Skip to main content
There Goes the Neighbourhood Legalised Marijuana.jpg

Economics & Finance

There Goes the Neighbourhood: Legalised Marijuana and Property Values

There Goes the Neighbourhood: Legalised Marijuana and Property Values

When a recreational marijuana dispensary opens, it depresses property prices in its immediate vicinity.

One third of Americans now live in a state that has legalised the use of recreational marijuana. A great deal of research has looked at the impact of this legalisation on people’s welfare, but results have been inconclusive. Some studies uncovered a positive impact, others a negative one. In many a case, the methodology used had inherent limitations. Due to the broad nature of the issues they study, social scientists often struggle to establish causation.

So when Danna Thomas (University of South Carolina) and I learned that Washington state was holding a lottery to award marijuana dispensary licences, it immediately caught our attention. For social scientists, lotteries themselves mean a jackpot as they create ideal, natural experiment conditions where the only difference between two groups is luck of the draw – the most attractive setting to pin down causation. We thus set out to investigate whether shops that legally sell marijuana would impact housing values in their neighbourhood.

In our new paper, we show that the prices of homes around a new dispensary (within 0.36 miles or 600m) fall by 3-4 percent on average, compared to control areas. This decrease particularly affected younger, more diverse neighbourhoods. For the average home sale, it translated into a price drop of between US$10,100 and US$13,500.

Winning and losing addresses

After Washington state legalised the sale of marijuana in November 2012, it opted to spread licences across jurisdictions based on population density. Given the contentious nature of the business, it also opted to award licences via an open lottery. Applicants were required to include the potential address of their store. A total of 1,173 applications were submitted for a total of 334 available licences. Lottery results were published in May 2014 and licensees could begin selling as early as July 2014.

Even though proposed business locations were not legally binding, 75 percent of firms opened their store within one third of a mile (500m) from an address stated in an application. This suggests that these addresses shared attributes making them similarly attractive for marijuana retail. Submitted addresses that were not awarded a licence in the lottery were our perfect control group.

We then analysed home sales from January 2012 to February 2016, using data from private vendors. Within a given neighbourhood, we compared properties situated nearby a dispensary to others located slightly further away, using a concentric-circle, or “ring”, approach. Many studies have used this geographical analysis to compare the immediate radius around a dispensary to the next wider ring. But in our case, the lottery process allowed us to also compare locations with a marijuana dispensary (the winning applications) to addresses that were entered in the lottery but didn’t have a nearby dispensary (the losing applications).

This extra layer of comparison allowed us to avoid the problem of self-selection – in the absence of a lottery, locations that attract this type of business may not be entirely random – and hence establish a causal link between the entry of a marijuana dispensary and changes in property prices.

A few bearing the cost of the benefits afforded to many

Currently a lot of discussion about marijuana legalisation focuses on its positive effects at the level of the general population. For instance, it can increase tax revenues, decrease incarceration rates and, as commonly known, combat stress, anxiety and depression. Our study draws attention to the fact that these society-wide benefits can have highly localised costs.

To put into perspective our observed 3 to 4 percent drop in property prices, we note from past studies that the arrival of a registered sex offender in a neighbourhood reduces property prices by 4 percent. Proximity to a power plant decreases housing values and rents by 3 to 7 percent, while the presence of a toxic plant lead to declines of 11 percent for homes within 0.5 miles (800m) of the plants.

Property values reflect, to a certain extent, buyers’ and sellers’ valuation of local amenities. As marijuana businesses are not legal at the federal level, they are often unable to open a bank account and thus forced to operate in cash. The decrease in property prices could be triggered by a concern that cash-only dispensaries may attract more burglars and criminals in the neighbourhood.

Looking at the particular case of Seattle, we found that while crime in general did not go up, there was a significant increase in nuisance-related crimes (e.g. disorderly conduct, loitering) in the neighbourhoods with a marijuana dispensary, compared to those devoid of them. These nuisance-related crime reports increased by 4.2 per 10,000 residents.

If a government wants to grow acceptance of marijuana legalisation, it should look at the population’s perception of its potential negative effects and design relevant public policies. Subsidies meant to mitigate negative impacts could be offered at the highly localised level where they are most likely – or at least perceived – to occur.

About the author(s)

View Comments

Anonymous User

05/06/2021, 03.20 pm
this needs to be compared to the effects bars/liquor stores have on property values. mitigation measures might include surveillance cameras surrounding these facilities and using tax revenues to compensate neighboring property owners.


Anonymous User

29/05/2021, 02.54 am

I'm sure anti-pot activists are going to LOVE your statement "there was a significant increase in nuisance-related crimes" and will never mention the data it was based on. However, in anyone's book, "4.2 per 10,000" increase IS NOT SIGNIFICANT!! I guess if you wrote "there was a .0004% increase in nuisance-related crimes" you couldn't make a big deal out of it, since it surely falls well below the statistical accuracy of the data you collected.

Leave a Comment
Please log in or sign up to comment.