When Scott Wiener released his proposal  to spur up to 3 million new homes near transit (Senate Bill 827 ), it received swift condemnation and praise from all corners. Whatever its merits for housing policy – and I happen to think they are quite good – it is not great transit-oriented development (TOD) policy.
The goal of good TOD must be to minimize driving’s share of trips that start or end in the area while also building a place that people actually want to live in and visit. It does this by blending homes, retail spaces, offices, and government buildings like schools into a single neighborhood. Each use supports the others: retail needs office workers for daytime activity; residents need retail so they won’t need to leave often for shopping trips; offices need homes, transit, and retail nearby to be attractive places to work; and transit connects the neighborhood into the broader urban fabric.
In a place like this, transit would probably mostly be used for commuting to and from the neighborhood. Walking should be a viable option for running errands, which make up fully 70 percent of a household’s trips . Driving is available to get around to areas outside the reach of the transit system or for those times when one just needs a car.
Because people are willing to walk further from home to a transit station than from a transit station to their work , and because retail should be accessible to the whole area, the layout of good TOD would look something like a donut:
The purple areas on this zoning map of San Anselmo are "neighborhood commercial" zones, leftover transit-oriented developments around the town's long-abandoned Interurban stations:
So, TOD involves a mix of uses with higher-intensity and more mixed uses nearer the transit station. Most non-work trips can be done on foot if so desired and transit allows easy commutes to and from the place. TOD under Senate Bill 827 (SB827) won’t do this.
SB827 eliminates residential density limits around transit, eliminates parking minimums, and raises the height limit to between 45 feet and 85 feet, depending on the context. But by solely focusing on residential, the mix of uses that makes good TOD is lost.
This is a typical failing in transit-oriented designs. Designers often focus on transit-oriented housing, forgetting that non-work trips are too short for transit to be useful. Further, placing only housing around transit will reinforce peak-direction transit use. Good transit needs people moving in both directions at peak hours – to and from the central business district – to take advantage of spare capacity on reverse-commute trains. Without jobs near suburban transit stations, the counter-commute capacity goes underused.
SB827 should give a development bonus rather than a strictly residential bonus. Raising height limits and lifting various restrictions on density and parking around transit stations would allow the market to determine what makes the most sense from a rental perspective. After all, San Francisco has more expensive office space than even Manhattan, indicating a severe shortage of commercial space. If the state wanted to get more prescriptive, it could define where the bonus could be applied to all uses and where it could only be applied to residential projects.
SB827 has other shortcomings, especially the lack of protections for transit systems, but its poor approach to TOD is big. Thankfully, the fix should be relatively simple. Here’s hoping for improvements as the bill moves through committee.
 Scott Wiener, “California Needs a Housing-First Agenda: My 2018 Housing Package,” Medium.com, Scott Wiener (blog), January 4, 2018.
 Scott Wiener, “Planning and Zoning: Transit-Rich Housing Bonus.,” Senate Bill 827 (2018).
 Adella Santos et al., “Summary of Travel Trends: 2009 National Household Travel Survey” (Washington, DC: Federal Highway Administration, June 2011).
 Kwoka, Gregory J., E. Eric Boschmann, and Andrew R. Goetz. “The Impact of Transit Station Areas on the Travel Behaviors of Workers in Denver, Colorado.” Transportation Research Part A: Policy and Practice 80, no. Supplement C (October 1, 2015): 277–87.