âčïž Skipped - page is already crawled
| Filter | Status | Condition | Details |
|---|---|---|---|
| HTTP status | PASS | download_http_code = 200 | HTTP 200 |
| Age cutoff | PASS | download_stamp > now() - 6 MONTH | 0.6 months ago |
| History drop | PASS | isNull(history_drop_reason) | No drop reason |
| Spam/ban | PASS | fh_dont_index != 1 AND ml_spam_score = 0 | ml_spam_score=0 |
| Canonical | PASS | meta_canonical IS NULL OR = '' OR = src_unparsed | Not set |
| Property | Value |
|---|---|
| URL | https://link.springer.com/article/10.1007/s00181-026-02894-6 |
| Last Crawled | 2026-03-23 01:33:34 (17 days ago) |
| First Indexed | not set |
| HTTP Status Code | 200 |
| Meta Title | Exploring the quantitative impact of medical marijuana dispensaries on residential sale prices in Oklahoma | Empirical Economics | Springer Nature Link |
| Meta Description | Medical marijuana was legalized in Oklahoma as recently as 2018, and since then Oklahoma has rapidly grown to have the largest number of medical marijuana |
| Meta Canonical | null |
| Boilerpipe Text | 1
Introduction
Oklahoma State Question 788 (Initiative Petition No. 412), or SQ 788, was filed with the Oklahoma Secretary of State on April 11, 2016, and proposed the legalization of the purchase and consumption of medical marijuana for adults with a state issued medical marijuana license. SQ 788 established that a regulatory officeâthe Oklahoma Medical Marijuana Authority (OMMA)âbe opened under the Oklahoma State Department of Health to regulate and receive applications for dispensaries, consumption, growing, and packaging licenses. SQ 788 passed on June 26, 2018, with a 56.86 percent majority vote, and took effect on August 25, 2018, making Oklahoma the 30th state in the USA to legalize medical marijuana. Since August 2018 when SQ 788 went into effect, Oklahoma has seen explosive growth in the number of medical marijuana dispensaries within the state. As of February 27, 2024, the State of Oklahoma has 2227 uniquely licensed medical marijuana dispensaries listed on the OMMA website, making Oklahoma the leading state for the most dispensaries of any kind in the USA. Compared to the next three states in orderâCalifornia, Colorado, and Michigan that have 1244, 1023, and 994 dispensaries, respectively, as of January 1, 2024 (dispenseapp.com)âit is clear how explosive the demand for medical marijuana dispensary licenses has been in the State of Oklahoma since legalization. What has been the impact of this rapid proliferation of medical marijuana dispensaries on the residents within the state? We seek to provide an answer to this question by investigating the local property effects of medical marijuana dispensaries in Oklahoma via the lens of a hedonic pricing model.
Clearly, this rapid proliferation of medical marijuana dispensaries in Oklahoma in the few years since legalization makes Oklahoma an ideal research setting. Beyond just this rapid proliferation, Oklahoma is somewhat of a surprise given that it is generally considered to be a strongly conservative state with much of its population identifying with the Republican Party. Voters in Oklahoma have consistently supported the Republican presidential candidate in every election since 1968, and in the 2020 US presidential election, barely two years after SQ 788 was enacted into state law, each of the 77 counties in Oklahoma voted in majority for the Republican presidential candidate. Since 2004, political party affiliation has been shown to be a statistically significant determinant in measuring support for the legalization of medical marijuana, with those who identify as Republicans expressing significantly less support than those who identify as Democrats or Independents (Denham
2019
).
Footnote
1
We will also show, in our descriptive statistics, that the medical marijuana dispensaries located across Oklahoma are not clustered purely in downtown areas; instead, the dispensaries are broadly located throughout the cities in Oklahoma as well as in both heavily urban and more rural areas. Therefore, the combination of the rapid increase in the quantity of medical marijuana dispensaries in Oklahoma, the strongly Republican electorate, and the broad scattering of dispensaries across the state makes Oklahoma an ideal context in which to explore the property value effects of legalized medical marijuana.
To date, there has been relatively little research done on the effects of (legalized) marijuana dispensaries on residential property prices, and so far, the empirical evidence is mixed. Cheng et al. (
2018
); Burkhardt and Flyr (
2019
) and Conklin et al. (
2020
) all find that legalized marijuana leads to a measurable increase in property values in Colorado, while (Boswell
2021
) finds a residential property value increase in California. Cheng et al. (
2018
) find a 6 percent increase of housing values, and Burkhardt and Flyr (
2019
) find that a marginal dispensary located within a half-mile of a new home increases home prices by about 7.7 percent. Conklin et al. (
2020
) study the conversion of medical marijuana dispensaries to recreational marijuana dispensaries, and find that single-family properties located within 0.1 miles of a converted dispensary realize an 8.4 percent increase in property value. They suggest that lower crime rates associated with dispensaries and higher housing demand associated with dispensary business activity may drive the boost in property values. Tyndall (
2021
), on the other hand, provides evidence to the contrary. Using a sample of repeat home sales from Vancouver, BC, he finds a decrease in property values ranging from 3.7 to 4.9 percent (depending on the regression specification) for properties located within 100Â m of a dispensary.
A related literature focuses on the spatial distribution and economic geography of dispensary locations to understand the extent to which dispensaries are viewed as undesirable. Boggess et al. (
2014
) find that medical marijuana dispensaries in Denver do not correlate with neighborhood racial composition or income and conclude that medical marijuana dispensaries are not locally undesirable.
Footnote
2
They do find that dispensaries are more likely to be located in areas with higher crime rates and greater density of retail employment. Somewhat differently, Morrison et al. (
2014
) find that medical marijuana dispensaries in California are more likely to be located in lower income areas, as well as areas with a greater number of alcohol outlets, and Brinkman and Mok-Lamme (
2019
) show that dispensaries are more likely to be located in Hispanic areas. Most related to our current work is Zimmer et al. (
2022
) and Cohn et al. (
2023
), both of which study dispensary siting in Oklahoma. Zimmer et al. (
2022
) find that medical marijuana dispensaries in Oklahoma are more likely to locate in areas with higher rates of crime, a greater number of uninsured individuals or individuals on disability, and in closer proximity to the Texas border. Cohn et al. (
2023
) find that Oklahoma dispensaries are more likely to be located in areas of lower socioeconomic status (racial minority and low-income) and in areas with more limited access to pharmacies.
Finally, some of the ânot in my backyardâ (or, NIMBY) literature is relevant to our work. Most immediately related is Iannacchione et al. (
2020
) who conducted a survey of dispensary operators (owners or managers) in both Colorado and Washington to assess the extent to which local residents may have been opposed to the opening of the dispensary; that is, the extent of NIMBY effects surrounding legalized medical or recreational marijuana dispensaries. They find that the NIMBY effect of legalized marijuana dispensaries is generally non-existent, and the cases in which dispensary operators faced pushback was short-lived and typically driven by a lack of education about marijuana dispensaries and/or uncertainty (that soon subsided). On the political affiliation side of NIMBY research, Whittemore and BenDor (
2019
) found that although both liberals and conservatives were likely to be concerned with changes in property values, conservatives were significantly more likely to show concern (at least in that context of new residents within the context of high-density residential infill). Extrapolating to our context, we might find evidence of harmful effects of dispensaries on nearby property values given the more conservative population.
Footnote
3
We consider models with data pooled across counties, as well as county-specific regressions; we consider both linear and nonlinear regression specifications. While we find some heterogeneity in the estimated effects of dispensaries on local home prices, a broad trend that emerges is that many residents prefer not to live in immediate proximity to a dispensary but prefer to maintain access to a dispensary within a moderate distance from the home (about 1â2 kilometers away). There are some instances in which the dispensaries do not have a significant effect on property values, and some cases wherein dispensary proximity raises housing values. These different results are discussed and in the context of different counties and relative robustness of the regression results (e.g., sample size).
Our work is important in at least two ways. First, we contribute to the measurement of the property impacts of legalized marijuana activity in Oklahoma. These estimates are important for citizens and policymakers in Oklahoma interested in understanding some of the effects of SQ 788 on local neighborhoods, particularly since we find evidence that many homeowners prefer not to live in close proximity to a dispensary despite the bill passing in Oklahoma with broad public support. Second, our estimates are also relevant for citizens and policymakers in other states that have already legalized marijuana, as well as in states in which legislation to legalize marijuana is (or may soon be) proposed. In either case, understanding (some of) the effects of legalized marijuana is of critical policy-relevance, and as motivated, Oklahoma is uniquely situated to provide insights that are relevant in the broader context. Finally, readers should recognize that our work focuses on localized housing market impacts of legalized medical marijuana and is not an overall welfare measurement.
2
Empirical approach
2.1
Hedonic price model
Our regression model is a standard hedonic pricing model (e.g., Rosen
1974
; Palmquist
1991
; Taylor
2003
; Parmeter and Pope
2013
) defined as
$$\begin{aligned} \ln Price_{it} = \beta _0 + Disp_{i}\beta _1 + X_{i} \delta + v_t + \varepsilon _{it} \hspace{2em} i=1,2,\dots ,n; \hspace{1em} t=1,2,\dots ,T \end{aligned}$$
(1)
where
\(\ln Price_{it}\)
is the natural log of the price of property
i
at the time of sale,
t
,
\(Disp_{i}\)
is a vector of variables measuring the proximity of the property to the dispensaries with corresponding parameter vector
\(\beta _1\)
,
\(X_{i}\)
is a
K
-dimensioned vector of property- and neighborhood-specific control variables with
\(\delta \)
as a corresponding vector of parameters,
\(v_t\)
is a year-specific effect common to all properties, and
\(\varepsilon _{it}\)
is the mean-zero error term.
Footnote
4
Note that we do not deflate property prices and instead prefer to control for differences in price levels across years through
\(v_t\)
. We are primarily interested in
\(\beta _1\)
as this parameter vector measures the impact of the dispensaries on the value of nearby properties.
2.2
Proximity to dispensaries
The proximity vector,
\(Disp_{i}\)
, contains several different measures of property proximity to the dispensaries to fully explore the effect of the dispensaries on property values. Our empirical strategy follows, for instance, Delgado et al. (
2016
) in that we consider both continuous and discrete measures of proximity, both individually and separately.
The first measure of proximity that we use is the continuous distance in meters from each property to the nearest dispensary, defined as
minDist
. Recent research focusing on hedonic regression methodology has proposed data-driven techniques for determining the optimal distance by which to model the effect of an amenity on housing prices. Specifically, Fitzpatrick and Parmeter (
2021
) propose a cross-validation procedure whereby the optimal distance is selected via a numerical optimization procedure; we apply their method in selecting the optimal distance of dispensary effect based on the
minDist
variable, and then use this optimal distance as a way of filtering our sample into the region of effect (within the threshold) and the relevant comparison area (within two times the threshold). We explore regression specifications with
minDist
measured both linearly and via a quadratic specification, with the latter allowing us to detect nonlinearities in the relationship between proximity to the nearest dispensary and the property price. Tyndall (
2021
), for example, finds localized effects of dispensary proximity on property values, and a nonlinear (quadratic) distance specification will help us capture such localized effects more readily than a linear specification. Using a continuous distance measure also allows us to recover the shape of the distance-impact gradient to determine over exactly what range of distance the dispensaries affect property values (if at all), and how rapidly the effect dissipates (if at all).
An alternative way to measure the effect of the dispensaries on property values is to measure the intensity of the dispensary proximity to each property using pre-specified distance bands. We select distance bands of 500Â m, 1000Â m, and 2000Â m and count the number of dispensaries that are within each distance band from each property as our measures of dispensary intensity. We denote these variables as
Count
500,
Count
1000, and
Count
2000 and consider regression specifications with these variables included together as well as separately.
Footnote
5
It is important to understand the interpretation of the regression coefficients when these variables are included individually or jointly: when these coefficients are included individually in the regression, the coefficient is simply the effect on
\(\ln Price\)
of an additional dispensary being added within the distance band. For instance, if we include
Count
1000 in the regression, then the coefficient reflects the effect of adding another dispensary within 1000
m
(or 1
km
) of the property on the
\(\ln Price\)
of the property (control variables held constant). On the other hand, if we include all three count variables in the model jointly, then the coefficient on
Count
1000 reflects the effect of adding another dispensary within 1000
m
of the property on the
\(\ln Price\)
of the property
holding constant
the number of dispensaries within 500
m
and 2000
m
of the property (all other control variables being held constant as well). In other words, the coefficient on
Count
1000 now captures the effect on
\(\ln Price\)
of adding another dispensary in the
\(>500m\)
but
\(\le 1000m\)
distance range.
Finally, it is worth clarifying that a regression specification including both the continuous and discrete proximity measures together via
\(Disp=(minDist,Count500, Count1000,Count2000)\)
bears the following interpretation. The coefficient on
minDist
is the effect on
\(\ln Price\)
of moving the dispensary one meter away from the property holding constant the number of dispensaries located within 500
m
, 1000
m
, and 2000
m
from the property (in addition to all other variables in the model being held constant). That is, we interpret the distance effect with intensity held constant. We interpret the coefficients on the count variables, say
Count
1000, as the effect of adding one more dispensary at a distance
\(>500m\)
but
\(\le 1000m\)
while holding the distance to the nearest dispensary constant. That is, an increase in intensity within the specified distance band while holding the distance to the nearest dispensary constant.
By considering the continuous
minDist
measure of proximity and the three count measures of intensity both individually and jointly, we are able to fully explore the nature of dispensary proximity on nearby property values. It is worth noting that further exploring these regressions separately by county allows us to explore the extent to which these relationships are heterogeneous in different parts of Oklahoma.
2.3
Control variables and fixed effects
Of course, the extent to which our estimates of
\(\beta _1\)
are reliable depends on our ability to adequately control for other factors that affect
\(\ln Price\)
and also correlate with the effect of the dispensaries on property values. In our main regression specifications, we deploy control variables and fixed effects to account for these other factors. Being a hedonic price model, we naturally control first for housing characteristics. In our primary dataset, we control for the number of bedrooms, the number of bathrooms, the age of the property, and the square footage of the property. With regard to the age of the property and square footage, we find the best fit via a quadratic specification. It is important to note that these particular variables are common across the four counties we consider jointly in our primary specifications; we explore additional covariate controls as the data are available in auxiliary regressions that are county-specific.
Footnote
6
Beyond these house-specific characteristics, we include the distance to the central business district in kilometers, as well as the average sales price of other properties within a 500Â m radius of each property sold
prior
to SQ 788. This variable is particularly important for identification, as it allows us to control for neighborhood-level differences that might otherwise be correlated with the location of dispensaries. We noted earlier that Zimmer et al. (
2022
) find that the location of medical marijuana dispensaries in Oklahoma is correlated with distance to Texas (where both medical and recreational marijuana is illegal), and rates of crime, the population that is uninsured, or has a disability, and Cohn et al. (
2023
) find that Oklahoma dispensary locations correlate with fewer hospitals, being uninsured, and having a lower socioeconomic status (defined by either race or income). In Denver, medical marijuana dispensary locations correlate with higher crime rates and having a higher rate of retail sale establishments (Boggess et al.
2014
). Regardless of the extent to which any of these particular factors causally affect dispensary location, it is clear that medical marijuana dispensariesâin Oklahoma and elsewhereâcorrelate with neighborhood factors. These factors are surely reflected by differences in neighborhood real estate prices, and so by controlling for average neighborhood property prices prior to SQ 788, we are able to adjust for differences in neighborhood effects that might otherwise correlate with dispensary locations.
In terms of the fixed effects, we control for the year the property was sold to account for macro-effects that are common to all properties sold within each year, as well as county-specific dummies to capture differences across counties.
2.4
Auxiliary regression specifications
We consider auxiliary regression specifications that serve as robustness checks. Specifically, we separate the data by county and consider county-specific regressions. This robustness check does two things. First, this approach allows us to estimate county-specific parameters for all of the coefficients in the model, to determine the extent to which the estimated relationships vary by county. Second, the data we obtained from the county assessor offices contain different sets of control variables, and by considering each county individually, we are able to add additional control variables that are available for each particular county. While the county-specific model specifications are not dramatically different, it is worth exploring the importance of these different control variables where possible. More specific details will be provided in subsequent sections where we explore these data and regressions.
Footnote
7
3
Data
3.1
Housing data
Residential property sales data were obtained from six county assessor offices in the State of Oklahomaâas noted in the introduction, Comanche, Oklahoma, Payne, Tulsa, Wagoner and Washington counties. The data include all single-family property sales from the beginning of 2016 to the end of 2023, providing information on the sales price, geographic location, and property attributes. In terms of the property attributes, the data includes typical housing attribute variables that are commonly found in the hedonic housing literature, including the age of the property measured in years, living area in square feet, the number of bedrooms, and the number of bathrooms. There is slight variation across county-specific datasets in terms of additional housing-level variables. Specifically, the data from Payne and Tulsa counties lack key variables, and so we consider the property sales from these counties in auxiliary regressions but not as part of our main benchmark models. Specifically, our data for Payne County lack data on the age of the property and property square footage, and our data for Tulsa County lacks data on the number of bedrooms of each property.
3.2
Dispensary data
Medical marijuana dispensary data are made publicly available to view and download online by the Oklahoma Medical Marijuana Authority (OMMA). Publicly available data as of 2024 include the name of the licensed dispensary, the license identification number, as well as the county and city the dispensary operates in within the state. These data are used by Zimmer et al. (
2022
) and Cohn et al. (
2023
), with Cohn et al. (
2023
) noting that they conducted online searches using the dispensary names provided by the OMMA to further identify exact geographical locations and current operational status. In our case, we use publicly available data provided by the Oklahoma Watch organization website (Brown and Howe
2019
) that is maintained by Trevor Brown and Jesse Howe; their data were obtained through the OMMA and mapped using Googleâs My Maps service.
Footnote
8
The data include 1,463 unique medical marijuana dispensary locations from across the entire state.
It is important to note that with these data, we do not know the date at which each dispensary opened. To link the dispensaries to the property sales, in both the benchmark and auxiliary datasets we include only property sales after July 2020 because the dispensary data we use comes from a cross-section of dispensaries that were in operation as of July 2020. By focusing on these particular sales transactions, we are able to ensure that our linkage of sales transactions to dispensary locations is concurrent in time.
3.3
Geolocation and mapping
For each county, each property comes with a geolocating identifier. Counties that provided a street address for each home were geocoded into latitudinal and longitudinal coordinates using Esri ArcGIS Pro 3.1. We use the geographical location of the property to compute the distance to the nearest central business district (CBD): we use the âgeosphereâ package in
R 4.2.2
to compute in kilometers the distance between the property and the closest CBD, which, in our data, is either downtown Oklahoma City or downtown Tulsa. We also use these locators to compute the distance in meters to the closest dispensary, as well as the number of dispensaries in binned radii around the property to provide a count of how many exist within 500, 1000, and 2000Â m. We also use these geographical locators to construct the historical neighborhood value variable: We construct a 500Â m radius around each property and then compute the average sale price of all properties sold within the radius between January 2016 and July 2018. These years correspond to our sales data prior to the opening of any dispensaries, thereby providing a market value-based measure of historical neighborhood quality. We also report the average number of properties within this 500Â m radius (the âdensityâ variable) for readers to understand the rough size of these local neighborhoods.
3.4
Descriptive statistics
Table
1
provides a descriptive summary of the 4-county benchmark data (the pooled sample) as well as the data broken down by county. More detailed descriptive statistics are in the appendix.
Table 1 Descriptive statistics for the pooled sample of four counties and separately by county
Full size table
Across the pooled sample, the average sale price is around $253,000, with average prices being slightly higher in Oklahoma and Tulsa counties, and lower in Comanche, Wagoner and Washington counties. These properties are, on average, three bedroom and two bathroom homes, that are around 40 years old and have just under 2000 square feet of living space. The average age of the home, in particular, varies quite a bit across counties. From the counties with available data, these are typically one story homes. We measure the central business district (CBD) as either Oklahoma City or Tulsa, and from Table
1
we see that the average distance to CBD is just under 40 kilometers.
The average historical neighborhood value is around $203,000, which is roughly $50,000 lower than the average sale price, and this computation comes from a local neighborhood density of, on average, 24.8 home sales. These numbers do not account for increases in property values through the intervening time period; this is an unconditional comparison of average values. To dig deeper into the relationship between historical neighborhood values and current property valuesâand thereby to provide some empirical justification for neighborhood to serve as an important control variable in our analysis and thereby aid in our identification strategyâwe turn to Fig.Â
1
and Table
2
. In Fig.Â
1
, we show a scatterplot correlation between historical neighborhood property values and current property prices, and Table
2
reports univariate regression coefficients (i.e., correlations) between these variables. We can see a strong, positive correlation between neighborhood prices pre-dispensaries and property values post-dispensaries; from the table, we can see that this correlation is statistically significant with a coefficient of about 0.7.
Fig. 1
Scatterplot of historical neighborhood values and current property prices
Full size image
Table 2 Univariate correlation between neighborhood values and
\(\ln (Price/100)\)
Full size table
In the spirit of Boggess et al. (
2014
); Zimmer et al. (
2022
) and Cohn et al. (
2023
), we consider the correlation between
Count
500 and historical neighborhood value. FigureÂ
2
provides a scatterplot and estimated trend line, and Table
3
provides the univariate regression estimate. There is a clear downward trend, whereby as the number of dispensaries within 500Â m increases, historical neighborhood value decreases; the regression table indicates that this correlation is statistically significant. Further, our estimated correlation is linear, though the figure clearly suggests a nonlinear pattern whereby the largest downward effect of dispensary proximity occurs with the siting of the first one or two dispensaries within 500Â m, with a dramatically reduced effect with marginal increases in dispensary intensity. Of course, this is merely a correlationâthus, a descriptive statistic that motivates our use of neighborhood as a control variableâand should not be taken as evidence that dispensaries are located in areas characterized in any particular way (e.g., by race, income, or other status).
Turning to the dispensary-related variables, we see that, on average, the minimum distance to the nearest dispensary is about 1.4 kilometers (Table
1
). These properties have, on average, 0.34 dispensaries within a 500Â m radius of the property; 1.5 dispensaries within 1Â km of the property; and 5.7 dispensaries within 2 kilometers of the property. It is worth noting that Washington, Oklahoma and Tulsa counties have properties that are in closest proximity, on average, to dispensaries and with the highest average density of proximate dispensaries.
Fig. 2
Scatterplot between
Count
500 and neighborhood value before legalization
Full size image
Table 3 Univariate correlation between
Count
500 and neighborhood
Full size table
3.5
Mapping the data
Fig. 3
Property sales and dispensary locations across Oklahoma. Red pins indicate dispensary locations, green pins indicate property locations for our four-county benchmark sample, and purple pins denote property locations from our auxiliary samples (Tulsa and Payne counties)
Full size image
We have described the rapid extent to which medical marijuana dispensaries have opened in Oklahoma since 2018, and the following figures help to illustrate this fact. In Fig.Â
3
we map all the properties in our dataset against the locations of all the dispensaries across the State of Oklahoma. Specifically, green circles represent the location of an individual residential sale from the benchmark dataset, and the red pins represent the locations of the dispensaries that were operating as of July 2020. It is clear from the map that the dispensaries are located all across Oklahoma, covering both urban and rural areas; given the short time span since the passage of the law, the number of operating dispensaries has quickly risen.
To gain deeper insight into the locations of the property sales and dispensaries, we plot the Oklahoma County data in Fig.Â
4
. We can clearly see the dispensaries located broadly across Oklahoma County (primarily this is Oklahoma City), located not just in the city center but in suburban areas as well.
Fig. 4
Property sales and dispensary locations in Oklahoma County. Red pins indicate dispensary locations, and green pins indicate property locations
Full size image
4
Results
4.1
Benchmark regression analysis
Table
4
displays the results from our benchmark regression model that includes both
minDist
and all three
Count
variables (
Count
500,
Count
1000,
Count
2000); a model that includes
minDist
but excludes the three
Count
variables; a model that is quadratic in
minDist
and excludes the
Count
variables; and a model that is quadratic in
minDist
, interacts
minDist
with historical neighborhood values, and excludes the three
Count
variables. The properties included in these regressions are all within two times the optimal cross-validated distance threshold, which for the pooled sample is 3110Â m. These models allow us to first set a benchmark for the housing market impacts of dispensary activities (Column 1) and then gain a deeper understanding into the relationship between
minDist
and property values (Columns 2â4). The data used in these regressions are the pooled sample of four counties for which we have complete and overlapping control variables (i.e., Comanche, Wagoner, Washington and Oklahoma counties). In all four models, we include housing characteristic variables that includes quadratic specifications for both age and square footage, as well as year of sale and county indicators.
Table 4 Benchmark model and sub-models to further assess
minDist
Full size table
Looking first at the benchmark model shown in Column 1 in Table
4
, we see that
minDist
,
Count
1000 and
Count
2000 are statistically significant.
Count
500 is not significant. Further,
minDist
and
Count
2000 have a positive coefficient, whereas
Count
1000 has a negative coefficient. Interpreting first the
minDist
coefficient: the model is a semi-log model, which means that a one-meter increase in the distance from the nearest dispensary leads to a
\(\hat{\beta }\times 100\)
percent increase in the sale price of the property (in the regression, prices are in hundreds of dollars). Our estimate thus implies a 0.002197 percentage change in price in hundreds with every one-meter increase in distance. Recalling from Table
1
that the average property price is 253,060.30, this estimated marginal effect implies that a one-meter increase in distance to nearest dispensary leads to a 5.56 dollar increase in sale price after converting into dollars. Again, bear in mind that this is controlling for the
Count
variables that capture dispensary intensity as well as the other control variables. If we extrapolate to a 100-m increase in distance to nearest dispensary, this is a property price increase of about 555.97 dollars. These estimates indicate that dispensaries are on net undesirable, though this is not a substantial price increase.
We turn next to the
Count
variables and their estimated impacts on property prices. Recall that since all three variables are included in the regression, the interpretation of each
Count
variable coefficient is incremental: adding one more dispensary within a 500Â m radius of a property, holding constant the number of dispensaries within the 500â2000Â m range, leads to a
\(\hat{\beta }_{Count500}\)
change in a propertyâs price in hundreds of dollars or a
\(\hat{\beta }_{Count500} \times 100\)
percentage change. Table
4
shows that
Count
500 is insignificant, while
Count
1000 is significantly negative and
Count
2000 is significantly positiveâholding
minDist
and other controls constant. Specifically, our estimates indicate that, all else constant, adding an additional dispensary within the 500â1000Â m distance from a property leads to a decrease in the average property price by about 0.429 percent; from the average home price, this is a decrease of about 1085.63 dollars. At the same time, the coefficient on
Count
2000 indicates that adding one more dispensary in the 1000â2000Â m range, holding
minDist
, the other
Count
variables, and controls constant, results in a 0.427 percent increase in property values (or about 1080.57 dollars at the average). Taken at face value, these estimates indicate that homeowners prefer to have a marginal dispensary located farther away from the home.
It is worth discussing the effects and significance of the control variables on property prices before digging deeper into the dispensary exposure variables and their relationship to property prices. Still looking at the benchmark model in Column 1 of Table
4
, we see statistically significant quadratic relationships between the age and square footage of the property and property values. In particular, property age has an inverse-U-shaped relationship with property prices, with a minimum property value occurring at a property age of 70 years. The average age of the home in the pooled sample is 40 years (Table
1
). Square footage has a U-shaped relationship with property values, with the maximum occurring at 15,046 square feet. From Table
1
, the average square footage in the pooled benchmark sample is 1904.99, and the detailed descriptive statistics tables in the appendix indicate that this turning point is greater than the maximum square footage in our data; in effect, the square footage effect on property prices is monotonically increasing but at a decreasing rate. We see that the number of bedrooms is significantly negative; given that we control for square footage, the negative coefficient on bedrooms means that home buyers prefer fewer but larger rooms for a home of a given size. The number of bathrooms and neighborhood home density are significantly positive, while distance to the nearest CBD is significantly negative. The historical neighborhood value is significantly positive, which is in line with the correlations shown in Fig.Â
1
and Table
2
, and indicates that historical average neighborhood values are positively and significantly correlated with current sales prices. This estimate, therefore, reinforces our intuition to include this variable as an important control for neighborhood quality that in part explains current property prices but might also correlate with dispensary activity (distance and exposure). Finally, we see some significance of the year of sale indicators relative to the baseline year of 2023, and significance of the county indicators relative to the Comanche County baseline. We note that the
\(R^2\)
is 0.640, reflecting a reasonable fit of the regression.
4.2
Distance to the nearest dispensary
Moving beyond the benchmark regression, in Columns 2â4 of Table
4
we explore more deeply the relationship between
minDist
and property values. In Column 2, we remove the
Count
variables and explore the linear effect of
minDist
on the natural log of property values: We find the results remain qualitatively unchanged from the benchmark regression (Column 1), with a significantly positive effect of
minDist
on property values. This estimate is fairly smaller in magnitude compared to the estimate in Column 1, this time implying an effect of 3.71 dollars per meter, or about 371 dollars for a 100Â m distance.
In Column 3, we consider a quadratic specification and find that only the quadratic term is statistically significant, and in Column 4 we consider both the quadratic
minDist
specification including an interaction with the historical neighborhood index. In this latter model, we find that all coefficients are significant. Exploring these estimates further, using both the average property value and average neighborhood value index (203,309.40 from Table
1
), we find that on average the
minDist
marginal effect on property prices is about 224 dollars per 100Â m. In looking across the models in Table
4
, we see that sharpening our estimates (i.e., Column 4) results in a smaller estimated average effect of
minDist
on property prices. Yet, from the perspective of these models, dispensaries are viewed by homeowners as a negative amenity.
4.3
Intensity of dispensary exposure
Table 5 Models to further assess significance of count variables
Full size table
Next, we look more deeply into the relationship between the
Count
variables and property values; regression results are found in Table
5
. The first model is a version of the benchmark regression from Table
4
except that we omit
minDist
. Models 2â4 consider the three
Count
variables separately, one at a time. The dependent variable remains
\(\ln (Price/100)\)
.
The results in the first column are very similar to the results shown in the benchmark regression (Column 1 of Table
4
). That is, removing
minDist
does not change the results from the benchmark model. This result mirrors Column 2 of Table
4
in that removing the
Count
variables did not change the effect of
minDist
on property price. That is, we continue to find an insignificant coefficient on
Count
500, a significantly negative coefficient on
Count
1000, and a significantly positive coefficient on
Count
2000 all with magnitudes very similar to those discussed in the previous subsection.
It is interesting that when considering the
Count
variables separately, as shown in Models 2â4, only
Count
2000 remains significant (and still has a positive coefficient). It is important to remember that in these three regressions, the interpretations of the
Count
variables coefficients are different from the joint model. Since the other bands are not held constant, the coefficient on
Count
1000, for instance, indicates the effect on log price of an additional dispensary located anywhere within 1000Â m of the property, whereas previously it pertained only to an additional dispensary being located within the 500â1000Â m distance band from the property. When taken separately, these intensity variables indicate that a change in intensity of dispensary exposure does not significantly affect property values within 500Â m (see Column 1) or up to 1000Â m (see Columns 2â3) of the property. A marginal increase in intensity up to 2000Â m from the property results in an increase in the property value of about 0.19 percent, or about 480.81 dollars based on the average property sale price. Broadly speaking, these results also indicate that residents prefer a marginal dispensary to be located farther from the property, though they may be indifferent to a marginal dispensary being located nearer to the property.
4.4
County-specific regressions
Table 6 County-specific benchmark regressions
Full size table
Thus far we have focused exclusively on both distance and intensity measures of dispensary exposure for our four-county benchmark sample. To ascertain the extent to which there may be heterogeneity in the dispensary effects across counties, and to exploit data on two other counties, we now divide the sample and turn to county-specific regressions. Note that from the four benchmark counties, the majority of observations come from Oklahoma County, with a large group also coming from Comanche County. Relatively, few observations come from Wagoner and Washington counties. Thus, it is interesting to assess the extent to which the pooled sample results merely reflect Oklahoma and perhaps Comanche counties, or are representative of all counties more generally. For these county-specific models, we are no longer bound by a common set of control variables across counties; for each county-specific regression, we include as many control variables as are available for that county. Also, for each county we use the Fitzpatrick and Parmeter (
2021
) cross-validation method to select the optimal
minDist
threshold, and then use all property sales within 2 times that threshold for our county-optimal sample. We report the county-specific regression results in Table
6
. To save space, we withhold estimates of all control variables, and provide the estimates of the dispensary effects.
Footnote
9
Looking first the the coefficients on
minDist
across Table
6
, we see that
minDist
is significantly negative in Comanche and Payne counties, insignificant in Wagoner and Washington counties, and significantly positive in Oklahoma and Tulsa counties.
Count
500 is only significant in Tulsa County, wherein we find the estimate is positive.
Count
1000 is significantly negative in Oklahoma County and significantly positive in Tulsa County, while
Count
2000 is significantly positive in Oklahoma County and significantly negative in Payne County. Taking stock of the first four columns, relative to the benchmark pooled sample regressions, it seems apparent that the results found with the four pooled counties are largely driven by the Oklahoma County effects. Table
6
shows that residents in Comanche County experience a negative
minDist
effect, indicating that property prices fall when dispensaries are located farther away from the home, though the intensity of dispensary activity does not affect property values. Looking toward Payne and Tulsa counties, we see that Payne County bears similar effects to Comanche County, while Tulsa County accords with Oklahoma County. It is worth pointing out the following. First, there is quite a bit of heterogeneity across counties in terms of both the optimal
minDist
threshold and the
\(R^2\)
measures for each regression, and readers should bear in mind that both Payne and Tulsa counties are missing several key variables (which is why they were left out of the baseline four-county model). In terms of the
\(R^2\)
, Oklahoma County seems to have the best fit, relative to the regressions for the other counties.
5
Discussion and conclusion
We study the effects of medical marijuana dispensaries on residential property prices in Oklahoma as a means of measuring the extent to which dispensaries are seen as desirable or undesirable neighborhood amenities and thereby affect property prices. We have motivated the uniqueness of Oklahoma for our study setting: not only has Oklahoma very recently legalized medical marijuana (effective August 2018) but Oklahoma has seen a rapid rise in the proliferation of medical marijuana dispensaries and has become a leader in the nation in terms of the number of operating dispensaries. Moreover, Oklahoma is an interesting case because of the strongly conservative electorate, despite marijuana legalization being typically seen as a politically liberal interest. Thus, we believe that Oklahoma is an ideal setting for studying the effects of medical marijuana legalization on nearby property values with lessons for citizens of Oklahoma as well as the public or policymakers in other states. Using data from six counties in Oklahoma, we conduct a state-wide hedonic pricing analysis to investigate the net effects, as capitalized in property values, of local dispensary activities on residential property prices. We measure dispensary activity with both distance to nearest dispensary as well as intensity of dispensary activity in the proximate area (within 500Â m, 1000Â m, and 2000Â m from the property). We consider pooled sample regressions as well as county-specific regressions, linear and nonlinear specifications. In looking across these six counties, we find some heterogeneity in our results. However, our results point to a few broad trends across most of Oklahoma. First, a general pattern is that residents view local proximity to a medical marijuana dispensary as a net negative, with property prices being significantly lower when either the minimum distance to the nearest dispensary is closer to the home or when a marginal dispensary is opened within 0.5 kilometers from the home. We find the effects to be quite small, with even the largest effects being less than one percent of the propertyâs value. At the same time, residents seem to prefer relatively easy access to medical marijuana dispensaries, with property prices increasing for marginal dispensaries being located at a medium distance (i.e., between 1 and 2 kilometers) from the home. Second, it is worth noting that Washington and Wagoner counties generally have insignificant effects of dispensary activities on home prices, though the sample size in those counties is substantially smaller than for the other counties.
Our results are important for residents and policymakers alike. First, despite the majority voter approval for SQ 788, it appears that residents in Oklahoma have mixed views on dispensary activity. Many residents seem to not want to live in immediate proximity, whereas many want access. It is worth noting that thus far the literature has been mixed in terms of dispensary effects on local residents; our work adds the lens of the experience in Oklahoma, and find that most residents seem to view dispensaries within immediate proximity negatively. These differences in impacts are important for policymakers to recognize, and understand when making adjustments or revisions to Oklahoma laws. Finally, readers should keep in mind that our estimates are both local and net measurements, and do not specifically measure the impact of local dispensary activities on any singular measure of well-being (such as crime rates, related alcohol use, or child abuse) as considered by others in this literature.
References
Adda J, McConnell B, Rasul I (2014) Crime and the depenalization of cannabis possession: evidence from a policing experiment. J Polit Econ 122(5):1130â1202
Article
Â
Google Scholar
Â
Anderson DM, Hansen B, Rees DI (2013) Medical marijuana laws, traffic fatalities, and alcohol consumption. J Law Econ 56(2):333â369
Article
Â
Google Scholar
Â
Anderson DM, Hansen B, Rees DI (2015) Medical marijuana laws and teen marijuana use. Am Law Econ Rev 17(2):495â528
Article
Â
Google Scholar
Â
Boggess LN, PĂ©rez DM, Cope K, Root C, Stretesky PB (2014) Do medical marijuana centers behave like locally undesirable land uses? Implications for the geography of health and environmental justice. Urban Geogr 35(3):315â336
Article
Â
Google Scholar
Â
Boswell K (2021) Essays on Cannacis legalization in California. PhD Thesis, University of Miami
Brinkman J, Mok-Lamme D (2019) Not in my backyard? Not so fast. The effect of marijuana legalization on neighborhood crime. Reg Sci Urban Econ 78:103460
Article
Â
Google Scholar
Â
Brooks TJ, Humphreys BR, Nowak A (2018) Strip clubs, âsecondary effects" and residential property prices. Real Estate Econ 48(3):850â885
Article
Â
Google Scholar
Â
Brown T, Howe J (2019) Number of medical marijuana outlets soars. How many are in your area?
Burkhardt J, Flyr M (2019) The effect of marijuana dispensary openings on housing prices. Contemp Econ Policy 37(3):462â475
Article
Â
Google Scholar
Â
Chang TY, Jacobson M (2017) Going to pot? The impact of dispensary closures on crime. J Urban Econ 100:120â136
Article
Â
Google Scholar
Â
Cheng C, Mayer WJ, Mayer Y (2018) The effect of legalizing retail marijuana on housing values: evidence from Colorado. Econ Inq 56(3):1585â1601
Article
Â
Google Scholar
Â
Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113
Article
Â
Google Scholar
Â
Conklin J, Drop M, Li H (2020) Contact high: the external effects of retail marijuana establishments on house prices. Real Estate Econ 48(1):135â173
Article
Â
Google Scholar
Â
Delgado MS, Guilfoos T, Boslett A (2016) The cost of unconventional gas extraction: a hedonic analysis. Resour Energy Econ 46:1â22
Article
Â
Google Scholar
Â
Denham BE (2019) Attitudes toward legalization of marijuana in the United States, 1986â2016: changes in determinants of public opinion. Int J Drug Policy 71:78â90
Article
Â
Google Scholar
Â
Dong X, Tyndall J (2024) The impact of recreational marijuana dispensaries on crime: evidence from a lottery experiment. Ann Reg Sci 72:1383â1414
Article
Â
Google Scholar
Â
Fitzpatrick LG, Parmeter CF (2021) Data-driven estimation of treatment buffers in hedonic analysis: an examination of surface coal mines. Land Econ 97:528â547
Article
Â
Google Scholar
Â
Freisthler B, Gruenewald PJ, Wolf JP (2015) Examining the relationship between marijuana use, medical marijuana dispensaries, and abusive and neglectful parenting. Child Abuse & Neglect 48:170â178
Article
Â
Google Scholar
Â
Huber A III, Newman R, LaFave D (2016) Cannabis control and crime: medicinal use, depenalization and the war on drugs. BE J Econ Anal Policy 16(4):20150167
Article
Â
Google Scholar
Â
Iannacchione B, Ward KC, Evans MK (2020) Perceptions of NIMBY syndrome among Colorado and Washington dispensary owners and managers. Justice Policy J 17(2)
Kepple NJ, Freisthler B (2012) Exploring the ecological association between crime and medical marijuana dispensaries. J Stud Alcohol Drugs 73(4):523â530
Article
Â
Google Scholar
Â
Lu R, Willits D, Stohr MK, Makin D, Snyder J, Lovrich N, Meize M, Stanton D, Wu G, Hemmens C (2021) The cannabis effect on crime: time-series analysis of crime in Colorado and Washington State. Justice Q 38(4):565â595
Article
Â
Google Scholar
Â
Mair C, Freisthler B, Ponicki WR, Gaidus A (2015) The impacts of marijuana dispensary density and neighborhood ecology on marijuana abuse and dependence. Drug Alcohol Depend 154:111â116
Article
Â
Google Scholar
Â
Morris RG, TenEyck M, Barnes JC, Kovandzic TV (2014) The effect of medical marijuana laws on crime: evidence from state panel data, 1990â2006. PLoS ONE 9(3):e92816
Article
Â
Google Scholar
Â
Morrison C, Gruenewald PJ, Freisthler B, Ponicki WR, Remer LG (2014) The economic geography of medical cannabis dispensaries in California. Int J Drug Policy 25(3):508â515
Article
Â
Google Scholar
Â
Palmquist R D (1991) Hedonic methods. Measuring the demand for environmental quality
Parmeter C F, Pope J C (2013) Quasi-experiments and hedonic property value methods. In: Handbook on Experimental Economics and the Environment, pp 3â66. Edward Elgar Publishing
Rosen S (1974) Hedonic prices and implicit markets: product differentiation in pure competition. J Polit Econ 82(1):34â55
Article
Â
Google Scholar
Â
Shi Y (2016) The availability of medical marijuana dispensary and adolescent marijuana use. Prev Med 91:1â7
Article
Â
Google Scholar
Â
Taylor LO (2003) The hedonic method. A primer on nonmarket valuation. Springer, Cham, pp 331â393
Chapter
Â
Google Scholar
Â
Thomas D, Tian L (2021) Hits from the bong: the impact of recreational marijuana dispensaries on property values. Reg Sci Urban Econ 87:103655
Article
Â
Google Scholar
Â
Tyndall J (2021) Getting high and low prices: marijuana dispensaries and home values. Real Estate Econ 49(4):1093â1119
Article
Â
Google Scholar
Â
Whittemore AH, BenDor TK (2019) Reassessing NIMBY: the demographics, politics, and geography of opposition to high-density residential infill. J Urban Aff 41(4):423â442
Article
Â
Google Scholar
Â
Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University
Download references |
| Markdown | [Skip to main content](https://link.springer.com/article/10.1007/s00181-026-02894-6#main)
[](https://link.springer.com/)
[Account](https://link.springer.com/article/10.1007/s00181-026-02894-6)
[Menu](https://link.springer.com/article/10.1007/s00181-026-02894-6#eds-c-header-nav)
[Find a journal](https://link.springer.com/journals/) [Publish with us](https://www.springernature.com/gp/authors) [Track your research](https://link.springernature.com/home/)
[Search](https://link.springer.com/article/10.1007/s00181-026-02894-6#eds-c-header-popup-search)
[Saved research](https://link.springer.com/saved-research)
[Cart](https://order.springer.com/public/cart)
1. [Home](https://link.springer.com/)
2. [Empirical Economics](https://link.springer.com/journal/181)
3. Article
# Exploring the quantitative impact of medical marijuana dispensaries on residential sale prices in Oklahoma
- [Open access](https://www.springernature.com/gp/open-science/about/the-fundamentals-of-open-access-and-open-research)
- Published:
21 March 2026
- Volume 70, article number 61, (2026)
- [Cite this article](https://link.springer.com/article/10.1007/s00181-026-02894-6#citeas)
You have full access to this [open access](https://www.springernature.com/gp/open-science/about/the-fundamentals-of-open-access-and-open-research) article
[Download PDF](https://link.springer.com/content/pdf/10.1007/s00181-026-02894-6.pdf)
[Save article](https://link.springer.com/article/10.1007/s00181-026-02894-6/save-research?_csrf=gEUufnNoYlj84f3kzQxSi-fpj5a3JZzI)
[View saved research](https://link.springer.com/saved-research)
[ Empirical Economics](https://link.springer.com/journal/181)
[Aims and scope](https://link.springer.com/journal/181/aims-and-scope)
[Submit manuscript](https://www.editorialmanager.com/emec)
Exploring the quantitative impact of medical marijuana dispensaries on residential sale prices in Oklahoma
[Download PDF](https://link.springer.com/content/pdf/10.1007/s00181-026-02894-6.pdf)
- [Joshua Clark](https://link.springer.com/article/10.1007/s00181-026-02894-6#auth-Joshua-Clark-Aff1)[1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Aff1) &
- [Michael S. Delgado](https://link.springer.com/article/10.1007/s00181-026-02894-6#auth-Michael_S_-Delgado-Aff1)
[1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Aff1)
## Abstract
Medical marijuana was legalized in Oklahoma as recently as 2018, and since then Oklahoma has rapidly grown to have the largest number of medical marijuana dispensaries of any states in the USA. What have been the impacts of this rapid proliferation of legalized medical marijuana? We use a hedonic pricing model to assess how legalized marijuana dispensary activity (both distance to the nearest dispensary and the number of nearby dispensaries) has impacted residential property values. The dataset spans six different counties from across the State of Oklahoma, and we explore a variety of different regression specifications to rigorously explore the effects of dispensary activity on local residents. While we find evidence of some heterogeneity in the dispensary effects, one broad finding is that residents prefer not to live in immediate proximity to a dispensary but prefer access to dispensaries moderately distanced from their home. Understanding these broad trends and the localized heterogeneity in effects are important for residents and policymakers alike in Oklahoma and in other states that may be considering similar legislation.
### Explore related subjects
Discover the latest articles, books and news in related subjects, suggested using machine learning.
- [Drugs](https://link.springer.com/subjects/drugs)
- [Law and Economics](https://link.springer.com/subjects/law-and-economics)
- [Legal Geography](https://link.springer.com/subjects/legal-geography)
- [Natural Resource and Energy Economics](https://link.springer.com/subjects/natural-resource-and-energy-economics)
- [Real Estate Economics](https://link.springer.com/subjects/real-estate-economics)
- [Spatial Economics](https://link.springer.com/subjects/spatial-economics)
## 1 Introduction
Oklahoma State Question 788 (Initiative Petition No. 412), or SQ 788, was filed with the Oklahoma Secretary of State on April 11, 2016, and proposed the legalization of the purchase and consumption of medical marijuana for adults with a state issued medical marijuana license. SQ 788 established that a regulatory officeâthe Oklahoma Medical Marijuana Authority (OMMA)âbe opened under the Oklahoma State Department of Health to regulate and receive applications for dispensaries, consumption, growing, and packaging licenses. SQ 788 passed on June 26, 2018, with a 56.86 percent majority vote, and took effect on August 25, 2018, making Oklahoma the 30th state in the USA to legalize medical marijuana. Since August 2018 when SQ 788 went into effect, Oklahoma has seen explosive growth in the number of medical marijuana dispensaries within the state. As of February 27, 2024, the State of Oklahoma has 2227 uniquely licensed medical marijuana dispensaries listed on the OMMA website, making Oklahoma the leading state for the most dispensaries of any kind in the USA. Compared to the next three states in orderâCalifornia, Colorado, and Michigan that have 1244, 1023, and 994 dispensaries, respectively, as of January 1, 2024 (dispenseapp.com)âit is clear how explosive the demand for medical marijuana dispensary licenses has been in the State of Oklahoma since legalization. What has been the impact of this rapid proliferation of medical marijuana dispensaries on the residents within the state? We seek to provide an answer to this question by investigating the local property effects of medical marijuana dispensaries in Oklahoma via the lens of a hedonic pricing model.
Clearly, this rapid proliferation of medical marijuana dispensaries in Oklahoma in the few years since legalization makes Oklahoma an ideal research setting. Beyond just this rapid proliferation, Oklahoma is somewhat of a surprise given that it is generally considered to be a strongly conservative state with much of its population identifying with the Republican Party. Voters in Oklahoma have consistently supported the Republican presidential candidate in every election since 1968, and in the 2020 US presidential election, barely two years after SQ 788 was enacted into state law, each of the 77 counties in Oklahoma voted in majority for the Republican presidential candidate. Since 2004, political party affiliation has been shown to be a statistically significant determinant in measuring support for the legalization of medical marijuana, with those who identify as Republicans expressing significantly less support than those who identify as Democrats or Independents (Denham [2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR15 "Denham BE (2019) Attitudes toward legalization of marijuana in the United States, 1986â2016: changes in determinants of public opinion. Int J Drug Policy 71:78â90")).[Footnote 1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn1) We will also show, in our descriptive statistics, that the medical marijuana dispensaries located across Oklahoma are not clustered purely in downtown areas; instead, the dispensaries are broadly located throughout the cities in Oklahoma as well as in both heavily urban and more rural areas. Therefore, the combination of the rapid increase in the quantity of medical marijuana dispensaries in Oklahoma, the strongly Republican electorate, and the broad scattering of dispensaries across the state makes Oklahoma an ideal context in which to explore the property value effects of legalized medical marijuana.
To date, there has been relatively little research done on the effects of (legalized) marijuana dispensaries on residential property prices, and so far, the empirical evidence is mixed. Cheng et al. ([2018](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR11 "Cheng C, Mayer WJ, Mayer Y (2018) The effect of legalizing retail marijuana on housing values: evidence from Colorado. Econ Inq 56(3):1585â1601")); Burkhardt and Flyr ([2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR9 "Burkhardt J, Flyr M (2019) The effect of marijuana dispensary openings on housing prices. Contemp Econ Policy 37(3):462â475")) and Conklin et al. ([2020](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR13 "Conklin J, Drop M, Li H (2020) Contact high: the external effects of retail marijuana establishments on house prices. Real Estate Econ 48(1):135â173")) all find that legalized marijuana leads to a measurable increase in property values in Colorado, while (Boswell [2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR5 "Boswell K (2021) Essays on Cannacis legalization in California. PhD Thesis, University of Miami")) finds a residential property value increase in California. Cheng et al. ([2018](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR11 "Cheng C, Mayer WJ, Mayer Y (2018) The effect of legalizing retail marijuana on housing values: evidence from Colorado. Econ Inq 56(3):1585â1601")) find a 6 percent increase of housing values, and Burkhardt and Flyr ([2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR9 "Burkhardt J, Flyr M (2019) The effect of marijuana dispensary openings on housing prices. Contemp Econ Policy 37(3):462â475")) find that a marginal dispensary located within a half-mile of a new home increases home prices by about 7.7 percent. Conklin et al. ([2020](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR13 "Conklin J, Drop M, Li H (2020) Contact high: the external effects of retail marijuana establishments on house prices. Real Estate Econ 48(1):135â173")) study the conversion of medical marijuana dispensaries to recreational marijuana dispensaries, and find that single-family properties located within 0.1 miles of a converted dispensary realize an 8.4 percent increase in property value. They suggest that lower crime rates associated with dispensaries and higher housing demand associated with dispensary business activity may drive the boost in property values. Tyndall ([2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR32 "Tyndall J (2021) Getting high and low prices: marijuana dispensaries and home values. Real Estate Econ 49(4):1093â1119")), on the other hand, provides evidence to the contrary. Using a sample of repeat home sales from Vancouver, BC, he finds a decrease in property values ranging from 3.7 to 4.9 percent (depending on the regression specification) for properties located within 100 m of a dispensary.
A related literature focuses on the spatial distribution and economic geography of dispensary locations to understand the extent to which dispensaries are viewed as undesirable. Boggess et al. ([2014](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR4 "Boggess LN, PĂ©rez DM, Cope K, Root C, Stretesky PB (2014) Do medical marijuana centers behave like locally undesirable land uses? Implications for the geography of health and environmental justice. Urban Geogr 35(3):315â336")) find that medical marijuana dispensaries in Denver do not correlate with neighborhood racial composition or income and conclude that medical marijuana dispensaries are not locally undesirable.[Footnote 2](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn2) They do find that dispensaries are more likely to be located in areas with higher crime rates and greater density of retail employment. Somewhat differently, Morrison et al. ([2014](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR25 "Morrison C, Gruenewald PJ, Freisthler B, Ponicki WR, Remer LG (2014) The economic geography of medical cannabis dispensaries in California. Int J Drug Policy 25(3):508â515")) find that medical marijuana dispensaries in California are more likely to be located in lower income areas, as well as areas with a greater number of alcohol outlets, and Brinkman and Mok-Lamme ([2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR6 "Brinkman J, Mok-Lamme D (2019) Not in my backyard? Not so fast. The effect of marijuana legalization on neighborhood crime. Reg Sci Urban Econ 78:103460")) show that dispensaries are more likely to be located in Hispanic areas. Most related to our current work is Zimmer et al. ([2022](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR34 "Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University")) and Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")), both of which study dispensary siting in Oklahoma. Zimmer et al. ([2022](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR34 "Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University")) find that medical marijuana dispensaries in Oklahoma are more likely to locate in areas with higher rates of crime, a greater number of uninsured individuals or individuals on disability, and in closer proximity to the Texas border. Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")) find that Oklahoma dispensaries are more likely to be located in areas of lower socioeconomic status (racial minority and low-income) and in areas with more limited access to pharmacies.
Finally, some of the ânot in my backyardâ (or, NIMBY) literature is relevant to our work. Most immediately related is Iannacchione et al. ([2020](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR20 "Iannacchione B, Ward KC, Evans MK (2020) Perceptions of NIMBY syndrome among Colorado and Washington dispensary owners and managers. Justice Policy J 17(2)")) who conducted a survey of dispensary operators (owners or managers) in both Colorado and Washington to assess the extent to which local residents may have been opposed to the opening of the dispensary; that is, the extent of NIMBY effects surrounding legalized medical or recreational marijuana dispensaries. They find that the NIMBY effect of legalized marijuana dispensaries is generally non-existent, and the cases in which dispensary operators faced pushback was short-lived and typically driven by a lack of education about marijuana dispensaries and/or uncertainty (that soon subsided). On the political affiliation side of NIMBY research, Whittemore and BenDor ([2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR33 "Whittemore AH, BenDor TK (2019) Reassessing NIMBY: the demographics, politics, and geography of opposition to high-density residential infill. J Urban Aff 41(4):423â442")) found that although both liberals and conservatives were likely to be concerned with changes in property values, conservatives were significantly more likely to show concern (at least in that context of new residents within the context of high-density residential infill). Extrapolating to our context, we might find evidence of harmful effects of dispensaries on nearby property values given the more conservative population.[Footnote 3](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn3)
We consider models with data pooled across counties, as well as county-specific regressions; we consider both linear and nonlinear regression specifications. While we find some heterogeneity in the estimated effects of dispensaries on local home prices, a broad trend that emerges is that many residents prefer not to live in immediate proximity to a dispensary but prefer to maintain access to a dispensary within a moderate distance from the home (about 1â2 kilometers away). There are some instances in which the dispensaries do not have a significant effect on property values, and some cases wherein dispensary proximity raises housing values. These different results are discussed and in the context of different counties and relative robustness of the regression results (e.g., sample size).
Our work is important in at least two ways. First, we contribute to the measurement of the property impacts of legalized marijuana activity in Oklahoma. These estimates are important for citizens and policymakers in Oklahoma interested in understanding some of the effects of SQ 788 on local neighborhoods, particularly since we find evidence that many homeowners prefer not to live in close proximity to a dispensary despite the bill passing in Oklahoma with broad public support. Second, our estimates are also relevant for citizens and policymakers in other states that have already legalized marijuana, as well as in states in which legislation to legalize marijuana is (or may soon be) proposed. In either case, understanding (some of) the effects of legalized marijuana is of critical policy-relevance, and as motivated, Oklahoma is uniquely situated to provide insights that are relevant in the broader context. Finally, readers should recognize that our work focuses on localized housing market impacts of legalized medical marijuana and is not an overall welfare measurement.
## 2 Empirical approach
### 2\.1 Hedonic price model
Our regression model is a standard hedonic pricing model (e.g., Rosen [1974](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR28 "Rosen S (1974) Hedonic prices and implicit markets: product differentiation in pure competition. J Polit Econ 82(1):34â55"); Palmquist [1991](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR26 "Palmquist R D (1991) Hedonic methods. Measuring the demand for environmental quality"); Taylor [2003](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR30 "Taylor LO (2003) The hedonic method. A primer on nonmarket valuation. Springer, Cham, pp 331â393"); Parmeter and Pope [2013](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR27 "Parmeter C F, Pope J C (2013) Quasi-experiments and hedonic property value methods. In: Handbook on Experimental Economics and the Environment, pp 3â66. Edward Elgar Publishing")) defined as
\$\$\\begin{aligned} \\ln Price\_{it} = \\beta \_0 + Disp\_{i}\\beta \_1 + X\_{i} \\delta + v\_t + \\varepsilon \_{it} \\hspace{2em} i=1,2,\\dots ,n; \\hspace{1em} t=1,2,\\dots ,T \\end{aligned}\$\$
(1)
where \\(\\ln Price\_{it}\\) is the natural log of the price of property *i* at the time of sale, *t*, \\(Disp\_{i}\\) is a vector of variables measuring the proximity of the property to the dispensaries with corresponding parameter vector \\(\\beta \_1\\), \\(X\_{i}\\) is a *K*\-dimensioned vector of property- and neighborhood-specific control variables with \\(\\delta \\) as a corresponding vector of parameters, \\(v\_t\\) is a year-specific effect common to all properties, and \\(\\varepsilon \_{it}\\) is the mean-zero error term.[Footnote 4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn4) Note that we do not deflate property prices and instead prefer to control for differences in price levels across years through \\(v\_t\\). We are primarily interested in \\(\\beta \_1\\) as this parameter vector measures the impact of the dispensaries on the value of nearby properties.
### 2\.2 Proximity to dispensaries
The proximity vector, \\(Disp\_{i}\\), contains several different measures of property proximity to the dispensaries to fully explore the effect of the dispensaries on property values. Our empirical strategy follows, for instance, Delgado et al. ([2016](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR14 "Delgado MS, Guilfoos T, Boslett A (2016) The cost of unconventional gas extraction: a hedonic analysis. Resour Energy Econ 46:1â22")) in that we consider both continuous and discrete measures of proximity, both individually and separately.
The first measure of proximity that we use is the continuous distance in meters from each property to the nearest dispensary, defined as *minDist*. Recent research focusing on hedonic regression methodology has proposed data-driven techniques for determining the optimal distance by which to model the effect of an amenity on housing prices. Specifically, Fitzpatrick and Parmeter ([2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR17 "Fitzpatrick LG, Parmeter CF (2021) Data-driven estimation of treatment buffers in hedonic analysis: an examination of surface coal mines. Land Econ 97:528â547")) propose a cross-validation procedure whereby the optimal distance is selected via a numerical optimization procedure; we apply their method in selecting the optimal distance of dispensary effect based on the *minDist* variable, and then use this optimal distance as a way of filtering our sample into the region of effect (within the threshold) and the relevant comparison area (within two times the threshold). We explore regression specifications with *minDist* measured both linearly and via a quadratic specification, with the latter allowing us to detect nonlinearities in the relationship between proximity to the nearest dispensary and the property price. Tyndall ([2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR32 "Tyndall J (2021) Getting high and low prices: marijuana dispensaries and home values. Real Estate Econ 49(4):1093â1119")), for example, finds localized effects of dispensary proximity on property values, and a nonlinear (quadratic) distance specification will help us capture such localized effects more readily than a linear specification. Using a continuous distance measure also allows us to recover the shape of the distance-impact gradient to determine over exactly what range of distance the dispensaries affect property values (if at all), and how rapidly the effect dissipates (if at all).
An alternative way to measure the effect of the dispensaries on property values is to measure the intensity of the dispensary proximity to each property using pre-specified distance bands. We select distance bands of 500 m, 1000 m, and 2000 m and count the number of dispensaries that are within each distance band from each property as our measures of dispensary intensity. We denote these variables as *Count*500, *Count*1000, and *Count*2000 and consider regression specifications with these variables included together as well as separately.[Footnote 5](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn5) It is important to understand the interpretation of the regression coefficients when these variables are included individually or jointly: when these coefficients are included individually in the regression, the coefficient is simply the effect on \\(\\ln Price\\) of an additional dispensary being added within the distance band. For instance, if we include *Count*1000 in the regression, then the coefficient reflects the effect of adding another dispensary within 1000*m* (or 1*km*) of the property on the \\(\\ln Price\\) of the property (control variables held constant). On the other hand, if we include all three count variables in the model jointly, then the coefficient on *Count*1000 reflects the effect of adding another dispensary within 1000*m* of the property on the \\(\\ln Price\\) of the property *holding constant* the number of dispensaries within 500*m* and 2000*m* of the property (all other control variables being held constant as well). In other words, the coefficient on *Count*1000 now captures the effect on \\(\\ln Price\\) of adding another dispensary in the \\(\>500m\\) but \\(\\le 1000m\\) distance range.
Finally, it is worth clarifying that a regression specification including both the continuous and discrete proximity measures together via \\(Disp=(minDist,Count500, Count1000,Count2000)\\) bears the following interpretation. The coefficient on *minDist* is the effect on \\(\\ln Price\\) of moving the dispensary one meter away from the property holding constant the number of dispensaries located within 500*m*, 1000*m*, and 2000*m* from the property (in addition to all other variables in the model being held constant). That is, we interpret the distance effect with intensity held constant. We interpret the coefficients on the count variables, say *Count*1000, as the effect of adding one more dispensary at a distance \\(\>500m\\) but \\(\\le 1000m\\) while holding the distance to the nearest dispensary constant. That is, an increase in intensity within the specified distance band while holding the distance to the nearest dispensary constant.
By considering the continuous *minDist* measure of proximity and the three count measures of intensity both individually and jointly, we are able to fully explore the nature of dispensary proximity on nearby property values. It is worth noting that further exploring these regressions separately by county allows us to explore the extent to which these relationships are heterogeneous in different parts of Oklahoma.
### 2\.3 Control variables and fixed effects
Of course, the extent to which our estimates of \\(\\beta \_1\\) are reliable depends on our ability to adequately control for other factors that affect \\(\\ln Price\\) and also correlate with the effect of the dispensaries on property values. In our main regression specifications, we deploy control variables and fixed effects to account for these other factors. Being a hedonic price model, we naturally control first for housing characteristics. In our primary dataset, we control for the number of bedrooms, the number of bathrooms, the age of the property, and the square footage of the property. With regard to the age of the property and square footage, we find the best fit via a quadratic specification. It is important to note that these particular variables are common across the four counties we consider jointly in our primary specifications; we explore additional covariate controls as the data are available in auxiliary regressions that are county-specific.[Footnote 6](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn6)
Beyond these house-specific characteristics, we include the distance to the central business district in kilometers, as well as the average sales price of other properties within a 500 m radius of each property sold *prior* to SQ 788. This variable is particularly important for identification, as it allows us to control for neighborhood-level differences that might otherwise be correlated with the location of dispensaries. We noted earlier that Zimmer et al. ([2022](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR34 "Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University")) find that the location of medical marijuana dispensaries in Oklahoma is correlated with distance to Texas (where both medical and recreational marijuana is illegal), and rates of crime, the population that is uninsured, or has a disability, and Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")) find that Oklahoma dispensary locations correlate with fewer hospitals, being uninsured, and having a lower socioeconomic status (defined by either race or income). In Denver, medical marijuana dispensary locations correlate with higher crime rates and having a higher rate of retail sale establishments (Boggess et al. [2014](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR4 "Boggess LN, PĂ©rez DM, Cope K, Root C, Stretesky PB (2014) Do medical marijuana centers behave like locally undesirable land uses? Implications for the geography of health and environmental justice. Urban Geogr 35(3):315â336")). Regardless of the extent to which any of these particular factors causally affect dispensary location, it is clear that medical marijuana dispensariesâin Oklahoma and elsewhereâcorrelate with neighborhood factors. These factors are surely reflected by differences in neighborhood real estate prices, and so by controlling for average neighborhood property prices prior to SQ 788, we are able to adjust for differences in neighborhood effects that might otherwise correlate with dispensary locations.
In terms of the fixed effects, we control for the year the property was sold to account for macro-effects that are common to all properties sold within each year, as well as county-specific dummies to capture differences across counties.
### 2\.4 Auxiliary regression specifications
We consider auxiliary regression specifications that serve as robustness checks. Specifically, we separate the data by county and consider county-specific regressions. This robustness check does two things. First, this approach allows us to estimate county-specific parameters for all of the coefficients in the model, to determine the extent to which the estimated relationships vary by county. Second, the data we obtained from the county assessor offices contain different sets of control variables, and by considering each county individually, we are able to add additional control variables that are available for each particular county. While the county-specific model specifications are not dramatically different, it is worth exploring the importance of these different control variables where possible. More specific details will be provided in subsequent sections where we explore these data and regressions.[Footnote 7](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn7)
## 3 Data
### 3\.1 Housing data
Residential property sales data were obtained from six county assessor offices in the State of Oklahomaâas noted in the introduction, Comanche, Oklahoma, Payne, Tulsa, Wagoner and Washington counties. The data include all single-family property sales from the beginning of 2016 to the end of 2023, providing information on the sales price, geographic location, and property attributes. In terms of the property attributes, the data includes typical housing attribute variables that are commonly found in the hedonic housing literature, including the age of the property measured in years, living area in square feet, the number of bedrooms, and the number of bathrooms. There is slight variation across county-specific datasets in terms of additional housing-level variables. Specifically, the data from Payne and Tulsa counties lack key variables, and so we consider the property sales from these counties in auxiliary regressions but not as part of our main benchmark models. Specifically, our data for Payne County lack data on the age of the property and property square footage, and our data for Tulsa County lacks data on the number of bedrooms of each property.
### 3\.2 Dispensary data
Medical marijuana dispensary data are made publicly available to view and download online by the Oklahoma Medical Marijuana Authority (OMMA). Publicly available data as of 2024 include the name of the licensed dispensary, the license identification number, as well as the county and city the dispensary operates in within the state. These data are used by Zimmer et al. ([2022](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR34 "Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University")) and Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")), with Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")) noting that they conducted online searches using the dispensary names provided by the OMMA to further identify exact geographical locations and current operational status. In our case, we use publicly available data provided by the Oklahoma Watch organization website (Brown and Howe [2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR8 "Brown T, Howe J (2019) Number of medical marijuana outlets soars. How many are in your area?")) that is maintained by Trevor Brown and Jesse Howe; their data were obtained through the OMMA and mapped using Googleâs My Maps service.[Footnote 8](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn8) The data include 1,463 unique medical marijuana dispensary locations from across the entire state.
It is important to note that with these data, we do not know the date at which each dispensary opened. To link the dispensaries to the property sales, in both the benchmark and auxiliary datasets we include only property sales after July 2020 because the dispensary data we use comes from a cross-section of dispensaries that were in operation as of July 2020. By focusing on these particular sales transactions, we are able to ensure that our linkage of sales transactions to dispensary locations is concurrent in time.
### 3\.3 Geolocation and mapping
For each county, each property comes with a geolocating identifier. Counties that provided a street address for each home were geocoded into latitudinal and longitudinal coordinates using Esri ArcGIS Pro 3.1. We use the geographical location of the property to compute the distance to the nearest central business district (CBD): we use the âgeosphereâ package in R 4.2.2 to compute in kilometers the distance between the property and the closest CBD, which, in our data, is either downtown Oklahoma City or downtown Tulsa. We also use these locators to compute the distance in meters to the closest dispensary, as well as the number of dispensaries in binned radii around the property to provide a count of how many exist within 500, 1000, and 2000 m. We also use these geographical locators to construct the historical neighborhood value variable: We construct a 500 m radius around each property and then compute the average sale price of all properties sold within the radius between January 2016 and July 2018. These years correspond to our sales data prior to the opening of any dispensaries, thereby providing a market value-based measure of historical neighborhood quality. We also report the average number of properties within this 500 m radius (the âdensityâ variable) for readers to understand the rough size of these local neighborhoods.
### 3\.4 Descriptive statistics
Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1) provides a descriptive summary of the 4-county benchmark data (the pooled sample) as well as the data broken down by county. More detailed descriptive statistics are in the appendix.
**Table 1 Descriptive statistics for the pooled sample of four counties and separately by county**
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/1)
Across the pooled sample, the average sale price is around \$253,000, with average prices being slightly higher in Oklahoma and Tulsa counties, and lower in Comanche, Wagoner and Washington counties. These properties are, on average, three bedroom and two bathroom homes, that are around 40 years old and have just under 2000 square feet of living space. The average age of the home, in particular, varies quite a bit across counties. From the counties with available data, these are typically one story homes. We measure the central business district (CBD) as either Oklahoma City or Tulsa, and from Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1) we see that the average distance to CBD is just under 40 kilometers.
The average historical neighborhood value is around \$203,000, which is roughly \$50,000 lower than the average sale price, and this computation comes from a local neighborhood density of, on average, 24.8 home sales. These numbers do not account for increases in property values through the intervening time period; this is an unconditional comparison of average values. To dig deeper into the relationship between historical neighborhood values and current property valuesâand thereby to provide some empirical justification for neighborhood to serve as an important control variable in our analysis and thereby aid in our identification strategyâwe turn to Fig. [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig1) and Table [2](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab2). In Fig. [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig1), we show a scatterplot correlation between historical neighborhood property values and current property prices, and Table [2](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab2) reports univariate regression coefficients (i.e., correlations) between these variables. We can see a strong, positive correlation between neighborhood prices pre-dispensaries and property values post-dispensaries; from the table, we can see that this correlation is statistically significant with a coefficient of about 0.7.
**Fig. 1**
[](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/1)
Scatterplot of historical neighborhood values and current property prices
[Full size image](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/1)
**Table 2 Univariate correlation between neighborhood values and \\(\\ln (Price/100)\\)**
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/2)
In the spirit of Boggess et al. ([2014](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR4 "Boggess LN, PĂ©rez DM, Cope K, Root C, Stretesky PB (2014) Do medical marijuana centers behave like locally undesirable land uses? Implications for the geography of health and environmental justice. Urban Geogr 35(3):315â336")); Zimmer et al. ([2022](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR34 "Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University")) and Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")), we consider the correlation between *Count*500 and historical neighborhood value. Figure [2](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig2) provides a scatterplot and estimated trend line, and Table [3](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab3) provides the univariate regression estimate. There is a clear downward trend, whereby as the number of dispensaries within 500 m increases, historical neighborhood value decreases; the regression table indicates that this correlation is statistically significant. Further, our estimated correlation is linear, though the figure clearly suggests a nonlinear pattern whereby the largest downward effect of dispensary proximity occurs with the siting of the first one or two dispensaries within 500 m, with a dramatically reduced effect with marginal increases in dispensary intensity. Of course, this is merely a correlationâthus, a descriptive statistic that motivates our use of neighborhood as a control variableâand should not be taken as evidence that dispensaries are located in areas characterized in any particular way (e.g., by race, income, or other status).
Turning to the dispensary-related variables, we see that, on average, the minimum distance to the nearest dispensary is about 1.4 kilometers (Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1)). These properties have, on average, 0.34 dispensaries within a 500 m radius of the property; 1.5 dispensaries within 1 km of the property; and 5.7 dispensaries within 2 kilometers of the property. It is worth noting that Washington, Oklahoma and Tulsa counties have properties that are in closest proximity, on average, to dispensaries and with the highest average density of proximate dispensaries.
**Fig. 2**
[](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/2)
Scatterplot between *Count*500 and neighborhood value before legalization
[Full size image](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/2)
**Table 3 Univariate correlation between *Count*500 and neighborhood**
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/3)
### 3\.5 Mapping the data
**Fig. 3**
[](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/3)
Property sales and dispensary locations across Oklahoma. Red pins indicate dispensary locations, green pins indicate property locations for our four-county benchmark sample, and purple pins denote property locations from our auxiliary samples (Tulsa and Payne counties)
[Full size image](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/3)
We have described the rapid extent to which medical marijuana dispensaries have opened in Oklahoma since 2018, and the following figures help to illustrate this fact. In Fig. [3](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig3) we map all the properties in our dataset against the locations of all the dispensaries across the State of Oklahoma. Specifically, green circles represent the location of an individual residential sale from the benchmark dataset, and the red pins represent the locations of the dispensaries that were operating as of July 2020. It is clear from the map that the dispensaries are located all across Oklahoma, covering both urban and rural areas; given the short time span since the passage of the law, the number of operating dispensaries has quickly risen.
To gain deeper insight into the locations of the property sales and dispensaries, we plot the Oklahoma County data in Fig. [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig4). We can clearly see the dispensaries located broadly across Oklahoma County (primarily this is Oklahoma City), located not just in the city center but in suburban areas as well.
**Fig. 4**
[](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/4)
Property sales and dispensary locations in Oklahoma County. Red pins indicate dispensary locations, and green pins indicate property locations
[Full size image](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/4)
## 4 Results
### 4\.1 Benchmark regression analysis
Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4) displays the results from our benchmark regression model that includes both *minDist* and all three *Count* variables (*Count*500, *Count*1000, *Count*2000\); a model that includes *minDist* but excludes the three *Count* variables; a model that is quadratic in *minDist* and excludes the *Count* variables; and a model that is quadratic in *minDist*, interacts *minDist* with historical neighborhood values, and excludes the three *Count* variables. The properties included in these regressions are all within two times the optimal cross-validated distance threshold, which for the pooled sample is 3110 m. These models allow us to first set a benchmark for the housing market impacts of dispensary activities (Column 1) and then gain a deeper understanding into the relationship between *minDist* and property values (Columns 2â4). The data used in these regressions are the pooled sample of four counties for which we have complete and overlapping control variables (i.e., Comanche, Wagoner, Washington and Oklahoma counties). In all four models, we include housing characteristic variables that includes quadratic specifications for both age and square footage, as well as year of sale and county indicators.
**Table 4 Benchmark model and sub-models to further assess *minDist***
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/4)
Looking first at the benchmark model shown in Column 1 in Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4), we see that *minDist*, *Count*1000 and *Count*2000 are statistically significant. *Count*500 is not significant. Further, *minDist* and *Count*2000 have a positive coefficient, whereas *Count*1000 has a negative coefficient. Interpreting first the *minDist* coefficient: the model is a semi-log model, which means that a one-meter increase in the distance from the nearest dispensary leads to a \\(\\hat{\\beta }\\times 100\\) percent increase in the sale price of the property (in the regression, prices are in hundreds of dollars). Our estimate thus implies a 0.002197 percentage change in price in hundreds with every one-meter increase in distance. Recalling from Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1) that the average property price is 253,060.30, this estimated marginal effect implies that a one-meter increase in distance to nearest dispensary leads to a 5.56 dollar increase in sale price after converting into dollars. Again, bear in mind that this is controlling for the *Count* variables that capture dispensary intensity as well as the other control variables. If we extrapolate to a 100-m increase in distance to nearest dispensary, this is a property price increase of about 555.97 dollars. These estimates indicate that dispensaries are on net undesirable, though this is not a substantial price increase.
We turn next to the *Count* variables and their estimated impacts on property prices. Recall that since all three variables are included in the regression, the interpretation of each *Count* variable coefficient is incremental: adding one more dispensary within a 500 m radius of a property, holding constant the number of dispensaries within the 500â2000 m range, leads to a \\(\\hat{\\beta }\_{Count500}\\) change in a propertyâs price in hundreds of dollars or a \\(\\hat{\\beta }\_{Count500} \\times 100\\) percentage change. Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4) shows that *Count*500 is insignificant, while *Count*1000 is significantly negative and *Count*2000 is significantly positiveâholding *minDist* and other controls constant. Specifically, our estimates indicate that, all else constant, adding an additional dispensary within the 500â1000 m distance from a property leads to a decrease in the average property price by about 0.429 percent; from the average home price, this is a decrease of about 1085.63 dollars. At the same time, the coefficient on *Count*2000 indicates that adding one more dispensary in the 1000â2000 m range, holding *minDist*, the other *Count* variables, and controls constant, results in a 0.427 percent increase in property values (or about 1080.57 dollars at the average). Taken at face value, these estimates indicate that homeowners prefer to have a marginal dispensary located farther away from the home.
It is worth discussing the effects and significance of the control variables on property prices before digging deeper into the dispensary exposure variables and their relationship to property prices. Still looking at the benchmark model in Column 1 of Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4), we see statistically significant quadratic relationships between the age and square footage of the property and property values. In particular, property age has an inverse-U-shaped relationship with property prices, with a minimum property value occurring at a property age of 70 years. The average age of the home in the pooled sample is 40 years (Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1)). Square footage has a U-shaped relationship with property values, with the maximum occurring at 15,046 square feet. From Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1), the average square footage in the pooled benchmark sample is 1904.99, and the detailed descriptive statistics tables in the appendix indicate that this turning point is greater than the maximum square footage in our data; in effect, the square footage effect on property prices is monotonically increasing but at a decreasing rate. We see that the number of bedrooms is significantly negative; given that we control for square footage, the negative coefficient on bedrooms means that home buyers prefer fewer but larger rooms for a home of a given size. The number of bathrooms and neighborhood home density are significantly positive, while distance to the nearest CBD is significantly negative. The historical neighborhood value is significantly positive, which is in line with the correlations shown in Fig. [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig1) and Table [2](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab2), and indicates that historical average neighborhood values are positively and significantly correlated with current sales prices. This estimate, therefore, reinforces our intuition to include this variable as an important control for neighborhood quality that in part explains current property prices but might also correlate with dispensary activity (distance and exposure). Finally, we see some significance of the year of sale indicators relative to the baseline year of 2023, and significance of the county indicators relative to the Comanche County baseline. We note that the \\(R^2\\) is 0.640, reflecting a reasonable fit of the regression.
### 4\.2 Distance to the nearest dispensary
Moving beyond the benchmark regression, in Columns 2â4 of Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4) we explore more deeply the relationship between *minDist* and property values. In Column 2, we remove the *Count* variables and explore the linear effect of *minDist* on the natural log of property values: We find the results remain qualitatively unchanged from the benchmark regression (Column 1), with a significantly positive effect of *minDist* on property values. This estimate is fairly smaller in magnitude compared to the estimate in Column 1, this time implying an effect of 3.71 dollars per meter, or about 371 dollars for a 100 m distance.
In Column 3, we consider a quadratic specification and find that only the quadratic term is statistically significant, and in Column 4 we consider both the quadratic *minDist* specification including an interaction with the historical neighborhood index. In this latter model, we find that all coefficients are significant. Exploring these estimates further, using both the average property value and average neighborhood value index (203,309.40 from Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1)), we find that on average the *minDist* marginal effect on property prices is about 224 dollars per 100 m. In looking across the models in Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4), we see that sharpening our estimates (i.e., Column 4) results in a smaller estimated average effect of *minDist* on property prices. Yet, from the perspective of these models, dispensaries are viewed by homeowners as a negative amenity.
### 4\.3 Intensity of dispensary exposure
**Table 5 Models to further assess significance of count variables**
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/5)
Next, we look more deeply into the relationship between the *Count* variables and property values; regression results are found in Table [5](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab5). The first model is a version of the benchmark regression from Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4) except that we omit *minDist*. Models 2â4 consider the three *Count* variables separately, one at a time. The dependent variable remains \\(\\ln (Price/100)\\).
The results in the first column are very similar to the results shown in the benchmark regression (Column 1 of Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4)). That is, removing *minDist* does not change the results from the benchmark model. This result mirrors Column 2 of Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4) in that removing the *Count* variables did not change the effect of *minDist* on property price. That is, we continue to find an insignificant coefficient on *Count*500, a significantly negative coefficient on *Count*1000, and a significantly positive coefficient on *Count*2000 all with magnitudes very similar to those discussed in the previous subsection.
It is interesting that when considering the *Count* variables separately, as shown in Models 2â4, only *Count*2000 remains significant (and still has a positive coefficient). It is important to remember that in these three regressions, the interpretations of the *Count* variables coefficients are different from the joint model. Since the other bands are not held constant, the coefficient on *Count*1000, for instance, indicates the effect on log price of an additional dispensary located anywhere within 1000 m of the property, whereas previously it pertained only to an additional dispensary being located within the 500â1000 m distance band from the property. When taken separately, these intensity variables indicate that a change in intensity of dispensary exposure does not significantly affect property values within 500 m (see Column 1) or up to 1000 m (see Columns 2â3) of the property. A marginal increase in intensity up to 2000 m from the property results in an increase in the property value of about 0.19 percent, or about 480.81 dollars based on the average property sale price. Broadly speaking, these results also indicate that residents prefer a marginal dispensary to be located farther from the property, though they may be indifferent to a marginal dispensary being located nearer to the property.
### 4\.4 County-specific regressions
**Table 6 County-specific benchmark regressions**
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/6)
Thus far we have focused exclusively on both distance and intensity measures of dispensary exposure for our four-county benchmark sample. To ascertain the extent to which there may be heterogeneity in the dispensary effects across counties, and to exploit data on two other counties, we now divide the sample and turn to county-specific regressions. Note that from the four benchmark counties, the majority of observations come from Oklahoma County, with a large group also coming from Comanche County. Relatively, few observations come from Wagoner and Washington counties. Thus, it is interesting to assess the extent to which the pooled sample results merely reflect Oklahoma and perhaps Comanche counties, or are representative of all counties more generally. For these county-specific models, we are no longer bound by a common set of control variables across counties; for each county-specific regression, we include as many control variables as are available for that county. Also, for each county we use the Fitzpatrick and Parmeter ([2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR17 "Fitzpatrick LG, Parmeter CF (2021) Data-driven estimation of treatment buffers in hedonic analysis: an examination of surface coal mines. Land Econ 97:528â547")) cross-validation method to select the optimal *minDist* threshold, and then use all property sales within 2 times that threshold for our county-optimal sample. We report the county-specific regression results in Table [6](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab6). To save space, we withhold estimates of all control variables, and provide the estimates of the dispensary effects.[Footnote 9](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn9)
Looking first the the coefficients on *minDist* across Table [6](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab6), we see that *minDist* is significantly negative in Comanche and Payne counties, insignificant in Wagoner and Washington counties, and significantly positive in Oklahoma and Tulsa counties. *Count*500 is only significant in Tulsa County, wherein we find the estimate is positive. *Count*1000 is significantly negative in Oklahoma County and significantly positive in Tulsa County, while *Count*2000 is significantly positive in Oklahoma County and significantly negative in Payne County. Taking stock of the first four columns, relative to the benchmark pooled sample regressions, it seems apparent that the results found with the four pooled counties are largely driven by the Oklahoma County effects. Table [6](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab6) shows that residents in Comanche County experience a negative *minDist* effect, indicating that property prices fall when dispensaries are located farther away from the home, though the intensity of dispensary activity does not affect property values. Looking toward Payne and Tulsa counties, we see that Payne County bears similar effects to Comanche County, while Tulsa County accords with Oklahoma County. It is worth pointing out the following. First, there is quite a bit of heterogeneity across counties in terms of both the optimal *minDist* threshold and the \\(R^2\\) measures for each regression, and readers should bear in mind that both Payne and Tulsa counties are missing several key variables (which is why they were left out of the baseline four-county model). In terms of the \\(R^2\\), Oklahoma County seems to have the best fit, relative to the regressions for the other counties.
## 5 Discussion and conclusion
We study the effects of medical marijuana dispensaries on residential property prices in Oklahoma as a means of measuring the extent to which dispensaries are seen as desirable or undesirable neighborhood amenities and thereby affect property prices. We have motivated the uniqueness of Oklahoma for our study setting: not only has Oklahoma very recently legalized medical marijuana (effective August 2018) but Oklahoma has seen a rapid rise in the proliferation of medical marijuana dispensaries and has become a leader in the nation in terms of the number of operating dispensaries. Moreover, Oklahoma is an interesting case because of the strongly conservative electorate, despite marijuana legalization being typically seen as a politically liberal interest. Thus, we believe that Oklahoma is an ideal setting for studying the effects of medical marijuana legalization on nearby property values with lessons for citizens of Oklahoma as well as the public or policymakers in other states. Using data from six counties in Oklahoma, we conduct a state-wide hedonic pricing analysis to investigate the net effects, as capitalized in property values, of local dispensary activities on residential property prices. We measure dispensary activity with both distance to nearest dispensary as well as intensity of dispensary activity in the proximate area (within 500 m, 1000 m, and 2000 m from the property). We consider pooled sample regressions as well as county-specific regressions, linear and nonlinear specifications. In looking across these six counties, we find some heterogeneity in our results. However, our results point to a few broad trends across most of Oklahoma. First, a general pattern is that residents view local proximity to a medical marijuana dispensary as a net negative, with property prices being significantly lower when either the minimum distance to the nearest dispensary is closer to the home or when a marginal dispensary is opened within 0.5 kilometers from the home. We find the effects to be quite small, with even the largest effects being less than one percent of the propertyâs value. At the same time, residents seem to prefer relatively easy access to medical marijuana dispensaries, with property prices increasing for marginal dispensaries being located at a medium distance (i.e., between 1 and 2 kilometers) from the home. Second, it is worth noting that Washington and Wagoner counties generally have insignificant effects of dispensary activities on home prices, though the sample size in those counties is substantially smaller than for the other counties.
Our results are important for residents and policymakers alike. First, despite the majority voter approval for SQ 788, it appears that residents in Oklahoma have mixed views on dispensary activity. Many residents seem to not want to live in immediate proximity, whereas many want access. It is worth noting that thus far the literature has been mixed in terms of dispensary effects on local residents; our work adds the lens of the experience in Oklahoma, and find that most residents seem to view dispensaries within immediate proximity negatively. These differences in impacts are important for policymakers to recognize, and understand when making adjustments or revisions to Oklahoma laws. Finally, readers should keep in mind that our estimates are both local and net measurements, and do not specifically measure the impact of local dispensary activities on any singular measure of well-being (such as crime rates, related alcohol use, or child abuse) as considered by others in this literature.
## Notes
1. Interestingly, Denham ([2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR15 "Denham BE (2019) Attitudes toward legalization of marijuana in the United States, 1986â2016: changes in determinants of public opinion. Int J Drug Policy 71:78â90")) shows that prior to 2004, political ideology, and not so much political party affiliation, was a strong correlate with support for or against marijuana legalization. After 2004, political party affiliation became a primary factor, and not political ideology. For reference, though the correlation between political ideology and party affiliation has been steadily increasing since 1986, the correlation spanning the 1986-2016 period ranges from about 0.24 to 0.44 (Denham [2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR15 "Denham BE (2019) Attitudes toward legalization of marijuana in the United States, 1986â2016: changes in determinants of public opinion. Int J Drug Policy 71:78â90")).
2. Boggess et al. ([2014](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR4 "Boggess LN, PĂ©rez DM, Cope K, Root C, Stretesky PB (2014) Do medical marijuana centers behave like locally undesirable land uses? Implications for the geography of health and environmental justice. Urban Geogr 35(3):315â336")) classify locally undesirable land uses to be those that disproportionately affect racial/ethnic minorities or the poor.
3. The broader literature has also explored the effects of legalized marijuana on rates of violent and property crimes (see, e.g.,Kepple and Freisthler [2012](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR21 "Kepple NJ, Freisthler B (2012) Exploring the ecological association between crime and medical marijuana dispensaries. J Stud Alcohol Drugs 73(4):523â530"); Anderson et al. [2013](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR2 "Anderson DM, Hansen B, Rees DI (2013) Medical marijuana laws, traffic fatalities, and alcohol consumption. J Law Econ 56(2):333â369"); Adda et al. [2014](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR1 "Adda J, McConnell B, Rasul I (2014) Crime and the depenalization of cannabis possession: evidence from a policing experiment. J Polit Econ 122(5):1130â1202"); Morris et al. [2014](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR24 "Morris RG, TenEyck M, Barnes JC, Kovandzic TV (2014) The effect of medical marijuana laws on crime: evidence from state panel data, 1990â2006. PLoS ONE 9(3):e92816"); Huber et al. [2016](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR19 "Huber A III, Newman R, LaFave D (2016) Cannabis control and crime: medicinal use, depenalization and the war on drugs. BE J Econ Anal Policy 16(4):20150167"); Chang and Jacobson [2017](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR10 "Chang TY, Jacobson M (2017) Going to pot? The impact of dispensary closures on crime. J Urban Econ 100:120â136"); Brinkman and Mok-Lamme [2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR6 "Brinkman J, Mok-Lamme D (2019) Not in my backyard? Not so fast. The effect of marijuana legalization on neighborhood crime. Reg Sci Urban Econ 78:103460"); Lu et al. [2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR22 "Lu R, Willits D, Stohr MK, Makin D, Snyder J, Lovrich N, Meize M, Stanton D, Wu G, Hemmens C (2021) The cannabis effect on crime: time-series analysis of crime in Colorado and Washington State. Justice Q 38(4):565â595"); Thomas and Tian [2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR31 "Thomas D, Tian L (2021) Hits from the bong: the impact of recreational marijuana dispensaries on property values. Reg Sci Urban Econ 87:103655"); Dong and Tyndall [2024](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR16 "Dong X, Tyndall J (2024) The impact of recreational marijuana dispensaries on crime: evidence from a lottery experiment. Ann Reg Sci 72:1383â1414")), as well as on rates of alcohol abuse, traffic fatalities, and child abuse (see, e.g., Anderson et al. [2013](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR2 "Anderson DM, Hansen B, Rees DI (2013) Medical marijuana laws, traffic fatalities, and alcohol consumption. J Law Econ 56(2):333â369"), [2015](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR3 "Anderson DM, Hansen B, Rees DI (2015) Medical marijuana laws and teen marijuana use. Am Law Econ Rev 17(2):495â528"); Freisthler et al. [2015](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR18 "Freisthler B, Gruenewald PJ, Wolf JP (2015) Examining the relationship between marijuana use, medical marijuana dispensaries, and abusive and neglectful parenting. Child Abuse & Neglect 48:170â178"); Mair et al. [2015](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR23 "Mair C, Freisthler B, Ponicki WR, Gaidus A (2015) The impacts of marijuana dispensary density and neighborhood ecology on marijuana abuse and dependence. Drug Alcohol Depend 154:111â116"); Shi [2016](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR29 "Shi Y (2016) The availability of medical marijuana dispensary and adolescent marijuana use. Prev Med 91:1â7")).
4. While properties are sold in different years, property and neighborhood characteristics are not time-varying. Additionally, the dispensary proximity variables are not time-varying for reasons described in Sect. [3](https://link.springer.com/article/10.1007/s00181-026-02894-6#Sec7). Thus, *Price* is subscripted with *i* and *t* to indicate the year of sale and allow us to include \\(v\_t\\), but *Disp* and *X* are subscripted by only *i*. Thus, our sample is best thought of as a cross-sectional dataset (though properties are sold in different years).
5. See also Brooks et al. ([2018](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR7 "Brooks TJ, Humphreys BR, Nowak A (2018) Strip clubs, âsecondary effects\" and residential property prices. Real Estate Econ 48(3):850â885")) where they measured proximity to strip clubs via distance bins of 500, 1000, and 2000 feet.
6. As we describe in the next section, for reasons of data consistency we define our benchmark set of counties to be Comanche, Oklahoma, Wagoner and Washington counties. We use data for Tulsa and Payne counties in our set of robustness checks.
7. In previous versions of this manuscript, we considered an additional set of regressions based on repeat sales transactions. Ultimately, those sample sizes were too small and noisy to yield any meaningful results, and so those regression models have been dropped. Interested readers should contact the authors for more information about the repeat sales models.
8. We requested the exact latitudinal and longitudinal coordinates for the dispensaries on November 29, 2023 through the Open Records Request on the OMMA website that operates in accordance with the Oklahoma Open Records Act. On February 5, 2024, the OMMA denied and closed the request to provide locational data for dispensaries within the state stating that current Oklahoma law does not allow the OMMA to release the site locations of medical marijuana businesses for reasons of confidentiality.
9. In terms of additional variables per county, we note that Comanche County has a dummy variable indicating whether the property has a garage, and a township location indicator; Wagoner County has the number of stories per home, the garage dummy, amount of non-ag acreage of the property, and measure of property depreciation loss; Washington County has the number of stories and garage dummy; Oklahoma County has the number of stories and rooms in the home; Payne County lacks data the age and square footage of the home; and Tulsa County lacks data on square footage but has available the number of stories in the home.
## References
- Adda J, McConnell B, Rasul I (2014) Crime and the depenalization of cannabis possession: evidence from a policing experiment. J Polit Econ 122(5):1130â1202
[Article](https://doi.org/10.1086%2F676932) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Crime%20and%20the%20depenalization%20of%20cannabis%20possession%3A%20evidence%20from%20a%20policing%20experiment&journal=J%20Polit%20Econ&doi=10.1086%2F676932&volume=122&issue=5&pages=1130-1202&publication_year=2014&author=Adda%2CJ&author=McConnell%2CB&author=Rasul%2CI)
- Anderson DM, Hansen B, Rees DI (2013) Medical marijuana laws, traffic fatalities, and alcohol consumption. J Law Econ 56(2):333â369
[Article](https://doi.org/10.1086%2F668812) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Medical%20marijuana%20laws%2C%20traffic%20fatalities%2C%20and%20alcohol%20consumption&journal=J%20Law%20Econ&doi=10.1086%2F668812&volume=56&issue=2&pages=333-369&publication_year=2013&author=Anderson%2CDM&author=Hansen%2CB&author=Rees%2CDI)
- Anderson DM, Hansen B, Rees DI (2015) Medical marijuana laws and teen marijuana use. Am Law Econ Rev 17(2):495â528
[Article](https://doi.org/10.1093%2Faler%2Fahv002) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Medical%20marijuana%20laws%20and%20teen%20marijuana%20use&journal=Am%20Law%20Econ%20Rev&doi=10.1093%2Faler%2Fahv002&volume=17&issue=2&pages=495-528&publication_year=2015&author=Anderson%2CDM&author=Hansen%2CB&author=Rees%2CDI)
- Boggess LN, PĂ©rez DM, Cope K, Root C, Stretesky PB (2014) Do medical marijuana centers behave like locally undesirable land uses? Implications for the geography of health and environmental justice. Urban Geogr 35(3):315â336
[Article](https://doi.org/10.1080%2F02723638.2014.881018) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Do%20medical%20marijuana%20centers%20behave%20like%20locally%20undesirable%20land%20uses%3F%20Implications%20for%20the%20geography%20of%20health%20and%20environmental%20justice&journal=Urban%20Geogr&doi=10.1080%2F02723638.2014.881018&volume=35&issue=3&pages=315-336&publication_year=2014&author=Boggess%2CLN&author=P%C3%A9rez%2CDM&author=Cope%2CK&author=Root%2CC&author=Stretesky%2CPB)
- Boswell K (2021) Essays on Cannacis legalization in California. PhD Thesis, University of Miami
- Brinkman J, Mok-Lamme D (2019) Not in my backyard? Not so fast. The effect of marijuana legalization on neighborhood crime. Reg Sci Urban Econ 78:103460
[Article](https://doi.org/10.1016%2Fj.regsciurbeco.2019.103460) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Not%20in%20my%20backyard%3F%20Not%20so%20fast.%20The%20effect%20of%20marijuana%20legalization%20on%20neighborhood%20crime&journal=Reg%20Sci%20Urban%20Econ&doi=10.1016%2Fj.regsciurbeco.2019.103460&volume=78&publication_year=2019&author=Brinkman%2CJ&author=Mok-Lamme%2CD)
- Brooks TJ, Humphreys BR, Nowak A (2018) Strip clubs, âsecondary effects" and residential property prices. Real Estate Econ 48(3):850â885
[Article](https://doi.org/10.1111%2F1540-6229.12236) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Strip%20clubs%2C%20%E2%80%9Csecondary%20effects%22%20and%20residential%20property%20prices&journal=Real%20Estate%20Econ&doi=10.1111%2F1540-6229.12236&volume=48&issue=3&pages=850-885&publication_year=2018&author=Brooks%2CTJ&author=Humphreys%2CBR&author=Nowak%2CA)
- Brown T, Howe J (2019) Number of medical marijuana outlets soars. How many are in your area?
- Burkhardt J, Flyr M (2019) The effect of marijuana dispensary openings on housing prices. Contemp Econ Policy 37(3):462â475
[Article](https://doi.org/10.1111%2Fcoep.12414) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20effect%20of%20marijuana%20dispensary%20openings%20on%20housing%20prices&journal=Contemp%20Econ%20Policy&doi=10.1111%2Fcoep.12414&volume=37&issue=3&pages=462-475&publication_year=2019&author=Burkhardt%2CJ&author=Flyr%2CM)
- Chang TY, Jacobson M (2017) Going to pot? The impact of dispensary closures on crime. J Urban Econ 100:120â136
[Article](https://doi.org/10.1016%2Fj.jue.2017.04.001) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Going%20to%20pot%3F%20The%20impact%20of%20dispensary%20closures%20on%20crime&journal=J%20Urban%20Econ&doi=10.1016%2Fj.jue.2017.04.001&volume=100&pages=120-136&publication_year=2017&author=Chang%2CTY&author=Jacobson%2CM)
- Cheng C, Mayer WJ, Mayer Y (2018) The effect of legalizing retail marijuana on housing values: evidence from Colorado. Econ Inq 56(3):1585â1601
[Article](https://doi.org/10.1111%2Fecin.12556) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20effect%20of%20legalizing%20retail%20marijuana%20on%20housing%20values%3A%20evidence%20from%20Colorado&journal=Econ%20Inq&doi=10.1111%2Fecin.12556&volume=56&issue=3&pages=1585-1601&publication_year=2018&author=Cheng%2CC&author=Mayer%2CWJ&author=Mayer%2CY)
- Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113
[Article](https://doi.org/10.26828%2Fcannabis%2F2023.01.008) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Population%20and%20neighborhood%20correlates%20of%20cannabis%20dispensary%20locations%20in%20Oklahoma&journal=Cannabis&doi=10.26828%2Fcannabis%2F2023.01.008&volume=6&issue=1&pages=99-113&publication_year=2023&author=Cohn%2CAM&author=Sedani%2CA&author=Niznik%2CT&author=Alexander%2CA&author=Lowery%2CB&author=McQuoid%2CJ&author=Campbell%2CJ)
- Conklin J, Drop M, Li H (2020) Contact high: the external effects of retail marijuana establishments on house prices. Real Estate Econ 48(1):135â173
[Article](https://doi.org/10.1111%2F1540-6229.12220) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Contact%20high%3A%20the%20external%20effects%20of%20retail%20marijuana%20establishments%20on%20house%20prices&journal=Real%20Estate%20Econ&doi=10.1111%2F1540-6229.12220&volume=48&issue=1&pages=135-173&publication_year=2020&author=Conklin%2CJ&author=Drop%2CM&author=Li%2CH)
- Delgado MS, Guilfoos T, Boslett A (2016) The cost of unconventional gas extraction: a hedonic analysis. Resour Energy Econ 46:1â22
[Article](https://doi.org/10.1016%2Fj.reseneeco.2016.07.001) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20cost%20of%20unconventional%20gas%20extraction%3A%20a%20hedonic%20analysis&journal=Resour%20Energy%20Econ&doi=10.1016%2Fj.reseneeco.2016.07.001&volume=46&pages=1-22&publication_year=2016&author=Delgado%2CMS&author=Guilfoos%2CT&author=Boslett%2CA)
- Denham BE (2019) Attitudes toward legalization of marijuana in the United States, 1986â2016: changes in determinants of public opinion. Int J Drug Policy 71:78â90
[Article](https://doi.org/10.1016%2Fj.drugpo.2019.06.007) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Attitudes%20toward%20legalization%20of%20marijuana%20in%20the%20United%20States%2C%201986%E2%80%932016%3A%20changes%20in%20determinants%20of%20public%20opinion&journal=Int%20J%20Drug%20Policy&doi=10.1016%2Fj.drugpo.2019.06.007&volume=71&pages=78-90&publication_year=2019&author=Denham%2CBE)
- Dong X, Tyndall J (2024) The impact of recreational marijuana dispensaries on crime: evidence from a lottery experiment. Ann Reg Sci 72:1383â1414
[Article](https://link.springer.com/doi/10.1007/s00168-023-01246-x) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20impact%20of%20recreational%20marijuana%20dispensaries%20on%20crime%3A%20evidence%20from%20a%20lottery%20experiment&journal=Ann%20Reg%20Sci&doi=10.1007%2Fs00168-023-01246-x&volume=72&pages=1383-1414&publication_year=2024&author=Dong%2CX&author=Tyndall%2CJ)
- Fitzpatrick LG, Parmeter CF (2021) Data-driven estimation of treatment buffers in hedonic analysis: an examination of surface coal mines. Land Econ 97:528â547
[Article](https://doi.org/10.3368%2Fle.97.3.528) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Data-driven%20estimation%20of%20treatment%20buffers%20in%20hedonic%20analysis%3A%20an%20examination%20of%20surface%20coal%20mines&journal=Land%20Econ&doi=10.3368%2Fle.97.3.528&volume=97&pages=528-547&publication_year=2021&author=Fitzpatrick%2CLG&author=Parmeter%2CCF)
- Freisthler B, Gruenewald PJ, Wolf JP (2015) Examining the relationship between marijuana use, medical marijuana dispensaries, and abusive and neglectful parenting. Child Abuse & Neglect 48:170â178
[Article](https://doi.org/10.1016%2Fj.chiabu.2015.07.008) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Examining%20the%20relationship%20between%20marijuana%20use%2C%20medical%20marijuana%20dispensaries%2C%20and%20abusive%20and%20neglectful%20parenting&journal=Child%20Abuse%20%26%20Neglect&doi=10.1016%2Fj.chiabu.2015.07.008&volume=48&pages=170-178&publication_year=2015&author=Freisthler%2CB&author=Gruenewald%2CPJ&author=Wolf%2CJP)
- Huber A III, Newman R, LaFave D (2016) Cannabis control and crime: medicinal use, depenalization and the war on drugs. BE J Econ Anal Policy 16(4):20150167
[Article](https://doi.org/10.1515%2Fbejeap-2015-0167) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Cannabis%20control%20and%20crime%3A%20medicinal%20use%2C%20depenalization%20and%20the%20war%20on%20drugs&journal=BE%20J%20Econ%20Anal%20Policy&doi=10.1515%2Fbejeap-2015-0167&volume=16&issue=4&publication_year=2016&author=Huber%2CA&author=Newman%2CR&author=LaFave%2CD)
- Iannacchione B, Ward KC, Evans MK (2020) Perceptions of NIMBY syndrome among Colorado and Washington dispensary owners and managers. Justice Policy J 17(2)
- Kepple NJ, Freisthler B (2012) Exploring the ecological association between crime and medical marijuana dispensaries. J Stud Alcohol Drugs 73(4):523â530
[Article](https://doi.org/10.15288%2Fjsad.2012.73.523) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Exploring%20the%20ecological%20association%20between%20crime%20and%20medical%20marijuana%20dispensaries&journal=J%20Stud%20Alcohol%20Drugs&doi=10.15288%2Fjsad.2012.73.523&volume=73&issue=4&pages=523-530&publication_year=2012&author=Kepple%2CNJ&author=Freisthler%2CB)
- Lu R, Willits D, Stohr MK, Makin D, Snyder J, Lovrich N, Meize M, Stanton D, Wu G, Hemmens C (2021) The cannabis effect on crime: time-series analysis of crime in Colorado and Washington State. Justice Q 38(4):565â595
[Article](https://doi.org/10.1080%2F07418825.2019.1666903) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20cannabis%20effect%20on%20crime%3A%20time-series%20analysis%20of%20crime%20in%20Colorado%20and%20Washington%20State&journal=Justice%20Q&doi=10.1080%2F07418825.2019.1666903&volume=38&issue=4&pages=565-595&publication_year=2021&author=Lu%2CR&author=Willits%2CD&author=Stohr%2CMK&author=Makin%2CD&author=Snyder%2CJ&author=Lovrich%2CN&author=Meize%2CM&author=Stanton%2CD&author=Wu%2CG&author=Hemmens%2CC)
- Mair C, Freisthler B, Ponicki WR, Gaidus A (2015) The impacts of marijuana dispensary density and neighborhood ecology on marijuana abuse and dependence. Drug Alcohol Depend 154:111â116
[Article](https://doi.org/10.1016%2Fj.drugalcdep.2015.06.019) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20impacts%20of%20marijuana%20dispensary%20density%20and%20neighborhood%20ecology%20on%20marijuana%20abuse%20and%20dependence&journal=Drug%20Alcohol%20Depend&doi=10.1016%2Fj.drugalcdep.2015.06.019&volume=154&pages=111-116&publication_year=2015&author=Mair%2CC&author=Freisthler%2CB&author=Ponicki%2CWR&author=Gaidus%2CA)
- Morris RG, TenEyck M, Barnes JC, Kovandzic TV (2014) The effect of medical marijuana laws on crime: evidence from state panel data, 1990â2006. PLoS ONE 9(3):e92816
[Article](https://doi.org/10.1371%2Fjournal.pone.0092816) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20effect%20of%20medical%20marijuana%20laws%20on%20crime%3A%20evidence%20from%20state%20panel%20data%2C%201990%E2%80%932006&journal=PLoS%20ONE&doi=10.1371%2Fjournal.pone.0092816&volume=9&issue=3&publication_year=2014&author=Morris%2CRG&author=TenEyck%2CM&author=Barnes%2CJC&author=Kovandzic%2CTV)
- Morrison C, Gruenewald PJ, Freisthler B, Ponicki WR, Remer LG (2014) The economic geography of medical cannabis dispensaries in California. Int J Drug Policy 25(3):508â515
[Article](https://doi.org/10.1016%2Fj.drugpo.2013.12.009) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20economic%20geography%20of%20medical%20cannabis%20dispensaries%20in%20California&journal=Int%20J%20Drug%20Policy&doi=10.1016%2Fj.drugpo.2013.12.009&volume=25&issue=3&pages=508-515&publication_year=2014&author=Morrison%2CC&author=Gruenewald%2CPJ&author=Freisthler%2CB&author=Ponicki%2CWR&author=Remer%2CLG)
- Palmquist R D (1991) Hedonic methods. Measuring the demand for environmental quality
- Parmeter C F, Pope J C (2013) Quasi-experiments and hedonic property value methods. In: Handbook on Experimental Economics and the Environment, pp 3â66. Edward Elgar Publishing
- Rosen S (1974) Hedonic prices and implicit markets: product differentiation in pure competition. J Polit Econ 82(1):34â55
[Article](https://doi.org/10.1086%2F260169) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Hedonic%20prices%20and%20implicit%20markets%3A%20product%20differentiation%20in%20pure%20competition&journal=J%20Polit%20Econ&doi=10.1086%2F260169&volume=82&issue=1&pages=34-55&publication_year=1974&author=Rosen%2CS)
- Shi Y (2016) The availability of medical marijuana dispensary and adolescent marijuana use. Prev Med 91:1â7
[Article](https://doi.org/10.1016%2Fj.ypmed.2016.07.015) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20availability%20of%20medical%20marijuana%20dispensary%20and%20adolescent%20marijuana%20use&journal=Prev%20Med&doi=10.1016%2Fj.ypmed.2016.07.015&volume=91&pages=1-7&publication_year=2016&author=Shi%2CY)
- Taylor LO (2003) The hedonic method. A primer on nonmarket valuation. Springer, Cham, pp 331â393
[Chapter](https://link.springer.com/doi/10.1007/978-94-007-0826-6_10) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20hedonic%20method&doi=10.1007%2F978-94-007-0826-6_10&pages=331-393&publication_year=2003&author=Taylor%2CLO)
- Thomas D, Tian L (2021) Hits from the bong: the impact of recreational marijuana dispensaries on property values. Reg Sci Urban Econ 87:103655
[Article](https://doi.org/10.1016%2Fj.regsciurbeco.2021.103655) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Hits%20from%20the%20bong%3A%20the%20impact%20of%20recreational%20marijuana%20dispensaries%20on%20property%20values&journal=Reg%20Sci%20Urban%20Econ&doi=10.1016%2Fj.regsciurbeco.2021.103655&volume=87&publication_year=2021&author=Thomas%2CD&author=Tian%2CL)
- Tyndall J (2021) Getting high and low prices: marijuana dispensaries and home values. Real Estate Econ 49(4):1093â1119
[Article](https://doi.org/10.1111%2F1540-6229.12302) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Getting%20high%20and%20low%20prices%3A%20marijuana%20dispensaries%20and%20home%20values&journal=Real%20Estate%20Econ&doi=10.1111%2F1540-6229.12302&volume=49&issue=4&pages=1093-1119&publication_year=2021&author=Tyndall%2CJ)
- Whittemore AH, BenDor TK (2019) Reassessing NIMBY: the demographics, politics, and geography of opposition to high-density residential infill. J Urban Aff 41(4):423â442
[Article](https://doi.org/10.1080%2F07352166.2018.1484255) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Reassessing%20NIMBY%3A%20the%20demographics%2C%20politics%2C%20and%20geography%20of%20opposition%20to%20high-density%20residential%20infill&journal=J%20Urban%20Aff&doi=10.1080%2F07352166.2018.1484255&volume=41&issue=4&pages=423-442&publication_year=2019&author=Whittemore%2CAH&author=BenDor%2CTK)
- Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University
[Download references](https://citation-needed.springer.com/v2/references/10.1007/s00181-026-02894-6?format=refman&flavour=references)
## Author information
### Authors and Affiliations
1. Department of Agricultural Economics, Purdue University, West Lafayette, IN, 47907-2056, USA
Joshua Clark & Michael S. Delgado
Authors
1. Joshua Clark
[View author publications](https://link.springer.com/search?sortBy=newestFirst&contributor=Joshua%20Clark)
Search author on:[PubMed](https://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=search&term=Joshua%20Clark) [Google Scholar](https://scholar.google.co.uk/scholar?as_q=&num=10&btnG=Search+Scholar&as_epq=&as_oq=&as_eq=&as_occt=any&as_sauthors=%22Joshua%20Clark%22&as_publication=&as_ylo=&as_yhi=&as_allsubj=all&hl=en)
2. Michael S. Delgado
[View author publications](https://link.springer.com/search?sortBy=newestFirst&contributor=Michael%20S.%20Delgado)
Search author on:[PubMed](https://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=search&term=Michael%20S.%20Delgado) [Google Scholar](https://scholar.google.co.uk/scholar?as_q=&num=10&btnG=Search+Scholar&as_epq=&as_oq=&as_eq=&as_occt=any&as_sauthors=%22Michael%20S.%20Delgado%22&as_publication=&as_ylo=&as_yhi=&as_allsubj=all&hl=en)
### Corresponding author
Correspondence to [Michael S. Delgado](mailto:delgado2@purdue.edu).
## Additional information
### Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
## Supplementary Information
Below is the link to the electronic supplementary material.
### [Supplementary file 1 (pdf 708 KB) (download PDF )](https://static-content.springer.com/esm/art%3A10.1007%2Fs00181-026-02894-6/MediaObjects/181_2026_2894_MOESM1_ESM.pdf)
## Rights and permissions
**Open Access** This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit <http://creativecommons.org/licenses/by/4.0/>.
[Reprints and permissions](https://s100.copyright.com/AppDispatchServlet?title=Exploring%20the%20quantitative%20impact%20of%20medical%20marijuana%20dispensaries%20on%20residential%20sale%20prices%20in%20Oklahoma&author=Joshua%20Clark%20et%20al&contentID=10.1007%2Fs00181-026-02894-6©right=The%20Author%28s%29&publication=0377-7332&publicationDate=2026-03-21&publisherName=SpringerNature&orderBeanReset=true&oa=CC%20BY)
## About this article
[](https://crossmark.crossref.org/dialog/?doi=10.1007/s00181-026-02894-6)
### Cite this article
Clark, J., Delgado, M.S. Exploring the quantitative impact of medical marijuana dispensaries on residential sale prices in Oklahoma. *Empir Econ* **70**, 61 (2026). https://doi.org/10.1007/s00181-026-02894-6
[Download citation](https://citation-needed.springer.com/v2/references/10.1007/s00181-026-02894-6?format=refman&flavour=citation)
- Received: 20 December 2024
- Accepted: 28 January 2026
- Published: 21 March 2026
- Version of record: 21 March 2026
- DOI: https://doi.org/10.1007/s00181-026-02894-6
### Share this article
Anyone you share the following link with will be able to read this content:
Get shareable link
Sorry, a shareable link is not currently available for this article.
Copy shareable link to clipboard
Provided by the Springer Nature SharedIt content-sharing initiative
### Keywords
- [Medical marijuana](https://link.springer.com/search?query=Medical%20marijuana&facet-discipline="Economics")
- [Hedonic](https://link.springer.com/search?query=Hedonic&facet-discipline="Economics")
- [Public policy](https://link.springer.com/search?query=Public%20policy&facet-discipline="Economics")
Advertisement
## Search
## Navigation
- [Find a journal](https://link.springer.com/journals/)
- [Publish with us](https://www.springernature.com/gp/authors)
- [Track your research](https://link.springernature.com/home/)
### Discover content
- [Journals A-Z](https://link.springer.com/journals/a/1)
- [Books A-Z](https://link.springer.com/books/a/1)
### Publish with us
- [Journal finder](https://link.springer.com/journals)
- [Publish your research](https://www.springernature.com/gp/authors)
- [Language editing](https://authorservices.springernature.com/go/sn/?utm_source=SNLinkfooter&utm_medium=Web&utm_campaign=SNReferral)
- [Open access publishing](https://www.springernature.com/gp/open-science/about/the-fundamentals-of-open-access-and-open-research)
### Products and services
- [Our products](https://www.springernature.com/gp/products)
- [Librarians](https://www.springernature.com/gp/librarians)
- [Societies](https://www.springernature.com/gp/societies)
- [Partners and advertisers](https://www.springernature.com/gp/partners)
### Our brands
- [Springer](https://link.springer.com/brands/springer)
- [Nature Portfolio](https://www.nature.com/)
- [BMC](https://link.springer.com/brands/bmc)
- [Palgrave Macmillan](https://link.springer.com/brands/palgrave)
- [Apress](https://link.springer.com/brands/apress)
- [Discover](https://link.springer.com/brands/discover)
- Your privacy choices/Manage cookies
- [Your US state privacy rights](https://www.springernature.com/gp/legal/ccpa)
- [Accessibility statement](https://link.springer.com/accessibility)
- [Terms and conditions](https://link.springer.com/termsandconditions)
- [Privacy policy](https://link.springer.com/privacystatement)
- [Help and support](https://support.springernature.com/en/support/home)
- [Legal notice](https://link.springer.com/legal-notice)
- [Cancel contracts here](https://support.springernature.com/en/support/solutions/articles/6000255911-subscription-cancellations)
Not affiliated
[](https://www.springernature.com/)
© 2026 Springer Nature |
| Readable Markdown | ## 1 Introduction
Oklahoma State Question 788 (Initiative Petition No. 412), or SQ 788, was filed with the Oklahoma Secretary of State on April 11, 2016, and proposed the legalization of the purchase and consumption of medical marijuana for adults with a state issued medical marijuana license. SQ 788 established that a regulatory officeâthe Oklahoma Medical Marijuana Authority (OMMA)âbe opened under the Oklahoma State Department of Health to regulate and receive applications for dispensaries, consumption, growing, and packaging licenses. SQ 788 passed on June 26, 2018, with a 56.86 percent majority vote, and took effect on August 25, 2018, making Oklahoma the 30th state in the USA to legalize medical marijuana. Since August 2018 when SQ 788 went into effect, Oklahoma has seen explosive growth in the number of medical marijuana dispensaries within the state. As of February 27, 2024, the State of Oklahoma has 2227 uniquely licensed medical marijuana dispensaries listed on the OMMA website, making Oklahoma the leading state for the most dispensaries of any kind in the USA. Compared to the next three states in orderâCalifornia, Colorado, and Michigan that have 1244, 1023, and 994 dispensaries, respectively, as of January 1, 2024 (dispenseapp.com)âit is clear how explosive the demand for medical marijuana dispensary licenses has been in the State of Oklahoma since legalization. What has been the impact of this rapid proliferation of medical marijuana dispensaries on the residents within the state? We seek to provide an answer to this question by investigating the local property effects of medical marijuana dispensaries in Oklahoma via the lens of a hedonic pricing model.
Clearly, this rapid proliferation of medical marijuana dispensaries in Oklahoma in the few years since legalization makes Oklahoma an ideal research setting. Beyond just this rapid proliferation, Oklahoma is somewhat of a surprise given that it is generally considered to be a strongly conservative state with much of its population identifying with the Republican Party. Voters in Oklahoma have consistently supported the Republican presidential candidate in every election since 1968, and in the 2020 US presidential election, barely two years after SQ 788 was enacted into state law, each of the 77 counties in Oklahoma voted in majority for the Republican presidential candidate. Since 2004, political party affiliation has been shown to be a statistically significant determinant in measuring support for the legalization of medical marijuana, with those who identify as Republicans expressing significantly less support than those who identify as Democrats or Independents (Denham [2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR15 "Denham BE (2019) Attitudes toward legalization of marijuana in the United States, 1986â2016: changes in determinants of public opinion. Int J Drug Policy 71:78â90")).[Footnote 1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn1) We will also show, in our descriptive statistics, that the medical marijuana dispensaries located across Oklahoma are not clustered purely in downtown areas; instead, the dispensaries are broadly located throughout the cities in Oklahoma as well as in both heavily urban and more rural areas. Therefore, the combination of the rapid increase in the quantity of medical marijuana dispensaries in Oklahoma, the strongly Republican electorate, and the broad scattering of dispensaries across the state makes Oklahoma an ideal context in which to explore the property value effects of legalized medical marijuana.
To date, there has been relatively little research done on the effects of (legalized) marijuana dispensaries on residential property prices, and so far, the empirical evidence is mixed. Cheng et al. ([2018](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR11 "Cheng C, Mayer WJ, Mayer Y (2018) The effect of legalizing retail marijuana on housing values: evidence from Colorado. Econ Inq 56(3):1585â1601")); Burkhardt and Flyr ([2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR9 "Burkhardt J, Flyr M (2019) The effect of marijuana dispensary openings on housing prices. Contemp Econ Policy 37(3):462â475")) and Conklin et al. ([2020](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR13 "Conklin J, Drop M, Li H (2020) Contact high: the external effects of retail marijuana establishments on house prices. Real Estate Econ 48(1):135â173")) all find that legalized marijuana leads to a measurable increase in property values in Colorado, while (Boswell [2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR5 "Boswell K (2021) Essays on Cannacis legalization in California. PhD Thesis, University of Miami")) finds a residential property value increase in California. Cheng et al. ([2018](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR11 "Cheng C, Mayer WJ, Mayer Y (2018) The effect of legalizing retail marijuana on housing values: evidence from Colorado. Econ Inq 56(3):1585â1601")) find a 6 percent increase of housing values, and Burkhardt and Flyr ([2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR9 "Burkhardt J, Flyr M (2019) The effect of marijuana dispensary openings on housing prices. Contemp Econ Policy 37(3):462â475")) find that a marginal dispensary located within a half-mile of a new home increases home prices by about 7.7 percent. Conklin et al. ([2020](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR13 "Conklin J, Drop M, Li H (2020) Contact high: the external effects of retail marijuana establishments on house prices. Real Estate Econ 48(1):135â173")) study the conversion of medical marijuana dispensaries to recreational marijuana dispensaries, and find that single-family properties located within 0.1 miles of a converted dispensary realize an 8.4 percent increase in property value. They suggest that lower crime rates associated with dispensaries and higher housing demand associated with dispensary business activity may drive the boost in property values. Tyndall ([2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR32 "Tyndall J (2021) Getting high and low prices: marijuana dispensaries and home values. Real Estate Econ 49(4):1093â1119")), on the other hand, provides evidence to the contrary. Using a sample of repeat home sales from Vancouver, BC, he finds a decrease in property values ranging from 3.7 to 4.9 percent (depending on the regression specification) for properties located within 100 m of a dispensary.
A related literature focuses on the spatial distribution and economic geography of dispensary locations to understand the extent to which dispensaries are viewed as undesirable. Boggess et al. ([2014](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR4 "Boggess LN, PĂ©rez DM, Cope K, Root C, Stretesky PB (2014) Do medical marijuana centers behave like locally undesirable land uses? Implications for the geography of health and environmental justice. Urban Geogr 35(3):315â336")) find that medical marijuana dispensaries in Denver do not correlate with neighborhood racial composition or income and conclude that medical marijuana dispensaries are not locally undesirable.[Footnote 2](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn2) They do find that dispensaries are more likely to be located in areas with higher crime rates and greater density of retail employment. Somewhat differently, Morrison et al. ([2014](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR25 "Morrison C, Gruenewald PJ, Freisthler B, Ponicki WR, Remer LG (2014) The economic geography of medical cannabis dispensaries in California. Int J Drug Policy 25(3):508â515")) find that medical marijuana dispensaries in California are more likely to be located in lower income areas, as well as areas with a greater number of alcohol outlets, and Brinkman and Mok-Lamme ([2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR6 "Brinkman J, Mok-Lamme D (2019) Not in my backyard? Not so fast. The effect of marijuana legalization on neighborhood crime. Reg Sci Urban Econ 78:103460")) show that dispensaries are more likely to be located in Hispanic areas. Most related to our current work is Zimmer et al. ([2022](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR34 "Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University")) and Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")), both of which study dispensary siting in Oklahoma. Zimmer et al. ([2022](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR34 "Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University")) find that medical marijuana dispensaries in Oklahoma are more likely to locate in areas with higher rates of crime, a greater number of uninsured individuals or individuals on disability, and in closer proximity to the Texas border. Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")) find that Oklahoma dispensaries are more likely to be located in areas of lower socioeconomic status (racial minority and low-income) and in areas with more limited access to pharmacies.
Finally, some of the ânot in my backyardâ (or, NIMBY) literature is relevant to our work. Most immediately related is Iannacchione et al. ([2020](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR20 "Iannacchione B, Ward KC, Evans MK (2020) Perceptions of NIMBY syndrome among Colorado and Washington dispensary owners and managers. Justice Policy J 17(2)")) who conducted a survey of dispensary operators (owners or managers) in both Colorado and Washington to assess the extent to which local residents may have been opposed to the opening of the dispensary; that is, the extent of NIMBY effects surrounding legalized medical or recreational marijuana dispensaries. They find that the NIMBY effect of legalized marijuana dispensaries is generally non-existent, and the cases in which dispensary operators faced pushback was short-lived and typically driven by a lack of education about marijuana dispensaries and/or uncertainty (that soon subsided). On the political affiliation side of NIMBY research, Whittemore and BenDor ([2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR33 "Whittemore AH, BenDor TK (2019) Reassessing NIMBY: the demographics, politics, and geography of opposition to high-density residential infill. J Urban Aff 41(4):423â442")) found that although both liberals and conservatives were likely to be concerned with changes in property values, conservatives were significantly more likely to show concern (at least in that context of new residents within the context of high-density residential infill). Extrapolating to our context, we might find evidence of harmful effects of dispensaries on nearby property values given the more conservative population.[Footnote 3](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn3)
We consider models with data pooled across counties, as well as county-specific regressions; we consider both linear and nonlinear regression specifications. While we find some heterogeneity in the estimated effects of dispensaries on local home prices, a broad trend that emerges is that many residents prefer not to live in immediate proximity to a dispensary but prefer to maintain access to a dispensary within a moderate distance from the home (about 1â2 kilometers away). There are some instances in which the dispensaries do not have a significant effect on property values, and some cases wherein dispensary proximity raises housing values. These different results are discussed and in the context of different counties and relative robustness of the regression results (e.g., sample size).
Our work is important in at least two ways. First, we contribute to the measurement of the property impacts of legalized marijuana activity in Oklahoma. These estimates are important for citizens and policymakers in Oklahoma interested in understanding some of the effects of SQ 788 on local neighborhoods, particularly since we find evidence that many homeowners prefer not to live in close proximity to a dispensary despite the bill passing in Oklahoma with broad public support. Second, our estimates are also relevant for citizens and policymakers in other states that have already legalized marijuana, as well as in states in which legislation to legalize marijuana is (or may soon be) proposed. In either case, understanding (some of) the effects of legalized marijuana is of critical policy-relevance, and as motivated, Oklahoma is uniquely situated to provide insights that are relevant in the broader context. Finally, readers should recognize that our work focuses on localized housing market impacts of legalized medical marijuana and is not an overall welfare measurement.
## 2 Empirical approach
### 2\.1 Hedonic price model
Our regression model is a standard hedonic pricing model (e.g., Rosen [1974](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR28 "Rosen S (1974) Hedonic prices and implicit markets: product differentiation in pure competition. J Polit Econ 82(1):34â55"); Palmquist [1991](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR26 "Palmquist R D (1991) Hedonic methods. Measuring the demand for environmental quality"); Taylor [2003](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR30 "Taylor LO (2003) The hedonic method. A primer on nonmarket valuation. Springer, Cham, pp 331â393"); Parmeter and Pope [2013](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR27 "Parmeter C F, Pope J C (2013) Quasi-experiments and hedonic property value methods. In: Handbook on Experimental Economics and the Environment, pp 3â66. Edward Elgar Publishing")) defined as
\$\$\\begin{aligned} \\ln Price\_{it} = \\beta \_0 + Disp\_{i}\\beta \_1 + X\_{i} \\delta + v\_t + \\varepsilon \_{it} \\hspace{2em} i=1,2,\\dots ,n; \\hspace{1em} t=1,2,\\dots ,T \\end{aligned}\$\$
(1)
where \\(\\ln Price\_{it}\\) is the natural log of the price of property *i* at the time of sale, *t*, \\(Disp\_{i}\\) is a vector of variables measuring the proximity of the property to the dispensaries with corresponding parameter vector \\(\\beta \_1\\), \\(X\_{i}\\) is a *K*\-dimensioned vector of property- and neighborhood-specific control variables with \\(\\delta \\) as a corresponding vector of parameters, \\(v\_t\\) is a year-specific effect common to all properties, and \\(\\varepsilon \_{it}\\) is the mean-zero error term.[Footnote 4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn4) Note that we do not deflate property prices and instead prefer to control for differences in price levels across years through \\(v\_t\\). We are primarily interested in \\(\\beta \_1\\) as this parameter vector measures the impact of the dispensaries on the value of nearby properties.
### 2\.2 Proximity to dispensaries
The proximity vector, \\(Disp\_{i}\\), contains several different measures of property proximity to the dispensaries to fully explore the effect of the dispensaries on property values. Our empirical strategy follows, for instance, Delgado et al. ([2016](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR14 "Delgado MS, Guilfoos T, Boslett A (2016) The cost of unconventional gas extraction: a hedonic analysis. Resour Energy Econ 46:1â22")) in that we consider both continuous and discrete measures of proximity, both individually and separately.
The first measure of proximity that we use is the continuous distance in meters from each property to the nearest dispensary, defined as *minDist*. Recent research focusing on hedonic regression methodology has proposed data-driven techniques for determining the optimal distance by which to model the effect of an amenity on housing prices. Specifically, Fitzpatrick and Parmeter ([2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR17 "Fitzpatrick LG, Parmeter CF (2021) Data-driven estimation of treatment buffers in hedonic analysis: an examination of surface coal mines. Land Econ 97:528â547")) propose a cross-validation procedure whereby the optimal distance is selected via a numerical optimization procedure; we apply their method in selecting the optimal distance of dispensary effect based on the *minDist* variable, and then use this optimal distance as a way of filtering our sample into the region of effect (within the threshold) and the relevant comparison area (within two times the threshold). We explore regression specifications with *minDist* measured both linearly and via a quadratic specification, with the latter allowing us to detect nonlinearities in the relationship between proximity to the nearest dispensary and the property price. Tyndall ([2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR32 "Tyndall J (2021) Getting high and low prices: marijuana dispensaries and home values. Real Estate Econ 49(4):1093â1119")), for example, finds localized effects of dispensary proximity on property values, and a nonlinear (quadratic) distance specification will help us capture such localized effects more readily than a linear specification. Using a continuous distance measure also allows us to recover the shape of the distance-impact gradient to determine over exactly what range of distance the dispensaries affect property values (if at all), and how rapidly the effect dissipates (if at all).
An alternative way to measure the effect of the dispensaries on property values is to measure the intensity of the dispensary proximity to each property using pre-specified distance bands. We select distance bands of 500 m, 1000 m, and 2000 m and count the number of dispensaries that are within each distance band from each property as our measures of dispensary intensity. We denote these variables as *Count*500, *Count*1000, and *Count*2000 and consider regression specifications with these variables included together as well as separately.[Footnote 5](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn5) It is important to understand the interpretation of the regression coefficients when these variables are included individually or jointly: when these coefficients are included individually in the regression, the coefficient is simply the effect on \\(\\ln Price\\) of an additional dispensary being added within the distance band. For instance, if we include *Count*1000 in the regression, then the coefficient reflects the effect of adding another dispensary within 1000*m* (or 1*km*) of the property on the \\(\\ln Price\\) of the property (control variables held constant). On the other hand, if we include all three count variables in the model jointly, then the coefficient on *Count*1000 reflects the effect of adding another dispensary within 1000*m* of the property on the \\(\\ln Price\\) of the property *holding constant* the number of dispensaries within 500*m* and 2000*m* of the property (all other control variables being held constant as well). In other words, the coefficient on *Count*1000 now captures the effect on \\(\\ln Price\\) of adding another dispensary in the \\(\>500m\\) but \\(\\le 1000m\\) distance range.
Finally, it is worth clarifying that a regression specification including both the continuous and discrete proximity measures together via \\(Disp=(minDist,Count500, Count1000,Count2000)\\) bears the following interpretation. The coefficient on *minDist* is the effect on \\(\\ln Price\\) of moving the dispensary one meter away from the property holding constant the number of dispensaries located within 500*m*, 1000*m*, and 2000*m* from the property (in addition to all other variables in the model being held constant). That is, we interpret the distance effect with intensity held constant. We interpret the coefficients on the count variables, say *Count*1000, as the effect of adding one more dispensary at a distance \\(\>500m\\) but \\(\\le 1000m\\) while holding the distance to the nearest dispensary constant. That is, an increase in intensity within the specified distance band while holding the distance to the nearest dispensary constant.
By considering the continuous *minDist* measure of proximity and the three count measures of intensity both individually and jointly, we are able to fully explore the nature of dispensary proximity on nearby property values. It is worth noting that further exploring these regressions separately by county allows us to explore the extent to which these relationships are heterogeneous in different parts of Oklahoma.
### 2\.3 Control variables and fixed effects
Of course, the extent to which our estimates of \\(\\beta \_1\\) are reliable depends on our ability to adequately control for other factors that affect \\(\\ln Price\\) and also correlate with the effect of the dispensaries on property values. In our main regression specifications, we deploy control variables and fixed effects to account for these other factors. Being a hedonic price model, we naturally control first for housing characteristics. In our primary dataset, we control for the number of bedrooms, the number of bathrooms, the age of the property, and the square footage of the property. With regard to the age of the property and square footage, we find the best fit via a quadratic specification. It is important to note that these particular variables are common across the four counties we consider jointly in our primary specifications; we explore additional covariate controls as the data are available in auxiliary regressions that are county-specific.[Footnote 6](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn6)
Beyond these house-specific characteristics, we include the distance to the central business district in kilometers, as well as the average sales price of other properties within a 500 m radius of each property sold *prior* to SQ 788. This variable is particularly important for identification, as it allows us to control for neighborhood-level differences that might otherwise be correlated with the location of dispensaries. We noted earlier that Zimmer et al. ([2022](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR34 "Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University")) find that the location of medical marijuana dispensaries in Oklahoma is correlated with distance to Texas (where both medical and recreational marijuana is illegal), and rates of crime, the population that is uninsured, or has a disability, and Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")) find that Oklahoma dispensary locations correlate with fewer hospitals, being uninsured, and having a lower socioeconomic status (defined by either race or income). In Denver, medical marijuana dispensary locations correlate with higher crime rates and having a higher rate of retail sale establishments (Boggess et al. [2014](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR4 "Boggess LN, PĂ©rez DM, Cope K, Root C, Stretesky PB (2014) Do medical marijuana centers behave like locally undesirable land uses? Implications for the geography of health and environmental justice. Urban Geogr 35(3):315â336")). Regardless of the extent to which any of these particular factors causally affect dispensary location, it is clear that medical marijuana dispensariesâin Oklahoma and elsewhereâcorrelate with neighborhood factors. These factors are surely reflected by differences in neighborhood real estate prices, and so by controlling for average neighborhood property prices prior to SQ 788, we are able to adjust for differences in neighborhood effects that might otherwise correlate with dispensary locations.
In terms of the fixed effects, we control for the year the property was sold to account for macro-effects that are common to all properties sold within each year, as well as county-specific dummies to capture differences across counties.
### 2\.4 Auxiliary regression specifications
We consider auxiliary regression specifications that serve as robustness checks. Specifically, we separate the data by county and consider county-specific regressions. This robustness check does two things. First, this approach allows us to estimate county-specific parameters for all of the coefficients in the model, to determine the extent to which the estimated relationships vary by county. Second, the data we obtained from the county assessor offices contain different sets of control variables, and by considering each county individually, we are able to add additional control variables that are available for each particular county. While the county-specific model specifications are not dramatically different, it is worth exploring the importance of these different control variables where possible. More specific details will be provided in subsequent sections where we explore these data and regressions.[Footnote 7](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn7)
## 3 Data
### 3\.1 Housing data
Residential property sales data were obtained from six county assessor offices in the State of Oklahomaâas noted in the introduction, Comanche, Oklahoma, Payne, Tulsa, Wagoner and Washington counties. The data include all single-family property sales from the beginning of 2016 to the end of 2023, providing information on the sales price, geographic location, and property attributes. In terms of the property attributes, the data includes typical housing attribute variables that are commonly found in the hedonic housing literature, including the age of the property measured in years, living area in square feet, the number of bedrooms, and the number of bathrooms. There is slight variation across county-specific datasets in terms of additional housing-level variables. Specifically, the data from Payne and Tulsa counties lack key variables, and so we consider the property sales from these counties in auxiliary regressions but not as part of our main benchmark models. Specifically, our data for Payne County lack data on the age of the property and property square footage, and our data for Tulsa County lacks data on the number of bedrooms of each property.
### 3\.2 Dispensary data
Medical marijuana dispensary data are made publicly available to view and download online by the Oklahoma Medical Marijuana Authority (OMMA). Publicly available data as of 2024 include the name of the licensed dispensary, the license identification number, as well as the county and city the dispensary operates in within the state. These data are used by Zimmer et al. ([2022](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR34 "Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University")) and Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")), with Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")) noting that they conducted online searches using the dispensary names provided by the OMMA to further identify exact geographical locations and current operational status. In our case, we use publicly available data provided by the Oklahoma Watch organization website (Brown and Howe [2019](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR8 "Brown T, Howe J (2019) Number of medical marijuana outlets soars. How many are in your area?")) that is maintained by Trevor Brown and Jesse Howe; their data were obtained through the OMMA and mapped using Googleâs My Maps service.[Footnote 8](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn8) The data include 1,463 unique medical marijuana dispensary locations from across the entire state.
It is important to note that with these data, we do not know the date at which each dispensary opened. To link the dispensaries to the property sales, in both the benchmark and auxiliary datasets we include only property sales after July 2020 because the dispensary data we use comes from a cross-section of dispensaries that were in operation as of July 2020. By focusing on these particular sales transactions, we are able to ensure that our linkage of sales transactions to dispensary locations is concurrent in time.
### 3\.3 Geolocation and mapping
For each county, each property comes with a geolocating identifier. Counties that provided a street address for each home were geocoded into latitudinal and longitudinal coordinates using Esri ArcGIS Pro 3.1. We use the geographical location of the property to compute the distance to the nearest central business district (CBD): we use the âgeosphereâ package in R 4.2.2 to compute in kilometers the distance between the property and the closest CBD, which, in our data, is either downtown Oklahoma City or downtown Tulsa. We also use these locators to compute the distance in meters to the closest dispensary, as well as the number of dispensaries in binned radii around the property to provide a count of how many exist within 500, 1000, and 2000 m. We also use these geographical locators to construct the historical neighborhood value variable: We construct a 500 m radius around each property and then compute the average sale price of all properties sold within the radius between January 2016 and July 2018. These years correspond to our sales data prior to the opening of any dispensaries, thereby providing a market value-based measure of historical neighborhood quality. We also report the average number of properties within this 500 m radius (the âdensityâ variable) for readers to understand the rough size of these local neighborhoods.
### 3\.4 Descriptive statistics
Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1) provides a descriptive summary of the 4-county benchmark data (the pooled sample) as well as the data broken down by county. More detailed descriptive statistics are in the appendix.
**Table 1 Descriptive statistics for the pooled sample of four counties and separately by county**
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/1)
Across the pooled sample, the average sale price is around \$253,000, with average prices being slightly higher in Oklahoma and Tulsa counties, and lower in Comanche, Wagoner and Washington counties. These properties are, on average, three bedroom and two bathroom homes, that are around 40 years old and have just under 2000 square feet of living space. The average age of the home, in particular, varies quite a bit across counties. From the counties with available data, these are typically one story homes. We measure the central business district (CBD) as either Oklahoma City or Tulsa, and from Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1) we see that the average distance to CBD is just under 40 kilometers.
The average historical neighborhood value is around \$203,000, which is roughly \$50,000 lower than the average sale price, and this computation comes from a local neighborhood density of, on average, 24.8 home sales. These numbers do not account for increases in property values through the intervening time period; this is an unconditional comparison of average values. To dig deeper into the relationship between historical neighborhood values and current property valuesâand thereby to provide some empirical justification for neighborhood to serve as an important control variable in our analysis and thereby aid in our identification strategyâwe turn to Fig. [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig1) and Table [2](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab2). In Fig. [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig1), we show a scatterplot correlation between historical neighborhood property values and current property prices, and Table [2](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab2) reports univariate regression coefficients (i.e., correlations) between these variables. We can see a strong, positive correlation between neighborhood prices pre-dispensaries and property values post-dispensaries; from the table, we can see that this correlation is statistically significant with a coefficient of about 0.7.
**Fig. 1**
[](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/1)
Scatterplot of historical neighborhood values and current property prices
[Full size image](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/1)
**Table 2 Univariate correlation between neighborhood values and \\(\\ln (Price/100)\\)**
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/2)
In the spirit of Boggess et al. ([2014](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR4 "Boggess LN, PĂ©rez DM, Cope K, Root C, Stretesky PB (2014) Do medical marijuana centers behave like locally undesirable land uses? Implications for the geography of health and environmental justice. Urban Geogr 35(3):315â336")); Zimmer et al. ([2022](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR34 "Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University")) and Cohn et al. ([2023](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR12 "Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113")), we consider the correlation between *Count*500 and historical neighborhood value. Figure [2](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig2) provides a scatterplot and estimated trend line, and Table [3](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab3) provides the univariate regression estimate. There is a clear downward trend, whereby as the number of dispensaries within 500 m increases, historical neighborhood value decreases; the regression table indicates that this correlation is statistically significant. Further, our estimated correlation is linear, though the figure clearly suggests a nonlinear pattern whereby the largest downward effect of dispensary proximity occurs with the siting of the first one or two dispensaries within 500 m, with a dramatically reduced effect with marginal increases in dispensary intensity. Of course, this is merely a correlationâthus, a descriptive statistic that motivates our use of neighborhood as a control variableâand should not be taken as evidence that dispensaries are located in areas characterized in any particular way (e.g., by race, income, or other status).
Turning to the dispensary-related variables, we see that, on average, the minimum distance to the nearest dispensary is about 1.4 kilometers (Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1)). These properties have, on average, 0.34 dispensaries within a 500 m radius of the property; 1.5 dispensaries within 1 km of the property; and 5.7 dispensaries within 2 kilometers of the property. It is worth noting that Washington, Oklahoma and Tulsa counties have properties that are in closest proximity, on average, to dispensaries and with the highest average density of proximate dispensaries.
**Fig. 2**
[](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/2)
Scatterplot between *Count*500 and neighborhood value before legalization
[Full size image](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/2)
**Table 3 Univariate correlation between *Count*500 and neighborhood**
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/3)
### 3\.5 Mapping the data
**Fig. 3**
[](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/3)
Property sales and dispensary locations across Oklahoma. Red pins indicate dispensary locations, green pins indicate property locations for our four-county benchmark sample, and purple pins denote property locations from our auxiliary samples (Tulsa and Payne counties)
[Full size image](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/3)
We have described the rapid extent to which medical marijuana dispensaries have opened in Oklahoma since 2018, and the following figures help to illustrate this fact. In Fig. [3](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig3) we map all the properties in our dataset against the locations of all the dispensaries across the State of Oklahoma. Specifically, green circles represent the location of an individual residential sale from the benchmark dataset, and the red pins represent the locations of the dispensaries that were operating as of July 2020. It is clear from the map that the dispensaries are located all across Oklahoma, covering both urban and rural areas; given the short time span since the passage of the law, the number of operating dispensaries has quickly risen.
To gain deeper insight into the locations of the property sales and dispensaries, we plot the Oklahoma County data in Fig. [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig4). We can clearly see the dispensaries located broadly across Oklahoma County (primarily this is Oklahoma City), located not just in the city center but in suburban areas as well.
**Fig. 4**
[](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/4)
Property sales and dispensary locations in Oklahoma County. Red pins indicate dispensary locations, and green pins indicate property locations
[Full size image](https://link.springer.com/article/10.1007/s00181-026-02894-6/figures/4)
## 4 Results
### 4\.1 Benchmark regression analysis
Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4) displays the results from our benchmark regression model that includes both *minDist* and all three *Count* variables (*Count*500, *Count*1000, *Count*2000\); a model that includes *minDist* but excludes the three *Count* variables; a model that is quadratic in *minDist* and excludes the *Count* variables; and a model that is quadratic in *minDist*, interacts *minDist* with historical neighborhood values, and excludes the three *Count* variables. The properties included in these regressions are all within two times the optimal cross-validated distance threshold, which for the pooled sample is 3110 m. These models allow us to first set a benchmark for the housing market impacts of dispensary activities (Column 1) and then gain a deeper understanding into the relationship between *minDist* and property values (Columns 2â4). The data used in these regressions are the pooled sample of four counties for which we have complete and overlapping control variables (i.e., Comanche, Wagoner, Washington and Oklahoma counties). In all four models, we include housing characteristic variables that includes quadratic specifications for both age and square footage, as well as year of sale and county indicators.
**Table 4 Benchmark model and sub-models to further assess *minDist***
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/4)
Looking first at the benchmark model shown in Column 1 in Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4), we see that *minDist*, *Count*1000 and *Count*2000 are statistically significant. *Count*500 is not significant. Further, *minDist* and *Count*2000 have a positive coefficient, whereas *Count*1000 has a negative coefficient. Interpreting first the *minDist* coefficient: the model is a semi-log model, which means that a one-meter increase in the distance from the nearest dispensary leads to a \\(\\hat{\\beta }\\times 100\\) percent increase in the sale price of the property (in the regression, prices are in hundreds of dollars). Our estimate thus implies a 0.002197 percentage change in price in hundreds with every one-meter increase in distance. Recalling from Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1) that the average property price is 253,060.30, this estimated marginal effect implies that a one-meter increase in distance to nearest dispensary leads to a 5.56 dollar increase in sale price after converting into dollars. Again, bear in mind that this is controlling for the *Count* variables that capture dispensary intensity as well as the other control variables. If we extrapolate to a 100-m increase in distance to nearest dispensary, this is a property price increase of about 555.97 dollars. These estimates indicate that dispensaries are on net undesirable, though this is not a substantial price increase.
We turn next to the *Count* variables and their estimated impacts on property prices. Recall that since all three variables are included in the regression, the interpretation of each *Count* variable coefficient is incremental: adding one more dispensary within a 500 m radius of a property, holding constant the number of dispensaries within the 500â2000 m range, leads to a \\(\\hat{\\beta }\_{Count500}\\) change in a propertyâs price in hundreds of dollars or a \\(\\hat{\\beta }\_{Count500} \\times 100\\) percentage change. Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4) shows that *Count*500 is insignificant, while *Count*1000 is significantly negative and *Count*2000 is significantly positiveâholding *minDist* and other controls constant. Specifically, our estimates indicate that, all else constant, adding an additional dispensary within the 500â1000 m distance from a property leads to a decrease in the average property price by about 0.429 percent; from the average home price, this is a decrease of about 1085.63 dollars. At the same time, the coefficient on *Count*2000 indicates that adding one more dispensary in the 1000â2000 m range, holding *minDist*, the other *Count* variables, and controls constant, results in a 0.427 percent increase in property values (or about 1080.57 dollars at the average). Taken at face value, these estimates indicate that homeowners prefer to have a marginal dispensary located farther away from the home.
It is worth discussing the effects and significance of the control variables on property prices before digging deeper into the dispensary exposure variables and their relationship to property prices. Still looking at the benchmark model in Column 1 of Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4), we see statistically significant quadratic relationships between the age and square footage of the property and property values. In particular, property age has an inverse-U-shaped relationship with property prices, with a minimum property value occurring at a property age of 70 years. The average age of the home in the pooled sample is 40 years (Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1)). Square footage has a U-shaped relationship with property values, with the maximum occurring at 15,046 square feet. From Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1), the average square footage in the pooled benchmark sample is 1904.99, and the detailed descriptive statistics tables in the appendix indicate that this turning point is greater than the maximum square footage in our data; in effect, the square footage effect on property prices is monotonically increasing but at a decreasing rate. We see that the number of bedrooms is significantly negative; given that we control for square footage, the negative coefficient on bedrooms means that home buyers prefer fewer but larger rooms for a home of a given size. The number of bathrooms and neighborhood home density are significantly positive, while distance to the nearest CBD is significantly negative. The historical neighborhood value is significantly positive, which is in line with the correlations shown in Fig. [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fig1) and Table [2](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab2), and indicates that historical average neighborhood values are positively and significantly correlated with current sales prices. This estimate, therefore, reinforces our intuition to include this variable as an important control for neighborhood quality that in part explains current property prices but might also correlate with dispensary activity (distance and exposure). Finally, we see some significance of the year of sale indicators relative to the baseline year of 2023, and significance of the county indicators relative to the Comanche County baseline. We note that the \\(R^2\\) is 0.640, reflecting a reasonable fit of the regression.
### 4\.2 Distance to the nearest dispensary
Moving beyond the benchmark regression, in Columns 2â4 of Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4) we explore more deeply the relationship between *minDist* and property values. In Column 2, we remove the *Count* variables and explore the linear effect of *minDist* on the natural log of property values: We find the results remain qualitatively unchanged from the benchmark regression (Column 1), with a significantly positive effect of *minDist* on property values. This estimate is fairly smaller in magnitude compared to the estimate in Column 1, this time implying an effect of 3.71 dollars per meter, or about 371 dollars for a 100 m distance.
In Column 3, we consider a quadratic specification and find that only the quadratic term is statistically significant, and in Column 4 we consider both the quadratic *minDist* specification including an interaction with the historical neighborhood index. In this latter model, we find that all coefficients are significant. Exploring these estimates further, using both the average property value and average neighborhood value index (203,309.40 from Table [1](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab1)), we find that on average the *minDist* marginal effect on property prices is about 224 dollars per 100 m. In looking across the models in Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4), we see that sharpening our estimates (i.e., Column 4) results in a smaller estimated average effect of *minDist* on property prices. Yet, from the perspective of these models, dispensaries are viewed by homeowners as a negative amenity.
### 4\.3 Intensity of dispensary exposure
**Table 5 Models to further assess significance of count variables**
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/5)
Next, we look more deeply into the relationship between the *Count* variables and property values; regression results are found in Table [5](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab5). The first model is a version of the benchmark regression from Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4) except that we omit *minDist*. Models 2â4 consider the three *Count* variables separately, one at a time. The dependent variable remains \\(\\ln (Price/100)\\).
The results in the first column are very similar to the results shown in the benchmark regression (Column 1 of Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4)). That is, removing *minDist* does not change the results from the benchmark model. This result mirrors Column 2 of Table [4](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab4) in that removing the *Count* variables did not change the effect of *minDist* on property price. That is, we continue to find an insignificant coefficient on *Count*500, a significantly negative coefficient on *Count*1000, and a significantly positive coefficient on *Count*2000 all with magnitudes very similar to those discussed in the previous subsection.
It is interesting that when considering the *Count* variables separately, as shown in Models 2â4, only *Count*2000 remains significant (and still has a positive coefficient). It is important to remember that in these three regressions, the interpretations of the *Count* variables coefficients are different from the joint model. Since the other bands are not held constant, the coefficient on *Count*1000, for instance, indicates the effect on log price of an additional dispensary located anywhere within 1000 m of the property, whereas previously it pertained only to an additional dispensary being located within the 500â1000 m distance band from the property. When taken separately, these intensity variables indicate that a change in intensity of dispensary exposure does not significantly affect property values within 500 m (see Column 1) or up to 1000 m (see Columns 2â3) of the property. A marginal increase in intensity up to 2000 m from the property results in an increase in the property value of about 0.19 percent, or about 480.81 dollars based on the average property sale price. Broadly speaking, these results also indicate that residents prefer a marginal dispensary to be located farther from the property, though they may be indifferent to a marginal dispensary being located nearer to the property.
### 4\.4 County-specific regressions
**Table 6 County-specific benchmark regressions**
[Full size table](https://link.springer.com/article/10.1007/s00181-026-02894-6/tables/6)
Thus far we have focused exclusively on both distance and intensity measures of dispensary exposure for our four-county benchmark sample. To ascertain the extent to which there may be heterogeneity in the dispensary effects across counties, and to exploit data on two other counties, we now divide the sample and turn to county-specific regressions. Note that from the four benchmark counties, the majority of observations come from Oklahoma County, with a large group also coming from Comanche County. Relatively, few observations come from Wagoner and Washington counties. Thus, it is interesting to assess the extent to which the pooled sample results merely reflect Oklahoma and perhaps Comanche counties, or are representative of all counties more generally. For these county-specific models, we are no longer bound by a common set of control variables across counties; for each county-specific regression, we include as many control variables as are available for that county. Also, for each county we use the Fitzpatrick and Parmeter ([2021](https://link.springer.com/article/10.1007/s00181-026-02894-6#ref-CR17 "Fitzpatrick LG, Parmeter CF (2021) Data-driven estimation of treatment buffers in hedonic analysis: an examination of surface coal mines. Land Econ 97:528â547")) cross-validation method to select the optimal *minDist* threshold, and then use all property sales within 2 times that threshold for our county-optimal sample. We report the county-specific regression results in Table [6](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab6). To save space, we withhold estimates of all control variables, and provide the estimates of the dispensary effects.[Footnote 9](https://link.springer.com/article/10.1007/s00181-026-02894-6#Fn9)
Looking first the the coefficients on *minDist* across Table [6](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab6), we see that *minDist* is significantly negative in Comanche and Payne counties, insignificant in Wagoner and Washington counties, and significantly positive in Oklahoma and Tulsa counties. *Count*500 is only significant in Tulsa County, wherein we find the estimate is positive. *Count*1000 is significantly negative in Oklahoma County and significantly positive in Tulsa County, while *Count*2000 is significantly positive in Oklahoma County and significantly negative in Payne County. Taking stock of the first four columns, relative to the benchmark pooled sample regressions, it seems apparent that the results found with the four pooled counties are largely driven by the Oklahoma County effects. Table [6](https://link.springer.com/article/10.1007/s00181-026-02894-6#Tab6) shows that residents in Comanche County experience a negative *minDist* effect, indicating that property prices fall when dispensaries are located farther away from the home, though the intensity of dispensary activity does not affect property values. Looking toward Payne and Tulsa counties, we see that Payne County bears similar effects to Comanche County, while Tulsa County accords with Oklahoma County. It is worth pointing out the following. First, there is quite a bit of heterogeneity across counties in terms of both the optimal *minDist* threshold and the \\(R^2\\) measures for each regression, and readers should bear in mind that both Payne and Tulsa counties are missing several key variables (which is why they were left out of the baseline four-county model). In terms of the \\(R^2\\), Oklahoma County seems to have the best fit, relative to the regressions for the other counties.
## 5 Discussion and conclusion
We study the effects of medical marijuana dispensaries on residential property prices in Oklahoma as a means of measuring the extent to which dispensaries are seen as desirable or undesirable neighborhood amenities and thereby affect property prices. We have motivated the uniqueness of Oklahoma for our study setting: not only has Oklahoma very recently legalized medical marijuana (effective August 2018) but Oklahoma has seen a rapid rise in the proliferation of medical marijuana dispensaries and has become a leader in the nation in terms of the number of operating dispensaries. Moreover, Oklahoma is an interesting case because of the strongly conservative electorate, despite marijuana legalization being typically seen as a politically liberal interest. Thus, we believe that Oklahoma is an ideal setting for studying the effects of medical marijuana legalization on nearby property values with lessons for citizens of Oklahoma as well as the public or policymakers in other states. Using data from six counties in Oklahoma, we conduct a state-wide hedonic pricing analysis to investigate the net effects, as capitalized in property values, of local dispensary activities on residential property prices. We measure dispensary activity with both distance to nearest dispensary as well as intensity of dispensary activity in the proximate area (within 500 m, 1000 m, and 2000 m from the property). We consider pooled sample regressions as well as county-specific regressions, linear and nonlinear specifications. In looking across these six counties, we find some heterogeneity in our results. However, our results point to a few broad trends across most of Oklahoma. First, a general pattern is that residents view local proximity to a medical marijuana dispensary as a net negative, with property prices being significantly lower when either the minimum distance to the nearest dispensary is closer to the home or when a marginal dispensary is opened within 0.5 kilometers from the home. We find the effects to be quite small, with even the largest effects being less than one percent of the propertyâs value. At the same time, residents seem to prefer relatively easy access to medical marijuana dispensaries, with property prices increasing for marginal dispensaries being located at a medium distance (i.e., between 1 and 2 kilometers) from the home. Second, it is worth noting that Washington and Wagoner counties generally have insignificant effects of dispensary activities on home prices, though the sample size in those counties is substantially smaller than for the other counties.
Our results are important for residents and policymakers alike. First, despite the majority voter approval for SQ 788, it appears that residents in Oklahoma have mixed views on dispensary activity. Many residents seem to not want to live in immediate proximity, whereas many want access. It is worth noting that thus far the literature has been mixed in terms of dispensary effects on local residents; our work adds the lens of the experience in Oklahoma, and find that most residents seem to view dispensaries within immediate proximity negatively. These differences in impacts are important for policymakers to recognize, and understand when making adjustments or revisions to Oklahoma laws. Finally, readers should keep in mind that our estimates are both local and net measurements, and do not specifically measure the impact of local dispensary activities on any singular measure of well-being (such as crime rates, related alcohol use, or child abuse) as considered by others in this literature.
## References
- Adda J, McConnell B, Rasul I (2014) Crime and the depenalization of cannabis possession: evidence from a policing experiment. J Polit Econ 122(5):1130â1202
[Article](https://doi.org/10.1086%2F676932) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Crime%20and%20the%20depenalization%20of%20cannabis%20possession%3A%20evidence%20from%20a%20policing%20experiment&journal=J%20Polit%20Econ&doi=10.1086%2F676932&volume=122&issue=5&pages=1130-1202&publication_year=2014&author=Adda%2CJ&author=McConnell%2CB&author=Rasul%2CI)
- Anderson DM, Hansen B, Rees DI (2013) Medical marijuana laws, traffic fatalities, and alcohol consumption. J Law Econ 56(2):333â369
[Article](https://doi.org/10.1086%2F668812) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Medical%20marijuana%20laws%2C%20traffic%20fatalities%2C%20and%20alcohol%20consumption&journal=J%20Law%20Econ&doi=10.1086%2F668812&volume=56&issue=2&pages=333-369&publication_year=2013&author=Anderson%2CDM&author=Hansen%2CB&author=Rees%2CDI)
- Anderson DM, Hansen B, Rees DI (2015) Medical marijuana laws and teen marijuana use. Am Law Econ Rev 17(2):495â528
[Article](https://doi.org/10.1093%2Faler%2Fahv002) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Medical%20marijuana%20laws%20and%20teen%20marijuana%20use&journal=Am%20Law%20Econ%20Rev&doi=10.1093%2Faler%2Fahv002&volume=17&issue=2&pages=495-528&publication_year=2015&author=Anderson%2CDM&author=Hansen%2CB&author=Rees%2CDI)
- Boggess LN, PĂ©rez DM, Cope K, Root C, Stretesky PB (2014) Do medical marijuana centers behave like locally undesirable land uses? Implications for the geography of health and environmental justice. Urban Geogr 35(3):315â336
[Article](https://doi.org/10.1080%2F02723638.2014.881018) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Do%20medical%20marijuana%20centers%20behave%20like%20locally%20undesirable%20land%20uses%3F%20Implications%20for%20the%20geography%20of%20health%20and%20environmental%20justice&journal=Urban%20Geogr&doi=10.1080%2F02723638.2014.881018&volume=35&issue=3&pages=315-336&publication_year=2014&author=Boggess%2CLN&author=P%C3%A9rez%2CDM&author=Cope%2CK&author=Root%2CC&author=Stretesky%2CPB)
- Boswell K (2021) Essays on Cannacis legalization in California. PhD Thesis, University of Miami
- Brinkman J, Mok-Lamme D (2019) Not in my backyard? Not so fast. The effect of marijuana legalization on neighborhood crime. Reg Sci Urban Econ 78:103460
[Article](https://doi.org/10.1016%2Fj.regsciurbeco.2019.103460) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Not%20in%20my%20backyard%3F%20Not%20so%20fast.%20The%20effect%20of%20marijuana%20legalization%20on%20neighborhood%20crime&journal=Reg%20Sci%20Urban%20Econ&doi=10.1016%2Fj.regsciurbeco.2019.103460&volume=78&publication_year=2019&author=Brinkman%2CJ&author=Mok-Lamme%2CD)
- Brooks TJ, Humphreys BR, Nowak A (2018) Strip clubs, âsecondary effects" and residential property prices. Real Estate Econ 48(3):850â885
[Article](https://doi.org/10.1111%2F1540-6229.12236) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Strip%20clubs%2C%20%E2%80%9Csecondary%20effects%22%20and%20residential%20property%20prices&journal=Real%20Estate%20Econ&doi=10.1111%2F1540-6229.12236&volume=48&issue=3&pages=850-885&publication_year=2018&author=Brooks%2CTJ&author=Humphreys%2CBR&author=Nowak%2CA)
- Brown T, Howe J (2019) Number of medical marijuana outlets soars. How many are in your area?
- Burkhardt J, Flyr M (2019) The effect of marijuana dispensary openings on housing prices. Contemp Econ Policy 37(3):462â475
[Article](https://doi.org/10.1111%2Fcoep.12414) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20effect%20of%20marijuana%20dispensary%20openings%20on%20housing%20prices&journal=Contemp%20Econ%20Policy&doi=10.1111%2Fcoep.12414&volume=37&issue=3&pages=462-475&publication_year=2019&author=Burkhardt%2CJ&author=Flyr%2CM)
- Chang TY, Jacobson M (2017) Going to pot? The impact of dispensary closures on crime. J Urban Econ 100:120â136
[Article](https://doi.org/10.1016%2Fj.jue.2017.04.001) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Going%20to%20pot%3F%20The%20impact%20of%20dispensary%20closures%20on%20crime&journal=J%20Urban%20Econ&doi=10.1016%2Fj.jue.2017.04.001&volume=100&pages=120-136&publication_year=2017&author=Chang%2CTY&author=Jacobson%2CM)
- Cheng C, Mayer WJ, Mayer Y (2018) The effect of legalizing retail marijuana on housing values: evidence from Colorado. Econ Inq 56(3):1585â1601
[Article](https://doi.org/10.1111%2Fecin.12556) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20effect%20of%20legalizing%20retail%20marijuana%20on%20housing%20values%3A%20evidence%20from%20Colorado&journal=Econ%20Inq&doi=10.1111%2Fecin.12556&volume=56&issue=3&pages=1585-1601&publication_year=2018&author=Cheng%2CC&author=Mayer%2CWJ&author=Mayer%2CY)
- Cohn AM, Sedani A, Niznik T, Alexander A, Lowery B, McQuoid J, Campbell J (2023) Population and neighborhood correlates of cannabis dispensary locations in Oklahoma. Cannabis 6(1):99â113
[Article](https://doi.org/10.26828%2Fcannabis%2F2023.01.008) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Population%20and%20neighborhood%20correlates%20of%20cannabis%20dispensary%20locations%20in%20Oklahoma&journal=Cannabis&doi=10.26828%2Fcannabis%2F2023.01.008&volume=6&issue=1&pages=99-113&publication_year=2023&author=Cohn%2CAM&author=Sedani%2CA&author=Niznik%2CT&author=Alexander%2CA&author=Lowery%2CB&author=McQuoid%2CJ&author=Campbell%2CJ)
- Conklin J, Drop M, Li H (2020) Contact high: the external effects of retail marijuana establishments on house prices. Real Estate Econ 48(1):135â173
[Article](https://doi.org/10.1111%2F1540-6229.12220) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Contact%20high%3A%20the%20external%20effects%20of%20retail%20marijuana%20establishments%20on%20house%20prices&journal=Real%20Estate%20Econ&doi=10.1111%2F1540-6229.12220&volume=48&issue=1&pages=135-173&publication_year=2020&author=Conklin%2CJ&author=Drop%2CM&author=Li%2CH)
- Delgado MS, Guilfoos T, Boslett A (2016) The cost of unconventional gas extraction: a hedonic analysis. Resour Energy Econ 46:1â22
[Article](https://doi.org/10.1016%2Fj.reseneeco.2016.07.001) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20cost%20of%20unconventional%20gas%20extraction%3A%20a%20hedonic%20analysis&journal=Resour%20Energy%20Econ&doi=10.1016%2Fj.reseneeco.2016.07.001&volume=46&pages=1-22&publication_year=2016&author=Delgado%2CMS&author=Guilfoos%2CT&author=Boslett%2CA)
- Denham BE (2019) Attitudes toward legalization of marijuana in the United States, 1986â2016: changes in determinants of public opinion. Int J Drug Policy 71:78â90
[Article](https://doi.org/10.1016%2Fj.drugpo.2019.06.007) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Attitudes%20toward%20legalization%20of%20marijuana%20in%20the%20United%20States%2C%201986%E2%80%932016%3A%20changes%20in%20determinants%20of%20public%20opinion&journal=Int%20J%20Drug%20Policy&doi=10.1016%2Fj.drugpo.2019.06.007&volume=71&pages=78-90&publication_year=2019&author=Denham%2CBE)
- Dong X, Tyndall J (2024) The impact of recreational marijuana dispensaries on crime: evidence from a lottery experiment. Ann Reg Sci 72:1383â1414
[Article](https://link.springer.com/doi/10.1007/s00168-023-01246-x) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20impact%20of%20recreational%20marijuana%20dispensaries%20on%20crime%3A%20evidence%20from%20a%20lottery%20experiment&journal=Ann%20Reg%20Sci&doi=10.1007%2Fs00168-023-01246-x&volume=72&pages=1383-1414&publication_year=2024&author=Dong%2CX&author=Tyndall%2CJ)
- Fitzpatrick LG, Parmeter CF (2021) Data-driven estimation of treatment buffers in hedonic analysis: an examination of surface coal mines. Land Econ 97:528â547
[Article](https://doi.org/10.3368%2Fle.97.3.528) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Data-driven%20estimation%20of%20treatment%20buffers%20in%20hedonic%20analysis%3A%20an%20examination%20of%20surface%20coal%20mines&journal=Land%20Econ&doi=10.3368%2Fle.97.3.528&volume=97&pages=528-547&publication_year=2021&author=Fitzpatrick%2CLG&author=Parmeter%2CCF)
- Freisthler B, Gruenewald PJ, Wolf JP (2015) Examining the relationship between marijuana use, medical marijuana dispensaries, and abusive and neglectful parenting. Child Abuse & Neglect 48:170â178
[Article](https://doi.org/10.1016%2Fj.chiabu.2015.07.008) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Examining%20the%20relationship%20between%20marijuana%20use%2C%20medical%20marijuana%20dispensaries%2C%20and%20abusive%20and%20neglectful%20parenting&journal=Child%20Abuse%20%26%20Neglect&doi=10.1016%2Fj.chiabu.2015.07.008&volume=48&pages=170-178&publication_year=2015&author=Freisthler%2CB&author=Gruenewald%2CPJ&author=Wolf%2CJP)
- Huber A III, Newman R, LaFave D (2016) Cannabis control and crime: medicinal use, depenalization and the war on drugs. BE J Econ Anal Policy 16(4):20150167
[Article](https://doi.org/10.1515%2Fbejeap-2015-0167) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Cannabis%20control%20and%20crime%3A%20medicinal%20use%2C%20depenalization%20and%20the%20war%20on%20drugs&journal=BE%20J%20Econ%20Anal%20Policy&doi=10.1515%2Fbejeap-2015-0167&volume=16&issue=4&publication_year=2016&author=Huber%2CA&author=Newman%2CR&author=LaFave%2CD)
- Iannacchione B, Ward KC, Evans MK (2020) Perceptions of NIMBY syndrome among Colorado and Washington dispensary owners and managers. Justice Policy J 17(2)
- Kepple NJ, Freisthler B (2012) Exploring the ecological association between crime and medical marijuana dispensaries. J Stud Alcohol Drugs 73(4):523â530
[Article](https://doi.org/10.15288%2Fjsad.2012.73.523) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Exploring%20the%20ecological%20association%20between%20crime%20and%20medical%20marijuana%20dispensaries&journal=J%20Stud%20Alcohol%20Drugs&doi=10.15288%2Fjsad.2012.73.523&volume=73&issue=4&pages=523-530&publication_year=2012&author=Kepple%2CNJ&author=Freisthler%2CB)
- Lu R, Willits D, Stohr MK, Makin D, Snyder J, Lovrich N, Meize M, Stanton D, Wu G, Hemmens C (2021) The cannabis effect on crime: time-series analysis of crime in Colorado and Washington State. Justice Q 38(4):565â595
[Article](https://doi.org/10.1080%2F07418825.2019.1666903) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20cannabis%20effect%20on%20crime%3A%20time-series%20analysis%20of%20crime%20in%20Colorado%20and%20Washington%20State&journal=Justice%20Q&doi=10.1080%2F07418825.2019.1666903&volume=38&issue=4&pages=565-595&publication_year=2021&author=Lu%2CR&author=Willits%2CD&author=Stohr%2CMK&author=Makin%2CD&author=Snyder%2CJ&author=Lovrich%2CN&author=Meize%2CM&author=Stanton%2CD&author=Wu%2CG&author=Hemmens%2CC)
- Mair C, Freisthler B, Ponicki WR, Gaidus A (2015) The impacts of marijuana dispensary density and neighborhood ecology on marijuana abuse and dependence. Drug Alcohol Depend 154:111â116
[Article](https://doi.org/10.1016%2Fj.drugalcdep.2015.06.019) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20impacts%20of%20marijuana%20dispensary%20density%20and%20neighborhood%20ecology%20on%20marijuana%20abuse%20and%20dependence&journal=Drug%20Alcohol%20Depend&doi=10.1016%2Fj.drugalcdep.2015.06.019&volume=154&pages=111-116&publication_year=2015&author=Mair%2CC&author=Freisthler%2CB&author=Ponicki%2CWR&author=Gaidus%2CA)
- Morris RG, TenEyck M, Barnes JC, Kovandzic TV (2014) The effect of medical marijuana laws on crime: evidence from state panel data, 1990â2006. PLoS ONE 9(3):e92816
[Article](https://doi.org/10.1371%2Fjournal.pone.0092816) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20effect%20of%20medical%20marijuana%20laws%20on%20crime%3A%20evidence%20from%20state%20panel%20data%2C%201990%E2%80%932006&journal=PLoS%20ONE&doi=10.1371%2Fjournal.pone.0092816&volume=9&issue=3&publication_year=2014&author=Morris%2CRG&author=TenEyck%2CM&author=Barnes%2CJC&author=Kovandzic%2CTV)
- Morrison C, Gruenewald PJ, Freisthler B, Ponicki WR, Remer LG (2014) The economic geography of medical cannabis dispensaries in California. Int J Drug Policy 25(3):508â515
[Article](https://doi.org/10.1016%2Fj.drugpo.2013.12.009) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20economic%20geography%20of%20medical%20cannabis%20dispensaries%20in%20California&journal=Int%20J%20Drug%20Policy&doi=10.1016%2Fj.drugpo.2013.12.009&volume=25&issue=3&pages=508-515&publication_year=2014&author=Morrison%2CC&author=Gruenewald%2CPJ&author=Freisthler%2CB&author=Ponicki%2CWR&author=Remer%2CLG)
- Palmquist R D (1991) Hedonic methods. Measuring the demand for environmental quality
- Parmeter C F, Pope J C (2013) Quasi-experiments and hedonic property value methods. In: Handbook on Experimental Economics and the Environment, pp 3â66. Edward Elgar Publishing
- Rosen S (1974) Hedonic prices and implicit markets: product differentiation in pure competition. J Polit Econ 82(1):34â55
[Article](https://doi.org/10.1086%2F260169) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Hedonic%20prices%20and%20implicit%20markets%3A%20product%20differentiation%20in%20pure%20competition&journal=J%20Polit%20Econ&doi=10.1086%2F260169&volume=82&issue=1&pages=34-55&publication_year=1974&author=Rosen%2CS)
- Shi Y (2016) The availability of medical marijuana dispensary and adolescent marijuana use. Prev Med 91:1â7
[Article](https://doi.org/10.1016%2Fj.ypmed.2016.07.015) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20availability%20of%20medical%20marijuana%20dispensary%20and%20adolescent%20marijuana%20use&journal=Prev%20Med&doi=10.1016%2Fj.ypmed.2016.07.015&volume=91&pages=1-7&publication_year=2016&author=Shi%2CY)
- Taylor LO (2003) The hedonic method. A primer on nonmarket valuation. Springer, Cham, pp 331â393
[Chapter](https://link.springer.com/doi/10.1007/978-94-007-0826-6_10) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=The%20hedonic%20method&doi=10.1007%2F978-94-007-0826-6_10&pages=331-393&publication_year=2003&author=Taylor%2CLO)
- Thomas D, Tian L (2021) Hits from the bong: the impact of recreational marijuana dispensaries on property values. Reg Sci Urban Econ 87:103655
[Article](https://doi.org/10.1016%2Fj.regsciurbeco.2021.103655) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Hits%20from%20the%20bong%3A%20the%20impact%20of%20recreational%20marijuana%20dispensaries%20on%20property%20values&journal=Reg%20Sci%20Urban%20Econ&doi=10.1016%2Fj.regsciurbeco.2021.103655&volume=87&publication_year=2021&author=Thomas%2CD&author=Tian%2CL)
- Tyndall J (2021) Getting high and low prices: marijuana dispensaries and home values. Real Estate Econ 49(4):1093â1119
[Article](https://doi.org/10.1111%2F1540-6229.12302) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Getting%20high%20and%20low%20prices%3A%20marijuana%20dispensaries%20and%20home%20values&journal=Real%20Estate%20Econ&doi=10.1111%2F1540-6229.12302&volume=49&issue=4&pages=1093-1119&publication_year=2021&author=Tyndall%2CJ)
- Whittemore AH, BenDor TK (2019) Reassessing NIMBY: the demographics, politics, and geography of opposition to high-density residential infill. J Urban Aff 41(4):423â442
[Article](https://doi.org/10.1080%2F07352166.2018.1484255) [Google Scholar](http://scholar.google.com/scholar_lookup?&title=Reassessing%20NIMBY%3A%20the%20demographics%2C%20politics%2C%20and%20geography%20of%20opposition%20to%20high-density%20residential%20infill&journal=J%20Urban%20Aff&doi=10.1080%2F07352166.2018.1484255&volume=41&issue=4&pages=423-442&publication_year=2019&author=Whittemore%2CAH&author=BenDor%2CTK)
- Zimmer R, Hilburn S, Van Leuven A, Whitacre B (2022) Medical marijuana dispensary locations across Oklahoma. Agricultural Economics Department Report, Oklahoma State University
[Download references](https://citation-needed.springer.com/v2/references/10.1007/s00181-026-02894-6?format=refman&flavour=references) |
| Shard | 129 (laksa) |
| Root Hash | 17645177711233004329 |
| Unparsed URL | com,springer!link,/article/10.1007/s00181-026-02894-6 s443 |