The Effects of Upzoning American Cities: A Difference in Differences Study of Minneapolis-St. Paul and Kansas City



With trends of US cities passing zoning reforms to allow for high density to be constructed in residential areas, this paper seeks to quantify the economic effects which such reforms would have in the respective areas through a difference in differences analysis of the Minneapolis-St. Paul and Kansas City metropolitan areas. Through the difference in differences analysis used, which allows for comparisons of the changing of Minneapolis to the unchanging nature of Kansas City, of building permits, the consumer price index for housing price, the number of retail workers, the paper concludes that the increased housing density allowed through the zoning reforms of Minneapolis-St. Paul had a positive impact on the region’s economy; however, despite the shown benefits, the magnitude which Minneapolis-St.Paul experienced those benefits were almost negligible. The lack of notable change found in this paper suggests two things. Firstly, the paper is a very early analysis of Minneapolis’s zoning reforms, which might need more time to fully come to fruition. Secondly, a lack of change also suggests that more aggressive reforms and supplemental policies should be pursued not only by Minneapolis, but also the suburban cities in the metropolitan area.


Historical Background of Low-Density Zoning

As transportation technology became better and more affordable over time, people and economic activity in the United States would continue to move out of the traditional central cities and into the surrounding, less dense lands around it, forming what would eventually be the modern-day United States suburbs. According to Fischel, an important thing to note, when analyzing the history of zoning, is that, in most cases, a homeowner’s largest financial asset, by far, is the value of their home1. During the age of streetcar dominance, though no zoning was enacted, the suburbs of many cities, such as Boston and New Haven, had single-family dominated tracts of land. The reason behind how this uniformity was achieved stems from the streetcar’s integral role in suburbs at that time. Streetcars were not able to carry heavy industries, therefore it remained in city cores by railways, and apartments, or other forms of multi-family housing, were built next to streetcar lines to take full advantage of them. To avoid apartments, all developers needed to do was to move their development away from streetcar lines. The uniformity of these suburbs were then maintained through political influence of the developers and home-owners to prevent the construction of streetcar lines in their area and through informal agreements and mutual understandings.

As transportation technology, such as motor buses and trucks, allowed poorer city residents and industrial businesses to further decentralize into the suburbs, the integrity of the single-family suburbs were put in jeopardy.

In response to the feared possibility of undesirable buildings or industries moving their way into these suburbs, developers started to push for zoning to replace their insufficient model of formal agreements.With home prices being heavily influenced by the characteristics of the area in which it is located and with population growth being perceived to bring a lower quality of life, homeowners began to dominate local politics and zoning to protect the values of their properties, thus solidifying the dominance of exclusive, low-density zoning in American cities2,3.

The Movements in America Against Low-Density Zoning

Despite the established dominance of single-family zoning across American cities, recently, in the wake of rising housing prices and environmental concerns, several US cities and states have passed laws and zoning codes to increase zoning.

In the pursuit of the Minneapolis 2040 plan, the city of Minneapolis in 2018 effectively banned single-family zoning in their city limits by allowing the construction of duplexes and triplexes across the city4. Following this example, several other US governments have passed similar legislation. The most notable of which is the state of California’s Senate Bill 9 which prohibited single-family zoning across all of their cities, and one of the most recent developments Arlington, Virginia’s banning of single-family zoning in 20235,6.

CityDate Reform Was ApprovedReform Information
Arlington, VA3/22/2023Allowed buildings of 2-4 units on prior single-family zones, reduced residential parking requirements, allowed more construction of residential buildings of 5-6 units.
Anchorage, AK11/22/2022Removed parking minimums across the city with the intention of facilitating multi-family housing construction and reusing vacant lots.
San Francisco10/28/2022Allowed fourplexes to be built on single-family land and six-unit buildings on corners
Portland, OR6/01/2022Allowed duplex construction on single-family land. Also expanded the areas where fourplexes and detached homes could be built.
Charlotte, NC6/21/2021Allowed duplex and triplex construction on single-family lands.
Table 01: Examples of Recent Movements in US Cities to Promote Upzoning7

Research Question

What are the economic implications of US cities shifting their zoning policies away from the traditional single-family zoning to zoning practices more inclusive to denser development? With the developments of certain American jurisdictions moving towards a more friendly stance on denser and more inclusive zoning practices (upzoning), it becomes important to quantify the effectiveness and impact of these new policies. Though many of these developments are still new or ongoing, this paper aims to predict the economic impact of jurisdictions embracing upzoning through a difference in differences analysis of data between US cities who’ve already established more inclusive zoning policies and those who have not.

The most predictable change would be in housing. As stated previously, with barriers lifted, assuming no other factors are changed, an increase in housing units built, specifically multi-family units, should be seen after the enactment of the policy.

With an increase in housing supply, since there will be more competition in the housing market, the cost of housing in the treatment metropolitan area should decrease when compared to the control metropolitan area, regardless of the presence or absence of specific policies to encourage low-income housing development.

Because of a greater concentration of housing, more residents will be living closer to commercial and recreational areas. With access to these areas improved for more potential customers, commercial retail in upzoned areas should experience an increased activity. Thus, the need for retail employees to keep up with the demand, even if the zoning is solely for residential uses.

Literature Review of Upzoning

Increases in Housing Supply

The primary purpose of implementing new, denser zoning codes is that there is an assumption that upzoning would increase the housing supply of a city, which would then theoretically make housing more affordable in the area. The theory behind this assumption is based on the simple supply vs demand concept. Continuously, most of the United States metropolitan areas have been continually growing in size. This would increase the demand for housing units. In a perfect market economy, as the demand for housing increases, developers would meet the new demand for housing through the construction of new property, using a combination of building housing on new lands and building denser buildings on already existing residential lands. However, due to the strict nature of zoning laws across US metropolitan areas, in most areas developers are unable to build denser housing or buildings. By constricting the supply of housing units, the growth in demand for housing would outpace the growth in supply, causing increases in housing prices. However, if restrictive zoning laws are to be blamed for a lack of housing, then if the zoning codes of cities were made to be more accommodating to denser housing, then developers would in theory take advantage of said changes and builder denser housing in existing residential areas, thus increasing the supply of housing and alleviating the increasing prices of housing.

Existing literature seems to support the validity of this theory. In the study by Fan et al.8, several simulations were conducted to estimate the housing market in Seattle in 2050 with and without zoning policies that allowed for denser residential development. Based on their simulations, though both zoning scenarios would see a less affordable housing market in 2050 as compared to the start date of 2014, the scenario including denser zoning saw cheaper housing prices, mortgages, and rent payments than the scenario without. Though this study does visualize the impact which zoning has on the housing market and economy as a whole, the nature of the study hinders its external applications. Though case studies can be conclusive of large ideas, the simulated data from this study is not of the same quality as real data. When building the parameters of the simulation, seemingly minute details could be left out which would have had a noticeable impact. More importantly, however, human interaction, and therefore the market, isn’t 100\% logical like computer software. Though the study can be helpful in understanding the implications of different zoning practices, real-world data which is affected by the imperfectness of the real market would be needed to back up the claims of the study.

Another study by Li9 also suggests that the building of more housing units increases housing affordability in an area. The paper was written in an effort to analyze a long-standing criticism against density: increased density would attract more high-income earners and amenities which would thus make rents higher than before. The paper used data from New York City neighborhoods before and after construction of new high-rise buildings. The findings of the paper suggested that, though the new, denser buildings did attract more high-income residents and amenities, the increase in housing supply around the new high-rises still outweighed those effects, causing net decrease in rent around new high-rise buildings.

Furthermore, it has been found that restrictive zoning has contributed to gentrification. As land values naturally increase in urban areas, developers will look to improve the value of existing buildings on said land. If existing zoning didn’t limit density, developers could best utilize the increase in land valuation through the building of more, smaller housing units. However, because most residential land in the United States only allows for single-family zoning, such developments are impossible. Instead, unable to build smaller, more affordable units, developers convert existing single-family houses into larger, more expensive single-family housing to capitalize on the increases in land value10. Thus, not only does restrictive zoning fail to meet market demands for housing, it also contributes to gentrification of previously developed lands.

Improvement in Walking and Cycling

Another benefit of upzoning is that by building denser development, traveling in cities becomes more accessible to walking and cycling. Density is a large variable in promoting alternative modes of transportation for shorter commutes. Because of the lower density typically associated with American zoning, many residents are prevented from living within the ideal distance for walking or cycling of areas of interest (i.e. commercial buildings, social gatherings, places of work)11. Thus, it would be expected that if the upzoning of a city or area were to be effective, then one would be able to see more walking and the benefits of it in the community.

The most obvious economic benefit for promoting walking and cycling is that they are a more effective method of traveling short distances. Not only do these methods of traveling have health benefits and do not have the fuel consumption and emissions as automobiles, their monetary costs are far lower than that of driving12. If cities were to increase density and encourage walking and cycling, then the average resident should be able to save money from their transportational needs and spend such money on other wants.

Besides indirectly benefiting businesses by saving consumers money from commuting to potentially spend on other goods and services, increased walking and cycling also has direct benefits to the operations of local businesses. Case studies on New York City and San Francisco have found that businesses have benefited from increased pedestrian and cycling activities due to the fact that pedestrians and cyclists make more frequent trips, totally to a higher monthly expenditure than their motorist counterparts13.

The benefits of increased pedestrian and cycling is well known. Central to their promotion is an increase in density, allowing for more trips to be done at the ideal distance for walking and cycling. However, a five-minute walk in a city might not be ideal, even if the distance suggests so, if there is no proper infrastructure making the walker or cyclist feel safe from ongoing motor traffic. Therefore, in the analysis of the economic progress of American cities who have increased their density through upzoning, it is important to understand the effectiveness which jurisdictions have utilized their newfound density in providing the necessary supplements to foster more pedestrian and cycling activities so that they can reap the full benefits of it.

Other Potential Economic Benefits

Though its effectiveness will not be analyzed in this paper, it is also important to acknowledge that by increasing residential density, public transportation becomes a more viable option for commuters. Like walking and cycling, public transportation usage requires a minimum density for the service to be viable as commuters usually still have to walk or cycle from transit spots and their desired destinations14. Thus, the lack of density, ubiquitous in most American cities, is a large factor in the country’s seemingly underdeveloped transit network. Public transportation was found to have similar economic benefits in the forms of increased consumer spending and cheaper transportation costs as increased walking and cycling15. However, due to the nature of establishing public transportation lines and the novelty of most US cities adapting more inclusive zoning practices, the data for such analysis would be nearly impractical. Whereas cycling and walking are affected by public opinion and government spending on infrastructure, public transit is nearly wholly dependent on such. With the much greater cost of expanding transit networks and the limited time which much of these new zoning policies have had to take root, the facet of the benefits of upzoning in the US remains an issue which needs more time to address.

Another acknowledged benefit found in studies, although it also won’t be analyzed in this paper, is the shorter communication distance when cities are built to a greater density is that workers and firms become more productive. Even with the popularization of video conferences in professional workspaces, the shorter travel distances found in denser cities have been found to result in an increase in worker wage, due impart to greater work experiences16. However, the productivity spillover found in cities requires extensive density as, according to various studies, the productivity outside of a five-mile radius of a firm experiences a sharp decline; likewise, the same could be found between areas within one mile of a firm and outside one mile17.

Previous Reforms

Though these reforms that allow for multi-family construction on all single-family zoned areas has been a relatively new trend, previous zoning reforms have been enacted in American cities and states that aimed to make housing more affordable to residents though the results have been largely mixed. From these past examples, an idea, though data would still be needed, could be drawn from the effectiveness of these zoning reforms.

For example, Seattle’s Mandatory Housing Affordability Program (MHA) upzoned 33 neighborhoods from 2017-2019. However, a fundamental flaw in the program, and why it can’t be an apples to apples comparison to the current wave of zoning reforms is that the MHA specifically upzoned already dense neighborhoods in the attempt to preserve the existing urban environment. Thus, the cost for developers prevented the program from making significant progress in affordable housing construction18. The shortcomings of Seattle’s MHA shows the importance of this new wave of zoning reforms. By now targeting less dense neighborhoods across cities, the cost for developers to construct denser housing will be significantly cheaper, allowing for more widespread effects.

Another previous plan was enacted by the state of California in the state’s Regional Housing Needs Assessment process (RHNA). From the RHNA, the state required localities to develop plans on how additional housing would be constructed to meet the demands of various income-groups across the state. However, with the plan originating from the state, many cities have found ways to not meet the housing targets set by the RHNA plans; thus, the RHNA was deemed as a substantial, yet incomplete, success19. The pitfall of the state’s RHNA stems mainly from local cities and governments not willing to change its existing zoning. However, since much of the analyzed reforms were passed and enacted by cities themselves, this issue shouldn’t be as widespread. However, specific neighborhoods, especially more affluent ones, might still be aggressive towards denser housing being constructed in specific areas.

The most akin example to the current wave of upzoning is in the New York City suburb of Ramapo, NY. Since 1986, when the city first introduced extensive upzoning throughout the city, the city, extremely progressive in regards to density for an American city, used two very notable zoning districts for much of its multi-family zoning: R-15A and R-15C. The much more accommodating district, R-15C, allowed for four to six units per lot and eight to twelve units per double lot; in contrast, the less accommodating district, R-15A, only allowed for accessory dwellings. Thus, as a result, the R-15C district experienced much greater development and infill than the R-15A20. The success of the R-15C zones could be indicative of more aggressive upzoning being necessary, especially in the future, for cities to better realize the economic benefits from denser housing.

Potential Consequences of Increasing Density

Although existing literature is largely in favor of denser development in cities, there are proposed and seen consequences from cities increasing their density.

The most applicable problems that could arise could be seen in Ramapo, NY. As multi-family housing became more widespread throughout the city due to zoning reforms, the city has experienced growing concerns about the ability of its infrastructure to handle the increased density. Although it was found that traffic largely didn’t worsen, the city’s water and sewage systems have struggled to uphold the increased demands from the city21. This strain on a city’s infrastructure and the cost to upgrade such infrastructure to handle the increased demand will be a natural trade-off in the process of upzoning, and, if the economic benefits outweigh the costs, this shouldn’t be much of an issue.

Meanwhile, papers on urban pathologies have had mixed results on the correlation between urban density and health and crime. On one hand, studies have found that there is no significant correlation between criminal activities, juvenile delinquency, and mortality if all other factors are considered22. On the other hand, another study, which analyzed data from Dutch cities, has found that urban density has contributed to increased male heart disease, mortality, and criminal activity23. As the American cities implementing zoning changes become more densely populated, it would be of interest to analyze such effects. For now, the inconclusive nature of this problem should be approached with caution.

Gaps in the Literature

Although there has been plenty of research done to support the benefits of denser development, there isn’t much research on how the widely accepted benefits translate when a city, specifically an American city, decides to adopt a widespread change in its zoning in the efforts to increase its density across the city.

Furthermore, there has been plenty of research done on the effectiveness of various American cities implementing various programs and changes in zoning codes to increase housing construction and density; however, these studies are usually just focused on the immediate impact of the zoning changes: housing construction and housing affordability. There isn’t much analyzing the indirect impact of a city increasing its housing density. Also, due to the recentness of American cities effectively phasing out single-family zoning, there isn’t much precedent or knowledge on the effectiveness which these policies have.

Thus, this paper seeks to take an early analysis of the current trends amongst the zoning of American cities. By analyzing the economic impact, directly or indirectly from reforming single-family zoning, this paper should serve as a basis for future investigations.


Difference in Differences

In order to measure the potential benefits of American cities taking policies to increase zoning density, a difference in differences (DID) analysis will be used to quantify the influence of such policies on the economy of the samples. Because this paper sets out to investigate the change brought out by a single policy, the DID method would best suit the analysis of the data. Since the research question asks the effects of enacting zoning reforms, the comparison between a city which did enact reforms to a city which hasn’t suits the nature of the question.

To conduct the analysis, two metropolitan areas with parallel trends in data before the enacting of the zoning reforms in one area are chosen to be compared. One of the metropolitan areas will have had policies and zoning ordinances passed while the other continues on the same trajectory since the start of the analyzed period. Furthermore, ideally, the metropolitan areas would be of a sufficient distance geographically and economically that the growth of one area does not have a spillover effect which would impact the growth of the other area and vice versa.

For the difference in differences analysis, the four data values collected could be broken down into two different categories: before and after the time the policy took effect (time value zero or one) and if the data came from the treated area or the control area (treatment value zero or one). The values of each point could thus be described using the following equations.

Time value  =  0Time value = 1
Treatment value = 0Y=\beta_1Y=\beta_1 + \beta_3
Treatment value = 1Y=\beta_1 + \beta_2Y=\beta_1 + \beta_2 + \beta_3 + \beta_4
Table 01: The Equations of a Difference in Differences Analysis

In the equations, the \gamma value would be representative of the collected data. \beta_1 would indicate the starting value of the control area (in this case, it would be building permits, housing costs, and retail employees).

Although the metropolitan areas should exhibit parallel trends prior to the zoning reforms, the data most likely won’t be identical. Thus, to factor in the starting differences between subjects,\beta_2 would be the value used for factoring in the starting differences between the control and the treatment areas. Thus, the \beta_2 value accounts for potential autocorrelation concerns that might have arisen, a critical tool found in a DID analysis.

\beta_3 indicates the change which the control group had during the analyzed time; Because of the established parallel trends of the two subjects prior to the zoning reforms, it is assumed in a DID analysis that the two subjects would still have parallel trends if it were not for the analyzed change. Thus, the value is also the assumed change for the treatment area had the policy not been enacted and an estimate of the economic state which the treatment group would be in had it not pursued those policies could be estimated through adding the \beta_2 value to the existing equation for the treatment group at time value 0.

Finally, \beta_4 is the most important value in the equations as it answers the research question in hand: how much did the more density-friendly zoning practices have in shaping the economy of the region? By taking the difference of the starting values, \beta_1 and \beta_2, and the expected change had the reforms never been approved, \beta_3, the remaining value of \beta_4 would thus be the assumed change of the tested metropolitan area due solely to the changes in zoning (with the assumption that all other factors were held equal during the time frame). \beta_4 is the difference in differences value and is ultimately the value to be mostly looked at across all the different types of data? Thus, the \beta_4 value is ultimately the value of interest. This value will be used to determine the effectiveness of the zoning changes enacted in the tested metropolitan area

Minneapolis-St. Paul , MN-WI vs Kansas City, MO-KS

To perform the differences in differences analysis, the Minneapolis-St. Paul metropolitan area was selected as the treatment area and the Kansas City metropolitan area was selected as the control area.

The Minneapolis-St. Paul area was selected in large part due to the Minneapolis 2040 plan discussed above. It is one of the most sweeping acts passed in a US city in regards to increasing city-wide zoning density with enough time passed for adequate data collection. Thus, the year 2018 would be the benchmark for the time periods analyzed in the study as that was the year the city-wide zoning reform allowing duplexes and triplexes to be built around the city’s residential areas was passed. St. Paul would also pass a zoning reform dealing with accessory dwelling units in January, 2022; however, the time at which it was passed means that its effects can’t be analyzed as much24.

The Kansas City metropolitan area was selected due to its various similarities between it and the Minneapolis-St. Paul metropolitan area. Both areas are located in the Midwest, having a large presence in their respective areas in both population and economy. Furthermore, both areas are experiencing fast growth in both their population and economy. As will be explored later in the paper, the parallel trends found between the data of the two metropolitan areas would corroborate the merit of having Kansas city as the treatment metropolitan area.

Finally, the metropolitan areas of Minneapolis-St.Paul and Kansas City are not located too close to each other where growth in one region would have a direct effect on the growth of the other region. Therefore, any economic changes occurring in the Twin Cities should not have influence on the economy of Kansas City and vice versa. Thus, with sufficient separation and parallel trends, Minneapolis-St. Paul and Kansas City are good candidates for a DID analysis and the finds from the analysis.

Data Availability

To ensure the accuracy of the data, all of the following data used in the calculations were released by the US Census. However, because the US Census generally releases data for metropolitan areas and not individual cities, areas of the analyzed metropolitan areas, Minneapolis-St. Paul, will have a significant portion of their data from cities which haven’t adopted the zoning reforms of interest. All of the data has not been cleaned, instead it was taken as is.

A couple of potentially useful statistics which could have been used for the study were unable to be applied due to the lack of availability across the desired time period of analysis. Estimates of average consumer spending on retail and dining over time in the metropolitan areas could not be used due to the lack of resources of the study and a substitute had to be found. Another limitation would be that the releases of the consumer price index (CPI) for rent in the Kansas City metropolitan area would be discontinued after 201725.

Building Permits: Though city governments can allow more density and development to occur through changes in their zoning policies, it is ultimately the private developer, meeting the demands of the market, to capitalize on such changes. Thus, it is appropriate to analyze just how much new development actually occurred in the markets.

To begin construction, projects must get permits from the government. Annual building permits are provided by the US Census21. Because of the nature of this data, being case-by-case, this data will be very accurate in determining the immediate effectiveness of the zoning reform’s paramount goal, to increase multi-family housing construction. The Census breaks down the amount of single-family units, 2 family units, 3-4 family units, and 5+ family units which are permitted each year for the metropolitan areas. The building permits for the four years preceding and following 2018 would be used with both time periods having the total number of each unit type summed up.

House Price Index: Consumer price index is a calculated value used to find the average cost or expensiveness of a product in a certain market. As the CPI value increases, the correlating price of a product in the market is more expensive. The US Census gives already calculated CPI values for various goods and services in different metropolitan areas.

Though the CPI for rent was discontinued in 2017 for Kansas City, the CPI for house price would be available by quarter for the four years preceding and following 201822,23. To analyze the data, the average of the CPI values released for the intended group and time period would be used for the calculation of the difference in differences analysis.

Fig 1: Housing CPI of Minneapolis-St. Paul and Kansas City prior to 2018. Although the CPI values between the two metropolitan areas are different before the zoning reforms, the change between the two CPIs are nearly identical, indicating parallel trends. The initial discrepancies are accounted for in the \beta_2 value. The expected CPI for Minneapolis-St.Paul took the starting CPI value after the zoning reforms and added the \beta_3 value to the end date. Note that the graph shows the average rate of change for the expected value, which doesn’t account for the market fluctuations actually experienced by the real CPI values.

Number of Retail Employees: With estimates on consumer spendings unavailable, the number of retail employees in the metropolitan areas would be the next-best value to analyze. Though also largely influenced by a myriad of other factors, if the need for retail labor is high amongst an area, then it could be indicative of a successful retail industry in the location. Thus, the average number of employees, by the thousands, for the four years preceding and following 2018 will be used in the calculations. The data collected for this factor was from the US Bureau of Labor Statistics’s count of retail workers in general merchandise stores, to ensure the most accurate data as possible24,25.

Fig: 02 Number of Retail Employees in General Merchandise prior to 2018. The fluctuations in retail employees is largely mirrored between the two metropolitan areas before 2018, suggesting parallel trends.

Results and Analysis

Housing TypeDifference in Differences Value
1 Unit Housing7, 326 units
2 Unit Housing-158 units
3-4 Unit Housing226 units
5+ Unit Housing29,351 units
Table 03: Difference in Differences Results for Each Housing Type
Statistic AnalyzedDifference in Differences Value
Housing Consumer Price Index-6.3275
Number of Retail Employees (Thousands)0.317
Table 04: Difference in Differences Results for Housing Consumer Price Index and Retail Employees

Building Permits Findings

Though the difference in differences for single-family permits shows that the Twin Cities outpaces the Kansas City, which is impart due to over 10,000 single-family units being built after 2018 than before, looking at the single-family permits for the two largest, most opened to reform cities, Minneapolis and St.Paul, their total single-family permits saw a drop, suggesting that the increase was coming from communities outside of the zoning reforms. This shows that, for an entire region to fully realize the effects of upzoning, a wider adoption of zoning reforms would be needed so that the change is not just limited to select areas.

The most noticeable data point from the differences in differences analysis is the negative value associated with duplex construction. Going against the hypothesized increase of construction due to barriers to construction being lifted, such an unexpected change would have be due to a combination of the following reasons: the demand for duplexes didn’t incentive developers to construct more duplexes in Minneapolis-St.Paul as expected; with the zoning reforms happening so recently, developers are not able to capitalize on such changes yet; the inclusion of suburban cities outside of Minneapolis in the analysis (though, with Minneapolis’s size and footprint, even if just it was the only city to change its zoning, the difference in differences shouldn’t be negative).

Similarly to the construction of duplexes, the change in triplex and four-unit housing is also nearly negligible. Though this initial data suggests that the changes in zoning did not provide enough incentive by itself for more “middle housing” to be constructed, this data should still be reviewed as the policy gets more time to take effect.

The most notable change brought after the zoning reforms was the construction of 5+ unit housing. Although the bulk of the Minneapolis 2040 plan addressed construction of smaller, multi-family housing units, this denser form of housing saw the greatest increase in construction. Though not the main focus of the plan, denser housing, even at this high level, still coincides with one of its large ultimate goals, to provide more housing at a high density. Thus, with this attitude, the change in the way Minneapolis approached zoning and housing density did have a positive impact on the overall housing construction of the city. However, for the plan to fully reach its ultimate goal, more time and participation from other cities will be needed to determine its overall effectiveness properly.

House Price Index Findings

The negative value for the difference in differences for the housing price index of Minneapolis-St. Paul compared to the control area indicates that the zoning reforms did have its intended effect. However, with a value of -6.3275, it could be argued that the changes brought about by the current zoning reforms were not widespread enough to have had a significant impact on the overall increase in housing costs. For context, using the average yearly CPI for house prices in Minneapolis-St. Paul during the time period used for the initial time period of the difference in differences analysis, the average annual change in CPI would be 10.685. With the effects of the zoning reforms not even negating a year of rising costs, it could be argued that the changes were in the right direction but too little and not widespread enough.

As stated above in the analysis of building permits, the effects of the Minneapolis 2040 plan might still need time to be fully realized. If more duplexes and triplexes are built into the market in the future, the difference in differences value could show a great magnitude of change.

Another thing to note about the Minneapolis 2040 plan, the upzoning of neighborhoods that comes with the plan is very general, all previously single-family zoned areas are allowed to have duplexes and triplexes constructed on. In that specific part of the policy, there is little discrimination between neighborhoods and housing price ranges. Therefore, developers, to maximize profits, could specifically choose to build denser zoning in wealthier neighborhoods for wealthier clients. If this is the case, then the overall affordability of the city wouldn’t be lowered as much as if there was an emphasis on low and middle income housing.

Despite the small impact to lowering housing costs, the zoning reforms nonetheless still made the housing to be built more than what it would have been if the zoning reforms had not been implemented. Because of this, housing is more accessible to home-buyers in the Minneapolis-St. Paul metropolitan area; however, more density-encouraging policy should be explored for not just the largest cities of the metropolitan area, but also the suburbs with much less density to begin with.

Retail Employees Findings

The tested area retaining more than what it would have if not for the zoning reforms potentially indicates a healthier retail environment for businesses. Though not a perfect measurement, more employees retained suggests a larger demand for labor in the retail industry, signifying more business activity in the test area. However, like most of the previously analyzed data, the magnitude of which the upzoning affected retail is still nearly negligible. However, if given more resources, a different measurement besides number of employees should be used to gauge the change in retail, because, just looking at the number of employees does not factor in long working hours, increased productivity, or automation which could all demand less additional workers to meet increased demands.

The Economic Implications of Upzoning

Though the data from the study indicates that the theories for the potential benefits of upzoning do exist when implemented, the small magnitude of indicated change brought about by the zoning reforms suggest that there was little impact in such directions. The underlying problem faced by the Minneapolis-St.Paul metropolitan area in regards to the limited change is that the policies of changing zoning changes to incorporate more density is not widespread and developed enough.

American metropolitan areas are almost all fragmented by city jurisdictions. Although Minneapolis is the largest city in the metropolitan area, it still only makes up a fraction of its population. Most of the Minneapolis-St. Paul metropolitan area, like most American metropolitan areas, has the majority of its population living in various suburban cities. Though the largest cities made specific changes to their zoning codes to allow for more density (St. Paul’s only reform happening in 2022 should be noted), the suburban communities which make up the majority of the land and population of the area still have yet to make such reforms. Thus, while the policies of Minneapolis seem to be pushing the metropolitan area in a positive direction, further action must be taken for the benefits of upzoning to be fully realized.

The suburban cities and communities of Minneapolis-St. Paul, or any metropolitan area in the process of increasing density, needs to have concurrent reforms and upzoning with the largest cities for the urban density to increase across the metropolitan area instead of a limited, contained city limit.

However, even with the limited changes brought about by these reforms, it was still found that Minneapolis-St. Paul still had an overall increase in density. As density increases in a city or metropolitan area, the aforementioned economic benefits would naturally occur due to an increased number of housing closer to areas of work and retail; however, specific policies and projects could enhance such benefits. Most notably, a transition between solely residential zoning to mixed zoning (having both residential and commercial uses allowed in one area) could be of benefit. This way, travel distances are even shorter, allowing for more frequent pedestrian and cyclist interactions with commerce.

Other projects could further promote walking and cycling as viable transportational options with the newfound density. Designated infrastructure, proper bike lanes and ample sidewalks, would further ensure that these alternative methods are fully realized with the upzoning; although, with the recency of these zoning reforms and their effects themselves potentially not fully realized, it is understandable if these policies and projects take a few more years to be set in motion.


Data Availability

The largest limitation for the paper is the lack of data availability on multiple levels which could have enhanced the findings. The easiest issue in regards to data availability to solve is the number of metropolitan areas which have enough data for a proper difference in differences analysis. Since the movement of upzoning has been relatively recent in the United States, the number of major cities with a proper city-wide ordinance to upzone their residential communities with ample time to analyze the effects of such decisions is fairly limited. This issue would be resolved with the passage of time as more time would allow more cities to have sizable time periods for data analysis.

Other issues come from the way data from the US Census and Bureau of Labor is presented. Much of the data available, and all of the data which was analyzed in this study besides housing permits, did not have city-wide statistics. Because of this, the entire metropolitan area, which includes many cities that did not have zoning reforms in the test area, was included in the study, thus understating the effects of zoning reforms in the cities which did implement such changes.

Finally, as discussed previously, a lack of resources and discontinuation of certain values being collected prevented this study from utilizing favorable statistical values in its analysis. With potential future research done on this subject, access to more specific data that better represents the economic impact of upzoning could provide a fuller understanding of the subject matter.

Endogeneity Concerns

Another limitation of this paper, which should be improved upon in future studies related to this topic, is that the analysis did not address potential endogeneity concerns. The most obvious variable not accounted for is the process of which housing projects get approved.

In American cities, even if a proposed construction project adheres to the locality’s zoning and building codes, depending on the location, it might still have to be approved by the local government and residents. Because different localities have different processes, changes in building permits and the variables affected by increasing density might not change uniformly even if zoning reforms and market demands are similar.


With a recent wave of US cities passing zoning reforms that allow for denser housing construction on previously zoned cities, this paper aims to establish a basis on the effectiveness of these new policies. Using Minneapolis-St. Paul as the analyzed city, because it was one of the first cities to adopt such policies, this paper serves as an early observation of these policies which should be further explored as these zoning reforms become more widespread and become more established. Through the data collected from the study, it can be concluded that upzoning policies have a positive economic impact in its metropolitan areas. However, the data also suggests that for the intended economic benefits to be fully realized, more widespread action must be taken in both existing communities, which have already adopted upzoning policies, and communities which have not adopted. Through a difference in differences analysis between Minneapolis-St. Paul and Kansas City, it was concluded in this study that the zoning reforms adopted by Minneapolis-St.Paul to increase housing density largely worked. Although additional duplex construction lagged behind Kansas City’s, more triplex and 5+ housing units were built. This increase in overall housing density was responsible for a very slight decrease in housing costs and a very slight increase in retail activity. Although the changes appear to be almost negligible, it should be noted that not all of Minneapolis-St. Paul adopted the upzoning policies and that the implemented policies should be given further time to take root in the existing communities which have implemented such policies. There are two aspects which future studies could build off of this one. Firstly, the largest weakness of this study is in regards to the data available for analysis. If city-wide data could be found in sufficient amounts, a comparison between cities which have and have not adopted upzoning policies would be much more accurate than the comparison between metropolitan areas. Additionally, if more specific metrics for retail performance or rent could also be found, it would make for a stronger analysis of the economic changes brought about by upzoning. Secondly, simply letting more time to pass between the implementation of the zoning reforms and additional research would produce more nuanced analysis. For example, increasing the housing density of an area allows for public transportation to become more of a feasible option for governments to build in the area. However, opening up additional transit lines not only needs the density of an area to reach a certain magnitude, which takes time in itself, but the political procedures is another lengthy process which might not even come into fruition. Given enough time, there should be enough cities, however, with denser zoning for an analysis of economic benefits of upzoning with the factor of promoting transportation taken into consideration to be feasible. Additionally, although this paper explores the impact of consumers when given alternative methods of transportation, it doesn’t look into the immediate impacts which alternative methods to automobile transportation promote: vehicular emissions, traffic accidents, commute times, etc. These are all economic factors which have enough merit to be explored in future research of this topic. Fundamentally, two aspects are holding back this study which future studies should remedy. Data availability, the first aspect, could be resolved with a study given to access to greater resources. The second aspect is time; with more time eclipsing between the implementation of the policies and the gathering of data, both the number of cities able to be properly analyzed and the total effects of the changes in policy would increase. Thus, as American cities continue to move in their current trends of embracing upzoning, it would be of a great importance to further analyze the effects of upzoning cities. Finally, this paper’s lack of methods to address endogeneity concerns in the tested metropolitan areas should also be looked at in future studies with more resources.

  1. Arlington, Virginia. County Board Adopts Expanded Housing Options  (2023, March 22). Retrieved from []
  2. Bureau, U.C. BPS e Permits by MSA (n.d.). Retrieved from []
  3. Edward Pinto, Tobias Peter, How Government Policy Made Housing Expensive and Scarce, and How Unleashing Market Forces Can Address It. Cityscape, 25, 123-168 (2023). Retrieved from []
  4. Gilles Duranton, Diego Puga, The Economics of Urban Density. The Journal of Economic Perspectives, 34, 3-26 (Summer 2020). Retrieved from []
  5. Harvey M. Choldin, Urban Density and Pathology. Annual Review of Sociology, 4, 91-113 (1978). Retrieved from []
  6. Jacob Krimmel, Betty Wang, Upzoning With Strings Attached: Evidence From Seattle’s Affordable Housing Mandate. Cityscape, 25, 257-278 (2023). Retrieved from []
  7. J. Cantong, S. Menedian, & S. Gambhir. Zoning Reform Tracker. Othering and Belonging Institute; University of California Berkeley (2023, July 6). Retrieved from []
  8. J. M. Quigley, & L. A. Rosenthal, The Effects of Land Use Regulation on the Price of Housing: What Do We Know? What Can We Learn? Cityscape, 8, 69–137 (2005). Retrieved from []
  9. Joseph Weil Huennekens, Learning From Land Use Reforms: The Case of Ramapo, New York. Cityscape, 25, 225-256 (2023). Retrieved from []
  10. K. McCormick, Rezoning History Influential Minneapolis Policy Shift Links Affordability, Equity. Lincoln Institute of Land Policy (2020, January 16). Retrieved from []
  11. Leo Levy, & Allen N. Herzog, Effects of Population Density and Crowding on Health and Social Adaptation in the Netherlands. Journal of Health and Social Behavior, 15, 228-240 (September 1974). Retrieved from []
  12. M. Lewyn, Sprawl in Canada and the United States. The Urban Lawyer, 44, 85–133 (2012). Retrieved from []
  13. N. Fan, K. Kockelman, P. Caballero, J. Hawkins, & X. Chen. How Does Upzoning Impact Land Use and Transport: a Case Study of Seattle (n.d.). Retrieved from []
  14. Noah M. Kazis, Learning From Land Use Reforms: Housing Outcomes and Regulatory Change. Cityscape, 25, 93-106 (2023). Retrieved from []
  15. R. Tolley, Good for Business The benefits of making streets more walking and cycling friendly (2011, November 22). Retrieved from []
  16. Stuart S. Rosenthal, William C. Strange, How Close is Close? The Spatial Reach of Agglomeration Economies. The Journal of Economic Perspectives, 34, 27-49 (Summer 2020). Retrieved from []
  17. Times Editorial Board. Editorial: Patience, please. Undoing damage from decades of anti-housing policy in California won’t happen overnight. Los Angeles Time (2023, January 23). Retrieved from,lot%2C%20has%20barely%20been%20used. []
  18. Todd Littman,  Economic Value of Walkability (April 15, 2022). Retrieved from []
  19. U.S. Bureau of Labor Statistics. Consumer Price Index for All Urban Consumers: Rent of Primary Residence in Kansas City, MO-KS (CMSA) (DISCONTINUED) (1984, January 1). Retrieved from []
  20. U.S. Bureau of Labor Statistics, & Federal Reserve Bank of St. Louis. All Employees: Retail Trade: General Merchandise Stores in Kansas City, KS (1990, January 1). Retrieved from []
  21. U.S. Bureau of Labor Statistics, & Federal Reserve Bank of St. Louis. All Employees: Retail Trade: General Merchandise Stores in Minneapolis-St. Paul-Bloomington, MN-WI (MSA) (1990, January 1). Retrieved from [] []
  22. U.S. Federal Housing Finance Agency. All-Transactions House Price Index for Kansas City, MO-KS (MSA) (1976, January 1). Retrieved from [] []
  23. U.S. Federal Housing Finance Agency. All-Transactions House Price Index for Minneapolis-St. Paul-Bloomington, MN-WI (MSA) (1976, April 1). Retrieved from [] []
  24. W.A. Fischel, An Economic History of Zoning and a Cure for its Exclusionary Effects. Urban Studies, 41 , 317–340 (2004). Retrieved from [] []
  25. X. Li, (2022). Do new housing units in your backyard raise your rents? Journal of Economic Geography 22 (2022). Retrieved from [] []


Please enter your comment!
Please enter your name here