Civic Data

Data-oriented thinking about where and how people live.

Leave a comment

Review: “Zoned in the USA: The Origins and Implications of American Land-Use Regulation” (Sonia Hirt)

tl;dr: highly worthwhile book on the international uniqueness of American zoning codes and their cultural underpinnings. 4 out of 5 stars.

This book is an attempt to demonstrate and explain the features of modern zoning codes (and cities) that are almost unique to the United States: the very broad use of exclusively residential zones and exclusively single-family house zones. It combines an extended history of zoning, both internationally and in America, with a comparative study of modern zoning codes and legal regimes surrounding new construction in several countries, most notably England and Germany.

The book is fascinating overall, and highly recommended to any students of cities and/or American history. It is richly detailed and sourced. The perspective is that of an urban planning student from Bulgaria who moved to the US for career reasons and was baffled by the apparent contradictions between the narrative of American individualism and freedom and the very restrictive codes surrounding the built environments in which they lived (which is baffling to this native US citizen as well).

There are many hypotheses that attempt to explain American zoning codes. To name a few: the availability of plenty of cheap land in the US, the predominance of local as opposed to national control over planning and development, the protection of private property values, etc.

However, Hirt feels that these arguments are insufficiently unique to explain the genuine uniqueness of American zoning. The principal thesis of the book is that they are largely the result of a strong cultural undercurrent of agrarian and “frontier” values in the US. As a result, the single-family house on a generous plot was seen as the morally correct dwelling arrangement, and our zoning and legal codes responded to that desire.

The apparent contradiction described above results from the explosion of US cities in the 19th century, and the resulting collision between our preferences for different sorts of “freedoms”: political freedom (in particular the right to use private property without governmental interference) and what she calls “spatial freedom,” which is something like the desire to claim, explore, and patrol the boundaries of a sizable piece of land. It’s not exactly a spoiler to observe that spatial freedom won this rhetorical battle. However, political freedom was appeased in that the new legal structures were simple, scientific, rules-based systems that would treat each property the same and give each property owner the right to development without asking permission within the constraints of the rules, or were advertised as such anyway. And “economic” freedom was appealed to by the universal emphasis on stabilizing and increasing property values.

I did feel that the international comparisons beyond those to Germany and England were a bit overpromised and underdelivered. The sections on each of the other nations discussed (France, Russia, Sweden, Australia, Canada, Japan) amount to capsule histories of a page or two and are not the subject of extended comparisons throughout the rest of the book.

More information about the book is available at Cornell Press.

Leave a comment

Thoughts on 51B improvements at Bancroft and College

The intersection of Bancroft Way and College Ave is near the southeast corner of Berkeley’s campus. Caffe Strada, an extremely popular coffee shop, is on the southwest corner, and Freehouse, a relatively new gastropub that seems to be doing pretty well, is to the southeast. To the north is a large plaza which serves as a major entry point to the university. Tons of students walk through this intersection on the way to class from the dorms and apartments further down College Ave.

The intersection also serves the 51B, a major AC Transit bus line with a weekday ridership of over 19,000. The route goes north on College, turns west onto Bancroft at this intersection, and has a stop at the northeast corner. (I’m a regular rider — I take this line probably ~5 times a month to get to and from downtown and the Amtrak station.)

Currently, AC Transit is working with Berkeley to implement improvements along the 51B. This will entail moving bus stops to the far side of intersections, removal of a few stops, bus bulbs, “queue jumps” (removal of a few parking spaces in front of a signalized intersection, so that a bus can move ahead of stopped traffic), new traffic signals at some intersections, roadway widenings, possibly traffic signal priority or special signaling phases for buses. If AC Transit’s proposal is adopted, the line could see improvements of 7 minutes along the line — hugely significant, given that the total route length is only 30 minutes at rush hour.

AC Transit hopes to install a traffic signal at the Bancroft and College intersection to speed buses along. The many pedestrians at this intersection interfere with traffic, and a signal would force pedestrians to wait their turn while cross traffic moves through.

Despite enthusiastically supporting the speed-up plan, I dislike this particular proposal. My sense is that the main problem at Bancroft and College is mostly due to the fact that it’s simply way too large. The crosswalks are 45 to 55 feet from corner to corner, for an intersection which has two traffic lanes in each direction! Pedestrians take 12 to 18 seconds to cross, at typical walking speeds, and buses and cars legally have to wait the entire time that pedestrians are in the crosswalk. Then, because it takes so long for pedestrians to cross, another has generally entered the crosswalk by the time the first has left. It is extremely difficult for vehicles to get a nose in if they’re being polite (as they generally are, to Berkeley’s credit).

The image below shows a simple solution which uses sidewalk extensions to reclaim asphalt for pedestrians, reduce crossing distances to 25-30 feet (7-10 seconds), and generally make it easier for everyone (vehicles included) to cross the intersection.

shrunken bancroft and college

The shaded regions represent sidewalk extensions that would shrink pedestrian crossing distances to 25 to 30 feet.

The result would be an intersection that is more pleasant, safer, and more efficient for all parties — buses, pedestrians, and cars. (Many of the same considerations apply at Russell and College, another intersection which AC Transit has proposed to put in a traffic signal…)

Best of all, these changes could be implemented as a pilot project for a few hundred bucks with paint and planters. If after a few months it isn’t working, scrap it and start over!

Leave a comment

Visualizing population density: New York City

This is the third in a series of posts trying to figure out what various levels of population density look like at street level. Previous posts focused on the Bay Area and Boston (check the former for a mission statement and methodology).

7 people per acre (Munsey Park)

7, Munsey Park

8 (New Rochelle)

8, New Rochelle

11 (Woodmere)

11, Woodmere

20 (Port Washington)

20, Port Washington

23 (Jamaica, Queens)

23, Jamaica, Queens

25 (Long Beach)

25, Long Beach

27 (Jamaica, Queens)

27, Jamaica, Queens

32 (Long Beach)

32, Long Beach

36 (Glen Oaks, Queens)

36, Glen Oaks, Queens

44 (Gravesend, Brooklyn)

44, Gravesend, Brooklyn

56 (Jamaica, Queens)

56, Jamaica, Queens

59 (Maspeth, Queens)

59, Maspeth, Queens

73 (Bed-Stuy, Brooklyn)

73, Stuyvesant Heights, Brooklyn

80 (Harlem, Manhattan)

80, Harlem, Manhattan

94 (Bensonhurst, Brooklyn)

94, Bensonhurst, Brooklyn

110 (Crown Heights, Brooklyn)

110, Crown Heights, Brooklyn

113 (Park Slope, Brooklyn)

113, Park Slope, Brooklyn

140 (East Williamsburg, Brooklyn)

140, East Williamsburg, Brooklyn

183 (Sunset Park, Brooklyn)

183, Sunset Park, Brooklyn


Not sure what’s going on here — this is a surprisingly high density for townhouses. This is a mostly Asian neighborhood, and so there may be more people per household than in other areas.

184 (West Village, Manhattan)

184, West Village, Manhattan

190 (Chelsea, Manhattan)

190, Chelsea, Manhattan

199 (Harlem, Manhattan)

199, Harlem, Manhattan

200 (Lower East Side, Manhattan)

200, Lower East Side, Manhattan

207 (East Village, Manhattan)

207, Alphabet City, Manhattan

212 (Upper West Side, Manhattan)

212, Upper West Side, Manhattan

320 (Upper East Side, Manhattan)

320, Upper East Side

1 Comment

Visualizing population density: Boston

If you’re interested, you can check out the first post in this series (the San Francisco Bay Area) for a few words about I’m doing this, as well as the methodology I used. With that said, on to the pictures!

3 people per acre (Canton, MA)

3, Canton


9 (Brookline, MA)

9, Brookline


12 (Norwood)


12, Norwood


15 (West Roxbury)

15, West Roxbury


23 (Quincy)

23, Quincy

30 (West Roxbury)

30, West Roxbury

35 (Dorchester)

35, Dorchester

50 (Boston)

50, Boston

52 (Brookline)

52, Brookline

55 (Boston)

55, Boston

55 (Cambridge)

55, Cambridge

75 (Boston)

75, Boston

82 (Charlestown)

82, Charlestown

120 (Boston)

120, Boston


This block and the next are at substantially higher densities than comparable San Francisco townhouse neighborhoods. The width of the street accounts for a substantial part of this.

135 (Boston)

135, Boston

175 (Boston)

175, Boston

190 (Boston)

190, Boston

Again, narrow streets contribute substantially to higher densities than San Francisco.



Visualizing population density: Bay Area

Population density is frequently described as people (or housing units) per acre. But I don’t have a good feel for how “people per acre” plays out in a real-life context. What does a street with 15 people per acre look like? One with 30 people per acre? A hundred per acre sounds like a lot — is it a lot?

With that in mind, I recently spent quite a bit of time combining a map of census block-level population data in QGIS with the awesome power of Google Street View to get a sense for what population density feels like at street level. I focused on the U.S. because the data was accessible to me, but I’d love to see similar explorations of living arrangements around the world — especially European and Japanese cities.

My strategy was to pick blocks across the metro area that represented a variety of population densities. This was done in a quasi-random way — I tried to distribute the locations around the metro area, and I haven’t picked particularly nice or ugly blocks. I just looked at a map of population densities, picked a likely block, navigated there in Street View, and snapped a screen shot. The population density numbers I quote are the averages of the two adjacent facing blocks.

A caveat is that I focused on streets that were primarily residential, which will have higher population densities than blocks which contain other uses mixed in. On the other hand, residential uses constitute a large majority of the acreage of any city (e.g. 48% for Berkeley), and the figures I quote do include the square footage of the surrounding streets, which constitute 15-25% of the area of a typical city (24% for Berkeley, from the same source).  On net, if you’re interested in bulk population density on a city scale — including shops, schools, offices, parks, etc — you’ll need to deflate these figures by something like 25%.

On to the pictures

1.8 people per acre (Danville):

1.8, Danville

At the low end, these densities are necessarily imprecise — lot and block sizes aren’t so clear from the air.

3.7 (Danville):

3.7, Danville

5.8 (Danville):

5.8, Danville

8 (Danville):

8, Danville

12 (Santa Cruz):

12, Santa Cruz

15 (Santa Cruz):

15, Santa Cruz

18.5 (Santa Cruz):

18.5, Santa Cruz
25 (Berkeley):

25, Berkeley

26 (Berkeley):

26, Berkeley

Both Berkeley and Santa Cruz have a reasonably high fraction of legal or illegal “granny flats” — a separate living unit at the back of the lot. You can see a few in this image.

40 (Santa Cruz):

40, Santa Cruz

50 (San Francisco):

50, San Francisco

Our first townhouse neighborhood is not so high-density: mostly two-story with fairly large backyards

65 (Berkeley):

65, Berkeley

Individual largish apartment buildings can contribute substantial density.

70 (San Francisco):

70, San Francisco

Three-story townhouses.

78 (San Francisco):

78, San Francisco

Slightly smaller lots.

85 (Berkeley):

85, Berkeley

Fairly large apartment buildings mix with single-family and duplex houses.

125 (San Francisco):

125, San Francisco

Mostly apartment buildings here.

135 (San Francisco):

135, San Francisco

Townhouses and apartment buildings. The Bay Area doesn’t get much denser than this.

shattuck blake south


Safe space for bikes on Shattuck

For the attention-impaired, the short version of my thesis here is that Shattuck Ave, the main street of progressive and bike-friendly Berkeley, CA, is hostile to bicyclists who are not comfortable sharing the road with traffic moving at 30 mph. It is also incredibly wide, which gives us an opportunity to fix this problem. A little ways down the page, I describe a simple and relatively cheap proposal to fix this condition and increase the number of people shopping, socializing, and general moving about in downtown Berkeley. Please read on!

A bit of history

I’m not an expert on the history, so I’ll keep this brief. Shattuck Ave. has long been the Main Street of Berkeley — it’s sufficiently historical that it has a Wikipedia page.


The picture below shows Shattuck in 1892 — follow the link for some interesting local history.

Basically a mud pit with a rail line through it. As Berkeley grew, Shattuck Ave developed commensurately.


I can’t find a date for this picture, but I would guess late 20s — the 12-story building in the middle is the Wells Fargo Tower, completed in 1925. Talk about wide! There’s space for two entirely separate rail lines — heavy rail and streetcars. The heavy rail hung on until 1941 and the streetcars until 1948.

By the 1970s, like downtowns everywhere, Shattuck was in decline. In conjunction with the opening of the BART in 1973, Berkeley decided that a multi-way boulevard treatment — like Paris! — would revitalize its downtown.


This vivid photochrom image doesn’t actually show the boulevard, but it does show the lovely architecture and automobiles of the time. The area at the bottom of the picture is used as a bus stop instead of parking these days.

The Boulevard

Here’s the street view of a representative block of Shattuck:

You can see how incredibly wide Shattuck is — 4 lanes for moving traffic in the middle, a wide median (which becomes a turning lane at intersections), two access roads, each of which have a moving lane and an angled parking lane, and relatively wide sidewalks on top of that. From building front to building front, the right of way is approximately 160′ wide! At present, 50′ to 60′ of this right of way is for fast cars, 70′ is for slow and stopped cars, and 20′ is for people. (Plus 10′ to 20′ of medians — used for mid-block vigilante picnics further north near Cheeseboard! But not so here.)

Why? A Complete Bicycle Network

Berkeley has a reputation as a bike-friendly city — a reputation which, in the American context, it certainly deserves. The most notable cycling infrastructure is a network of bicycle boulevards. These are principally residential streets which have low traffic volumes and speeds due to traffic diverters, and are indeed the safest way to get to many parts of town by bicycle.

But Berkeley’s bike boulevards as presently implemented have a major drawback: they generally have no commercial development. As a result, one is forced to leave the boulevard network to get to shops and restaurants by bike, and use major streets like Shattuck. Compared to the places one can directly and conveniently access by car, the bicycle network is generally sparse and uninviting. Bicycling is the most efficient mode of urban transportation (in cost, land, and energy) and we have misprioritized our streets.

In this context, it’s distressing that Shattuck is hostile to cyclists. Cars commonly travel upwards of 30 miles an hour, though strictly speaking the speed limit here is 25, and there are red lights at most cross streets (an example of the “hurry up and wait” theory of traffic engineering). As noted above, there are low-speed access lanes mid-block, which the less brave cyclists such as myself tend to use, but this requires re-merging with traffic to cross every intersection. Needless to say, it doesn’t feel safe. Despite this, a respectable number of people bike to and from the downtown Berkeley BART station on a daily basis, and in fact the city has established a Bike Station with secure and cheap/free bicycle parking in a storefront on Shattuck. Berkeley should support these investments with streets that are safe for bicyclists.

A secondary reason is that it would actually be remarkably easy! Shattuck is very wide, at 160′ of right of way, and making a safe space for bicycles without taking away travel lanes or parking spaces is quite straightforward.

A fun aside: residents of a car-optional city have a ton of extra money to spend around town. If you can skip buying a car — at $9122/year ownership costs, by AAA estimates (pdf) — you can afford an extra $157,737 worth of house, on a 4% 30-year mortgage (pretty typical rates these days, though low by historical standards). Safe bike lanes are about dollars and cents.

Safe space for bikes, on the cheap

The red line on the following picture marks the nine-block stretch of Shattuck between Allston and Carleton which has the “boulevard” configuration described above, with access lanes and parking on both sides. The project would focus on this area.

boulevard status

My proposal would simply add curb cuts and bike lanes to connect the access lanes at the intersection. The access lanes would now simply be a throughway for bikes, while still being used by cars accessing the angled parking. In the access lanes, cars are moving sufficiently slowly that sharing space is tenable — this would be plenty comfortable for most people, even (and especially) for those who don’t think of themselves as “cyclists,” and would constitute a massive improvement over the current configuration.

I’ve edited some pictures to give you a sense of what I mean. This one is facing south, at the intersection of Blake and Shattuck:

shattuck-blake south

The space that would be repurposed for a bike lane is shaded green, and the space overlaid in white would be lowered to a height of ~1 inch so that the lowered bike lane doesn’t constitute a barrier to pedestrians (see below).

Same intersection, facing north:

shattuck blake north

Same as above. Apologies for my meager photo-editing skills!

Obviously the crossings don’t need to be green, but they should be distinctively marked in some way — with some combination of paint and/or pavement — to make them visually obvious to turning drivers.

Implementation-wise, the simplest option would be to simply to cut a path through the pedestrian bulb-outs, so that bicyclists continue through at street level. However, pedestrians should always have the right-of-way before cyclists, and a deep cyclist “trench” would hinder pedestrian movement. I think the main problem is really that the curbs here are too high. I would simply lower the curb for the entire bulb-out (say, from the right edge of the planter leftward in the picture above), so that the bike lane cut-through would only be ~1 inch below the pedestrian space. I ain’t a lawyer, but I imagine that this would also help with ADA-compliance.

low curb

A lowered curb for the entire pedestrian bulb-out would make the cut-through bike lanes much less of an intrusion into the pedestrian realm. The curb/pedestrian island pictured is in Copenhagen.

I see two obstacles to this. The first is fairly minor. The city saw fit — presumably in the last redesign in the early 1970s — to locate massive concrete planters on the southwest and northwest corners of each intersection. As you can see in the picture immediately above, these would need to be removed or relocated. I have no idea how expensive this would be but the planters are pretty imposing in that Brutalist kind of way. (This also serves as a warning about over-designing projects — small is better than big, and fast is better than slow, because if it doesn’t work you can always go in and change it later! Cities need to make more small, experimental investments that potentially have big payoffs.)

The second issue is bigger. There are a total of nine bus stops along this corridor, all of which are recessed from the street, so that buses can pull off to the side for loading/unloading without disrupting traffic. If the bike cut-throughs continued straight on as described above, they would pass directly through the bus loading area. I’ve tried to diagram the problem and my solution for a representative example below, at the corner of Shattuck and Dwight.

shattuck dwight

The green area shows where the bike crossing would be. At present, this area passes directly through the bus boarding area. The white area represents an extension of the sidewalk — a bus loading island.

I think the best solution here is to extend the bus stop to be flush with the rest of the access lanes, which would allow plenty of space for the bicycle crossing. This has the downside that a bus stopped for loading and unloading would be blocking a travel lane. However, this is actually a common operating condition for buses in the rest of the world, because this allows the bus to take a straight path, making for a much nicer, smoother, and more reliable ride. (Consider that after pulling over to a stop, buses have to wait for a break in traffic to get moving again!) Furthermore, traffic on this part of Shattuck is not all that severe — most of the delay for cars traveling through this area consists of waiting at red lights. Some political will would need to be mustered, but the benefits for pedestrians, cyclists and public transit outweigh minor delays for car travel.


How much would this cost? This will be totally uninformed speculation on my part — I have no experience with road or sidewalk construction costs. But maybe a ballpark estimate is better than nothing.

I count 30 bulb-outs along the boulevard-ized part of Shattuck similar to the intersection at Blake pictured above. A very rough estimate of the area we’re reconfiguring on each bulb-out is 1000 sq. ft. The proposal also involves the construction of 9 pedestrian islands at bus boarding locations, at about 1800 sf each. Thus the total square footage is roughly 50,000.

This site estimates that the cost — labor + materials — of replacing an old concrete driveway with stone pavers is about $7/sf. That’s describing a residential use, and it’s probably appropriate to expect that a heavily-used piece of civic infrastructure will cost a bit more — but a city implementing a relatively large construction project can also get economies of scale, so I’ll use $7/sf as a metric, totaling $350,000 for the pavement. Throw in some paint to mark the lanes at intersections and add sharrows in the access lanes, round up a bit, and I think $400,000 is a reasonable expectation for the total cost.

In another post, I’ll describe a slightly more radical proposal for a redesign of the access lanes with a substantial expansion of the pedestrian realm. But for a quick fix on the (relative) cheap, I think this would be a great start. What do you think, Berkeley?

Leave a comment

Fact-checking PolitiFact

(I bet nobody’s ever used THAT title for a blog post before…)

A few days ago, the Republican National Committee tweeted “Is it a coincidence that Americans are fleeing tax-and-spend states like CA for states that spend money responsibly? #jobs”. In response, published a blog post titled “Are Americans flocking to low-tax states?” in which they half-hearted argued that the data does not show what the RNC claimed.

The research post used a lot of words — it split the states into high- and low-tax groups, and looked at the states within those buckets. We can do better.


Here I’ve used the same data sets that PolitiFact did on net migration and state tax rates. That’s a pretty decent correlation. If you want to play with the data yourself, I’ve uploaded it to Filtergraph, a neat web service that tries to make it easy to work with datasets interactively.

But Why?

Well, who knows. The conservative narrative is that the higher tax rates of states like CA and NY drive workers out of the state. The data is suggestive but doesn’t necessarily prove this. The idea is basically plausible only if you stipulate that state/local governments are providing services that people don’t actually want, or providing desirable services incredibly inefficiently.

(NB: I recently read a paper which pointed out that low-skilled workers are the dominant contribution to the migration rates described above, but that high-skilled workers have very different migration patterns. Specifically, high-skill workers tend to migrate into the largest “knowledge-oriented” metros, such as NYC, San Francisco, or Boston. For skilled labor, productivity and wages are relatively much higher in those areas, but low-skilled workers don’t get enough of a wage bump to make the high cost of living a good trade-off. Unfortunately, I can’t find the paper!)

For fun, I’ve included state-level unemployment in the Filtergraph data set above. Unemployment rates have almost no correlation with either tax rates or out-migration (an R^2 of about 0.09 for each, though Filtergraph won’t automatically calculate best-fit trend lines). So it’s not all that clear that tax rates are closely tied to state economic performance.

Another Narrative: Restrictions on Housing Supply

Why does housing in Manhattan cost so much? The short answer is that supply doesn’t keep up with the demand. Part of this is due to physical restrictions — it’s an island, and building up is somewhat more expensive than building out. But another important contributor is regulatory restrictions. Glaeser et al showed rather elegantly that housing costs on Manhattan south of 125th St. are at least twice as high as the cost of constructing that housing. The authors infer that restrictions on development are contributing significantly to the lack of supply, and thus to the insanely high housing costs. (In 1960 alone, 13,000 new units were permitted in Manhattan, vs. 21,000 for the entire decade of the 90s.)

A less dramatic version of the same story plays out almost everywhere in America where there are people already living, most notably through wildly complicated use-based zoning codes that limit higher-density housing and require too much parking. In most areas, cheap housing can only be built on greenfields at the periphery of the urban area, where there are no neighbors to complain. But because there is no demand for high-density housing in these remote and cheap areas, only detached single-family homes get built. The primary result is auto-oriented sprawl.

Housing costs account for a much higher fraction of the average American’s income than do tax rates. And variation in housing costs is much higher than variation in tax rates. If a large fraction of housing costs in expensive areas is due to regulatory barriers (and not, say, higher local incomes), we might expect low-skilled workers to be driven to cheaper states. I grabbed Census data on median housing costs by state, taking a weighted average of owner-occupied and rental units, and then normalized by median 2009 household income from Wikipedia. (This data is also available on the aforementioned Filtergraph page.)


The evidence isn’t clear. The fit here is statistically weaker than for tax rates. But median housing costs isn’t a perfect proxy for regulation — some states have low housing costs because there’s no demand, not because of a less regulations on development. And low-wage workers don’t get the same salary boost from living in productive metros that high-wage workers typically do, so  median income may not be the appropriate normalization. We really need better data to distinguish between these explanations! In general, though, I’m more sympathetic to the “regulatory tax” narrative, simply because the costs involved can be much larger than the differential tax rates between states.