Five Takeaways From Ten Years of Minneapolis Planning Commission Approvals

Originally posted on Streets.MN.

Yesterday, I shared five takeaways from forty-one years of housing starts data in the Minneapolis-St. Paul metro region. This data, collected from the US Department of Housing and Urban Development, can be useful to get a look at how much housing is being built across a range of geographies, down to the municipality level.

But in order to answer more granular questions about what exactly is being built and where it’s being built within cities, you need a different source of data. For many years now, I’ve been regularly recording the work of the Minneapolis Planning Commission in order to get a better sense of what is happening project-by-project. The Planning Commission meets roughly twice a month (see their agendas here) and virtually all multi-family projects come through that body. The overwhelming majority are approved. The submissions are presented in a standard layout (in contrast, submissions are less accessible and the process is less standardized in St. Paul) that makes it easy to find key information. Specifically, for each approved project I’ve recorded the number of housing units proposed and the number of car and bicycle parking spaces attached to the housing. Since I started this record keeping, I’ve also gradually gone through agendas for past years and have compiled data going back to January of 2011.

As of the end of 2020, I had a spreadsheet with ten years of Minneapolis housing development data. You can check it out for yourself here. This is my work alone, and over time I have certainly made mistakes. I have also had to occasionally estimate numbers that were not explicitly stated, and so specific numbers should be treated as ballpark figures. (A more detailed explanation of methodology can be found at the very bottom of this article.)

Even so, I believe this data presents a perspective on development in Minneapolis that has not previously been public. Looking deeper than just the surface numbers, we can derive broader conclusions about housing in Minnesota’s largest and busiest urban center, and make inferences about how it may change in the future.

1: Within Minneapolis, there are enormous geographic disparities in where new housing is proposed.

Cumulative units approved per ward (2011-2020). Totals are all true to their boundaries as of this writing, which were redrawn after the 2010 census.

Cumulative units approved per ward (2011-2020). Totals are all true to their boundaries as of this writing, which were redrawn after the 2010 census.

Housing proposals surged in Minneapolis in end of the past decade, especially in 2018, 2019, and 2020. But while the city’s overall production numbers are impressive, they mask significant differences in where housing was located. Broadly stated, proposed housing projects were predominantly located in centrally-located, middle- or mixed-income areas of the city. The city’s peripheral, low- or high-income neighborhoods saw the least real estate interest.

Over 40% of housing units proposed in Minneapolis were proposed for Ward 3, the city’s superstar core. Comprising much of downtown, as well as the fast-densifying neighborhoods on the northeastern bank of the Mississippi River, Ward 3 is home to the city’s most expensive land and highest-demand neighborhoods. As a result, residents and developers are more likely to trade floor space for good location, leading to a high population density and big projects.

Wards 2, 7, and 10 all saw high totals of housing proposed, including a handful of towers. Much of the city’s growth of mid-rise apartment buildings were built in these wards, offering easy access to downtown and abundant amenities at prices made possible by “5 over 1" wood frame construction.

Six wards saw a medium amount of attention from housing developers. Wards 1, 5, 6, and 9 all boast a central location and good amenities, but historically were less wealthy than their counterparts in the higher grouping. That legacy of racial and economic segregation is still apparent today. Ward 12, by contrast, is not as close to downtown by physical distance, but it has seen a surge of proposals in the vicinity of Blue Line LRT stations, where access to the central city is as convenient and fast as in any of the closer-in neighborhoods.

Ward 13, the city’s wealthiest, also belongs to this middle-tier of wards. But its total requires a massive caveat. That’s because 57% of the housing units proposed in Ward 13 are contained in just a single project, the four-building Calhoun Tower mega-development next to the future West Lake LRT Station. This project has yet to break ground. Until then, the total of new units built in the most exclusive corner of the city will remain low, and no projects were proposed for this Ward in 2011, 2016, and 2020.

Three wards saw almost no development activity whatsoever. Ward 8, which includes the three street corners opposite the one where George Floyd was killed, has seen a little over three hundred units proposed, despite a great deal of housing proposed to the west, north, and east. To the south, however, Ward 11 has also seen very little interest. No housing projects were brought before the Planning Commission in Ward 11 in six of the last ten years, and overall it saw just 74 units proposed, the second lowest of any ward.

Only Ward 4, the north half of the northside, has been more ignored. Incredible, not a single housing projects was brought before the Planning Commission in Ward 4 in nine of the last ten years. The lone exception was in 2012, where the Commission approved a like-for-like replacement of three units in a building destroyed by fire.

2: Downtown Minneapolis is still humming along, but outside neighborhoods have seen a huge boom.

Cumulative number of units approved outside downtown and inside downtown (2011-2020).

Cumulative number of units approved outside downtown and inside downtown (2011-2020).

Another way of looking at the geographic distribution of housing proposals in Minneapolis is to split the city between downtown areas (defined by me as the areas within the freeway loop) and the rest. This division splits all of the downtown wards, and it helps to tell a slightly different story.

Year-over-year, the number of units proposed in the downtown core is surprisingly consistent. The long development process for the types of buildings proposed for this area may make them more resistant to small ups and downs of the market.

Outside of downtown is a different story. Coming out of the great recession, in 2011, fewer units were proposed outside of downtown than within it. In 2012 and 2013, buoyed a bit by student housing in the university area, proposals outside of downtown spiked, but quickly fell again. The two numbers then closely tracked each other for several years. But the explosion in proposals that has come in the past three years has been sparked by a huge surge in projects outside downtown. Perhaps the areas outside downtown are more resistant to economic conditions than initially appeared. In the teeth of the global pandemic and unrest across the city, 2020 saw a record number of proposed housing units outside of downtown.

It’s worth mentioning at this point that not everyone defines “downtown” the same way. The Minnepolis Downtown Council, for instance, includes part of the St. Anthony riverfront. Tenants of the recently-opened Expo tower are presumably considered to be among downtown’s over 53,000 residents, but tenants of the Rafter tower just a few blocks away are not. In my counting, both are outside downtown, and help contribute to those burgeoning numbers.

3. Within Minneapolis, less parking for cars is being built.

Ratio of car parking to housing units (2011-2020).

Ratio of car parking to housing units (2011-2020).

In 2015, the Minneapolis City Council passed a (then) landmark zoning reform which slashed or eliminated parking requirements for developments near high-frequency transit routes. Since then, many developments have taken advantage of these lowered requirements. Three years ago, (then) City Planning Commissioner Nick Magrino looked at the ratio of car parking spaces to housing units in proposed projects, and found that it had declined in the wake of the reforms.

After the three more years of proposals, that trend has held steady. Cumulatively, proposals that were approved in 2020 had a ratio of 0.76 car parking spaces to housing units, tied with 2015 for the lowest of the decade. But this result was in fact far more impressive, because the overall number of units and projects proposed was well-over double. In 2015, a single project, the no-parking City Club Apartments, brought that year’s ratio down a full eight points (Interestingly, 2016 would’ve had the lowest parking ratio of any year, but for the whopping Legacy condominium project, which boosted its ratio by fourteen points). In 2020, the low parking ratio was achieved across 70 separate projects. Net, 1,631 housing units were proposed without corresponding parking spaces in 2020, likely a record for the city since structured parking was invented.

In 2021, the City Council is poised to eliminate parking requirements entirely. It is hard to guess what the effect will be. Many current projects still propose more parking than is required, and projects that request variances for parking are almost always granted them. At the same time, having no requirements whatsoever may give some developers even greater license to push the envelope. The coming explosion of new rapid transit lines in the city and an increasing number of high-quality protected bikeways on critical corridors will make low parking projects easier to sell to prospective residents and the banks that fund their construction. I wouldn’t expect this ratio to plunge, but I would expect it to continue a slow, measured descent in coming years.

4. Smaller projects tend to have lower parking ratios, while the biggest projects vary wildly.

Every development proposal in Minneapolis from 2011-2020, plotted with housing units on the X axis and car parking spaces on the Y axis. Significant developments are specifically noted. Full disclosure, one development, “The Legacy” condominium proj…

Every development proposal in Minneapolis from 2011-2020, plotted with housing units on the X axis and car parking spaces on the Y axis. Significant developments are specifically noted. Full disclosure, one development, “The Legacy” condominium project near Gold Medal Park, would’ve broken the aesthetics of the graph if it had been included. It has 374 units of housing and 686 car parking spaces.

When plotted all together, on a graph of housing units and parking ratios, a few things become apparent. The most noticeable is that projects with lower unit counts tend to have lower parking ratios. This isn’t surprising, the bigger the project, the more economical it might be to stick a bunch of structured parking underneath your building.

What this graph helps visualize is where the breakeven points might be. Up until about eighty units, parking ratios are almost always extremely low. From 80 to 160 units (unless you are building an ultra-luxury skyscraper), a majority of projects still appear to feature low parking ratios, but there are more exceptions. Once above 160 units, things appear to be more of a wash, and much more developer specific.

For instance, Doran Companies’ low-rise proposals for two full-block lots near the Stone Arch Bridge (both featuring a building design known infamously as the “Dallas Donut”) are some of the most car-centric in the city. Despite being connected to downtown by the state’s best walking and biking path, and steps from the future route of the E Line aBRT, this mega-development will construct over 1.44 parking spaces per housing unit.

In contrast, a collection of mid-rise buildings planned for the area adjacent to the East Lake Station by Saturday Properties will include 432 units of housing and just 271 car parking spaces, a ratio of just under 0.63 spaces per unit. Combined with the first phase building on the site, the overall parking ratio of the development will be roughly 0.67.

Differences of this scale are reflective of the mix of units in each project and their targeted demographics. But they also seem to represent fundamentally different perspectives by each developer and their investors as to what kind of city Minneapolis is, or should become.

You can learn a bit more from this graph by viewing it in GIF form. This helps dramatize the explosion of proposals that the city has seen starting in 2018, as well as a subtle shift in attitudes towards lower parking developments over time.

5. Minneapolis is sitting on a massive backlog of work.

Despite a year of unprecedented disruptions, from the COVID pandemic to the civil unrest following the killing of George Floyd, the pace of housing project proposals in Minneapolis only increased in 2020. On one level, that fact is strong proof that people still want to live in the City of Lakes, and that big developers and financial institutions believe that demand will continue.

But there is also some evidence of a little bit of hedging. Minneapolis is sitting on an unprecedented backlog at the moment. Fifteen buildings or planned developments of over 200 proposed units have yet to break ground (although two are reportedly about to do so). Several might be dead, after a significant amount of time without movement. At least one new project will soon join the list.

Caveats aside, as it stands today, a massive 5,315 units of housing—more than Minneapolis has built in any of the past forty plus years—have been approved on paper but have yet to begin appearing in reality. Here’s what that list looks like:

Biggest Projects Approved, But Yet To Start Construction

  1. Calhoun Tower PUD — 774 units

  2. Expo Low Rise PUD — 574 units

  3. 315 Nicollet — 369 units

  4. 200 Central — 360 units

  5. 12th St Apartments — 357 units [Probably dead?]

  6. North Loop Green — 356 units [Possibly jeopardized by a debate over short-term-rental units, which I am not counting in this total.]

  7. Malcolm Yards PUD — 348 [Work has started on part of this project, but only on renovating an existing building on the site.]

  8. Block One — 342 units

  9. 240 Portland — 318 units

  10. Dinkytown McDonalds — 308 units

  11. Art and Architecture — 274 units [Probably dead?]

  12. Vison Loss Resources — 256 units

  13. Thrivent Block Phase II — 240 units [Reportedly about to break ground]

  14. HenCen Apartments — 229 units [Reportedly about to break ground]

  15. Snelling Yards — 210 units

As vaccinations ramp up across the country and the grip of the pandemic hopefully eases, we may start to see more of these stalled projects move forward. However, it seems especially foolish, after the year we’ve had, to predict the future with any confidence. At this point, it seems safe only to say that there’s a huge amount of more housing that’s planned, and a huge amount of untapped potential that still sits on drawing boards, or in the heads of people like the readers of this article. Will it come to fruition?

Stay tuned for the next installment of this article in ten years!

***

Notes on methodology

Here’s how I collect this data and where it might not tell the full story or be inaccurate—

Housing Units: The easiest number to find is the unit count, which is often contained in the project description. If there are discrepancies, I dig a bit deeper to find which number is the correct number, and which is the typo. I track units in order to be consistent, not bedrooms. This sometimes means that projects (especially in the University area) get listed as having a small number of units, even though they will be home to a lot of people. I only count the number of units in the new development, not the net number of units overall. Sometimes a project will include the demolition of a number of homes in existing buildings, and that is not reflected in the data.

Car Parking Spaces: Often, the development packet will list the number of car parking spaces right alongside the number of housing units, and that usually makes my job easy. But mixed-use buildings present complications. Because my spreadsheet is focused on housing development, I’ve chosen to count the number of car parking spaces assigned to residential units. If a building has 100 car parking spaces, of which 50 are reserved for residents and 50 are reserved for patrons of the building’s retail space, that project goes into my spreadsheet as having 50 car parking spaces.

Sometimes, the development packet does not clearly assign car parking spaces across the various uses. The highly unscientific approach that I take in these cases is to determine the percentage of the total number of parking spaces required by the zoning code for residential use. If a building proposes 100 car parking spaces, and the zoning code says that 150 car parking spaces are required, with 75 required for the building’s residences, then I take that ratio (50%) and apply it to what is proposed. Such a project would go into my spreadsheet as having 50 car parking spaces. As a transportation planner, I know full well that parking requirements are total pseudoscience. But I don’t have a better or more consistent metric.

One final complication can come with projects that plan to use existing parking structures or lots. For instance, the proposed renovation of the Strutwear Building just south of US Bank Stadium, would represent a net gain of 179 housing units to the city, with the construction of no additional parking. But because the project intends to use 286 car parking spaces in the adjacent ramp, the residences in this building could in fact be quite car-dependent. Again, because I am interested in the relationship of housing to car parking, and not in compiling a master list of every net change of housing or parking in the city, I choose in these cases to list the project as having as many parking spaces as are intended to be reserved for the residents (286 in the Strutwear example).

Bike Parking Spaces: I haven’t mentioned bike parking spaces in this article, because I’m not sure there’s as much to learn from those numbers, and they tend to be swayed massively by how much student housing was built that year. But I do count these in my spreadsheet. In cases when “long-term” and “short-term” bicycle parking are separated, I count only the long-term parking.