Density and Transit: Some Numbers – Minneapolis St. Paul Edition

The Old Urbanist has a nice post from last year: Density and Transit: Some Numbers. I steal one of his graphs and quote his words:


“While there are a few transit overachievers, notably Portland, Seattle, Boston and Washington D.C., the correlation is overall very strong. No city with an overall density of less than 4,000 per square mile, and there are many, has broken a 10 percent commuting modal share. The most notable outlier, Miami, may possibly be explained by examining its CBD employment density and job concentration in the charts below. “

Looking at the data for the City of Minneapolis, we are in the terms of the Old Urbanist a “transit underachiever”. That is, we have both a CBD employment density that is above average (5th highest in the US at 146,000 jobs per square mile), and a residential density of 7000 persons per square mile. But we have a underperforming metropolitan area mode share of 4.67%.

Of course there is a lot going on, the City of Minneapolis Transit mode share by place of residence is much higher than the regional average, at 14%, (for comparison, St. Paul is 11.3%) while most of the rest of the region must be below 5%. So while the core cities of Minneapolis and St. Paul are decent (though by no means champion) Transit Cities, the rest of the region is dragging the metropolitan area’s statistics down.

In life, there are two basic strategies: Prop up the weak or reward the strong.


In transit, there are two investment strategies: invest in areas with poor ridership to increase it, or invest in areas with decent ridership to make it even better. From a transit agency financial perspective, the second is clearly better (it is what a private operator would do). From a spatial/political equity (or suburban welfare) perspective, the services get spread around. We get high-capacity service to third-ring suburbs and even exurbs. This would not be a problem but that resources are scarce and our eyes are bigger than our wallets.

Similarly, we can add services in areas which are already built up, or we can add services in places in advance of development, because of speculative “economic development”.

In a recent post, I suggested some alternative transit routes, e.g. Rosedale to Southdale, the Purple Line, or Fridley to St. Paul, the Pink Line. There is nothing magical about these proposed routes, and I certainly have no religious feelings about them. (Nor do I own real estate along the rights-of-way).

They just have nice mostly available rights-of-way, key demand points along the routes (large shopping destinations, universities, downtowns) and serve the core cities and first-ring suburbs which have above metropolitan average population densities and thus should be feasible to serve with a higher capacity service that does not get stuck in traffic.

14 thoughts on “Density and Transit: Some Numbers – Minneapolis St. Paul Edition

  1. David

    Here’s a letter I sent to the Strib last night. I think it applies well to your statements on SWLRT as well. What assumptions are you making of who is and is not a resident of Minneapolis and what their needs are?

    “Louise Erdrich is the latest in a parade of people spreading misinformation about the Southwest LRT line. Her most egregious statement is that the line does not serve the people of Minneapolis. She has quite the exclusive view of who is and is not a resident of Minneapolis.

    Start with downtown. Obviously, the Southwest LRT serves downtown Minneapolis. It will make the same downtown stops as the existing Blue Line as well as adding a Royalston station in the North Loop. Perhaps Ms. Erdrich has missed the population boom in our downtown.

    But even more appalling is the total ignorance of how this line creates a critical opportunity link for residents of North Minneapolis to jobs in the southwest suburbs. There is currently no reasonable transit service for the residents of Near North and beyond out to jobs in the southwest corridor. Southwest LRT is a game changer for the neighborhoods in our city with the worst levels of unemployment.

    Ms. Erdrich’s concern lies with the Kenilworth corridor, a strip of land that has been a freight rail corridor for over a century and a designated transitway for decades. Project engineers have done extraordinary work to keep users of the bike trails in mind during design of this very important transit project. I am willing to find solutions for bicyclists. I am not willing to continue cutting neighborhoods off from opportunity so that a small minority uncomfortable with change will feel better.”

  2. Kasia McMahonKasia

    “In transit, there are two investment strategies: invest in areas with poor ridership to increase it, or invest in areas with decent ridership to make it even better.”

    What about the strategy of collecting data, analyzing it, and designing a transit system that will work for the people that need it (getting them to jobs, entertainment, and shopping?) We live in a world where data is a nearly limitless resource–why aren’t we using it to plan a regional transit system? Transit planning shouldn’t be decided by what is politically popular, it should be based on Data.

    As far as strategy, that should be simple. Get transit dependent people to their jobs and back home again. I know that the supreme goal of transit planners is to increase the number of people that use public transit and decrease urban dependence on cars…thereby decreasing carbon emissions…preventing climate change….and saving the world!! But I think we should all just calm down, step back, take some of the “save the world” pressure” off of transit planners, and just plan something that actually works.

    1. Nathaniel

      In defense of transportation planners, I don’t know a single one who thinks they are saving the world by adding a light rail line or by improving bus service. The benefits you describe, such as a reduction of carbon emissions, are positive externalizes but are rarely (actually never) used as a primary or secondary reason for creating a line.

      Your argument about data usage is misguided. How do you collect extensive data on a line or service that does not exist? There are methods, admittedly not perfect, but our reliance on data and its usage can be misguided. In fact, one criticism many have of transportation planning is its reliance on too much speculative data during the creation of new transitways, roads, highways, suburban collectors, developments, etc.

      1. Nick MagrinoNick Magrino

        Yeah, I was going to say that lots of data and numbers and charts and formulas got us…1,000 boardings in Kenwood and 1,100 in Uptown on Southwest. Let’s not overthink things too much.

    2. Adam MillerAdam

      It’s not just save the world. Nor should it just be serving those who are “transit dependent.”

      It should be about increasing the number of people who can take advantage of transit and thereby improve their lives.

      Cheaper, more efficient transportation means having more to spend on things you don’t hate (i.e., not commuting).

  3. Kasia McMahonKasia

    I don’t mean to knock transit planners! There is a lot of political pressure to support any investments in transportation because of the urgency of climate change (which is urgent) and other political pressures. There barely seems enough time to think about what we are doing (SWLRT planners are proud of their record speed for example). And the costs are so high.

    Instead of trying to solve all of our transit problems, I think we should let data focus our efforts on what is possible and what is a priority. Here is an example of a recent project that used cell phone data to optimize bus routes:

    What about the go-to-card data? I am not even sure if they are collecting that, but I sure hope so! I believe “guesstimates” (in ridership numbers, TOD potential) play far too great a role in the decision making process. There are plenty of creative ways to collect data. How about a metro area survey? Anything….

  4. Alex CecchiniAlex Cecchini

    I think one of the bigger takeaways is… if you don’t have a land-use plan, you don’t have a transportation plan. Increasing residential density (alone) is not the only way to drive transit ridership (ie, increasing it today without changing/improving on transit, bike, etc options will not likely see an increase in that mode share). But the data shows it becomes a much more natural choice for commuters and daily trips as density increases (and, one would assume, the density and accessibility of daily needs like groceries, day care, restaurants, shops, etc by foot and transit). It becomes far more cost-effective to build transit lines that serve these amenities.

    I’m glad to see the Minneapolis Climate Action Plan ( ) targeting reducing some barriers to construction that allows more neighborhoods becoming transit-supportive. But honestly, other than removing the minimum lot-area per unit restriction, has there been any talk of removal of any other requirements and shifting toward more form-based codes? The fact that the DInkytown Opus development was so disputed and a relatively close call in being approved, despite the slam-dunk design, location, and replacement of existing land-use really makes me question whether the city is serious about making it possible for a larger share new residents (and businesses) to take hold within the city’s border rather than on the exurban fringes.

    Beyond that, the seeming disconnect between Minneapolis and St Paul as municipalities, MetCouncil, MnDOT, and any other bodies regarding investments in transportation (in amount and mode) boggle my mind. Conflicts between Hennepin County insisting streets like Hiawatha and Washington Ave must conform to their vehicle counts and speeds while mostly regarding other forms of transportation (notably foot, bike, transit) along stretches as irrelevant to the conversation underscores this (evidenced by Metro Transit not having any real input on the Washington Ave re-design).

    When can Minneapolis and St Paul become cities truly well-served by transit? By this I mean 80% of the people could choose to take transit to 80% of their daily needs and have it be cost and time competitive with owning a personal vehicle and doing the same.

  5. Adam MillerAdam

    Is one reason that DC is an overachiever that it chose the first option? The DC metro reaches fairly far out into the suburbs, and seems to have sparked significantly greater density in and around those suburban stops.

    Of course, DC is also going to be less dense in its core as long as it sticks with its height restrictions, which may also help lead to increased density around suburban metro stops.

    1. Cameron ConwayCameron

      DC’s ridership successes are for a few reasons:

      1. It’s heavy rail system was implemented incredibly well. Almost completely underground in the city and quite a bit in the suburbs as well, with most stops either right at a commercial intersection or in an area prime for Transit Oriented Development. Having it be one of the most extensive heavy rail networks in the country certainly doesn’t hurt either.
      2. Metro area traffic is so terrible that downtown employees form HOV sharing lines called slug lines (, so naturally folks are also drawn to transit. Metro access is so desirable that real estate tends to get more expensive the closer you get to a station.
      3. DC has one of the most bonkers real estate markets in the country. I lived in an Adams Morgan house that rose from $70,000 to $700,000 in 10 years. As traffic has gotten worse and walkable lifestyles have become more fashionable, the inner suburbs of DC have absolutely exploded with Transit Oriented Development. DC is also gentrifying frighteningly quickly, which is spurring large scale development in most areas of the city.

      Also, the height limit really only holds weight limiting development downtown, where the height limit is actually being reached. There’s a lot of NIMBYism at work keeping taller developments at bay.

      1. Adam MillerAdam

        There is a lot of NIMBYism, but the height restrictions do mean that downtown development will never be as dense as, say, midtown Manhattan.

        Anyway, I find thinking about DC’s experience to be an interesting comparison. It’s sometimes hard to tell how much it’s distorted by the real estate and gentrification boom of the last two decades or so, but if you just look at the area around each new metro stop before it opened compared to today, the boom in transit oriented development looks astounding.

        It’s probably not the best example, because it’s probably more the gentrification and less the rail access, but the first time I got off the metro at U street, in 1999, there was Ben’s, the Reeve’s center, a few other businesses and a lot of rubble. Today is totally different.

  6. Ian Bicking

    Using straight density figures instead of population-weighted density makes this graph kind of useless for establishing who is underperforming or overperforming their density.

    For instance, looking at the table at the bottom of that article, Austin and Minneapolis have about the same transit usage, even though Minneapolis is 7k/sqmi and Austin is 2.7k/sqmi. But if you look at population-weighted density ( Minneapolis is 3.4k/sqmi and Austin is 3.1k/sqmi. So it’s not so surprising they are similar.

    CBD density seems like an odd figure too – simple CBD size (CBD jobs as a percentage of all jobs) would seem like a more useful figure. A city that has double the physical size CBD while maintaining the same relative number of jobs is in largely the same place as the tighter CBD, as the real question we’re asking is how well can a spoke system serve the metropolitan area.

    1. Charlie Gardner

      Hi Ian — I’m flattered to see this post linked here (thanks David!), but looking back at it also wondered why I hadn’t used weighted density. It wouldn’t have affected the overall correlation, I think, but as you say would have been more helpful in identifying which cities are the true overachievers. As it turns out, the weighted density data from the Census hadn’t quite been issued yet — Chris picked up on the release a couple months later. It may be time to revisit these figures in light of the new data to see how that affects the results for individual cities. I did, however, run the transit data against CBD jobs as a percentage of all metro jobs: that information is in the third graph in the linked post.

Comments are closed.