked to council by members Rigel Robinson (D7) and Lori Droste (D8) contains some (to me, shocking and alarming) fabrications in the cornerstones of the authors’ arguments in favor.

Race-washing

The account of history they offer is highly distorted and, by misrepresenting actual racialized oppression, it “race-washes” the quadplex item, falsely presenting it as a major anti-racist move. People other than myself are already speaking about that and I won’t dwell on it here because it is too complicated for a short note. I will note that the policy promoted in the item originates with YIMBY corporate and major investor sponsors - big banks, large corporate developers, and so on - and so ironically, if we we are to believe the racial justice claims made in this item, we must accept those corporations as sudden and unexpected champions of racial justice. You know, Citibank for justice, I guess. 

Deeply flawed economic analysis

The memo for the item argues that recently constructed duplexes, triplexes, and quadplexes house surprisingly low income households, and thus building more will mostly serve lower income households. The gist of the argument is summarized in the first table from the memo.

The table purports to show median household incomes, broken down by housing typology (number of units) and year of housing construction. I think most people would find some of those numbers at least a bit surprising. They would be correct to think something is amiss. 

The first detail to note is the citation in the caption, which reads: “American Community Survey, 2014-18, Public Use Micro Data Set, US Census”. A few notes, first the survey in question is customarily referred to as “5 year American Community Survey, 2018”. Second, the data source cited is not called a “Public Use Microdata Set”, it is called a “Public Use Microdata Sample”. These errors would not mean much were it not for one “small” detail: 

The Public Use Microdata Sample cited simply does not contain data from which the table could be computed, even in principle. The table appears to be a complete fabrication, or the product of very faulty methodology. The memo does not say who actually prepared the table (it seems to come from the YIMBYs online but I am not certain). Droste has not provided any explanation of the methodology behind the table. The YIMBYs and author of this measure can easily show me up, if I am wrong. I will gracefully concede if they can. They know this. They have not attempted to defend what appears to be a fabricated table. 

Bogus environmental claims

“Last year, climate researchers in Berkeley quantified local and state opportunities to reduce greenhouse gases from a”comprehensive consumption-based perspective." The most impactful local policy to potentially reduce greenhouse gas consumption by 2030 is urban infill. In short, Berkeley can meaningfully address climate change if we allow the production of more homes near job centers and transit." 

Neither the underlying paper from the research group in question, nor their web site, makes that claim. Whoever wrote that paragraph misrepresents what that research was even trying to explore. 

Comically (or tragically) the second image offers a screenshot from that research team’s website (coolclimate.org), shown here. 

The graph is nonsense. I’ll explain why. First, let me note that I can easily recreate that same table. The tool used to do so is accessible from this page: 

https://coolclimate.org/scenarios 

If you poke around there, you’ll find a link to the interactive tool for creating such a table, a link to a tutorial about how to use that interactive tool, and a link to the underlying publication from the team that built the tool. YIMBYs started blasting that same graph ever since the tool was first published - though amusingly at least one local YIMBY has been deleting his old tweets where he mentions it. Anyway, here is my recreated bar chart with more of the context shown. 

Please notice the “sliders” underneath the graph. They represent local assumptions about the “adoption rate” of the planned outcomes of a given policy category. Frankly, the research team did a poor job of explaining how they define “adoption rate” but as you can see, the setting of “adoption rate” for urban infill policies is set for around 70%. I believe that is supposed to mean that by 2050, 70% of new housing in Berkeley will be built in what are today’s lowest carbon footprint neighborhoods - mainly the R1, R1A, and R2 neighborhoods in the flats. 

Why is that slider set for 70%? No particularly specific reason. It is just the default, used for every city included in the study. The entire point of this tool is that policy makers are supposed to make reasoned estimates of how those sliders should be set locally, and look at that resulting graph. The quadplex memo does not do that or attempt to. It appears that whoever wrote that section of the memo just had no idea what the tool was about or what its output means. 

To illustrate, here is a graph that results if we think that, no, not so much new housing will be built in those areas. This seems plausible to me because even with upzoning, it won’t be economically attractive to redevelop expensive, existing detached housing. My expectation would be that upzoning mainly leads to the destruction of rent stabilized properties. 

I have shifted the “urban infill” adoption rate to about 15%, leaving all other settings the same. Notice how “Urban infill” policies have a greatly reduced alleged impact in my adjusted bar chart. 

By playing with those sliders, the bar graph can be changed radically. What are the “right” settings? Who the hell knows - the tool and the methodology behind it is simply not well enough documented to make a principled decision – but the default settings, used by Droste et al., miss the entire point of the tool by failing to contemplate local conditions. 

Once again, Droste et al. have haughtily cited research that simply does not even claim to offer the data they describe in their conclusions. 

In other words, the Environmental Sustainability section of the memo also appears to be a complete fabrication. Droste et al. can again show me up by giving a robust and convincing description of their reasoning and methodology, and I will gracefully concede if they can do so. 

Is this quadplex item an example of all the seriousness and competence with which D8 and now the whole city is being represented by council? We are surely in sorry shape, if so.