As 2024 begins, Americans in twenty-four states and D.C. find themselves in jurisdictions where recreational marijuana is legal. Public opinion nationwide is now decisively in favor of legalization, with support hovering around 70%. Politicians have largely abandoned it as a polarizing issue. This was not always the case; at the turn of the millennium, it was decisively unpopular, with roughly a third of Americans supporting legalization. Indeed, the story of how recreational marijuana won over voters is largely one that takes place over the past decade.
Few issues have seen a more rapid rise in the court of public opinion than marijuana legalization. Support for legalization still reflects partisan divides, but is nonetheless depolarized relative to many other current issues — and has been for a long time. In 2012, Colorado and Washington became the first states to successfully legalize marijuana via referendum. Legalization won by about 11 percentage points in both states. In Colorado, that was a substantial overperformance over Obama’s win, but in Washington, it was a slight underperformance. It would not be the first time such differences in coalitions manifested.
While Democrats have long been more supportive of marijuana legalization, a significant portion of the party’s voters, particularly among nonwhites, are less warm to the idea. By contrast, Republicans have historically opposed legalization, though far from universally. Over time, legalization gained favor in both parties, broadly consolidating the Democratic vote and making significant inroads with Republicans. This has largely been reflected in referendum results over time.
Depolarization is a difficult concept to measure directly, but the most expedient way to do so may be to utilize geography. In 2013, the National Center for Health Statistics (NCHS) helpfully compiled an urban-rural classification scheme for the entire nation. This system utilizes data from the OMB and the Census Bureau to create several distinct categories of counties. It is not perfect; it could be argued that some counties are miscategorized and the data is somewhat outdated. However, it remains a systematic approach that applies consistent standards.
Some NCHS categories have been combined.
Large Central Metro: This group of counties are large and urban, containing the central cities of metropolitan areas that exceed 1 million residents. Owing to these traits, they are often diverse and lean strongly Democratic.
Large Outer Metro: These counties constitute the remaining suburban or exurban jurisdictions in large metropolitan areas. They are collectively well-educated and wealthy, and some of the counties in this category have moved substantially toward the Democratic Party in the modern era.
Small/Medium Metro: All metropolitan areas with a population less than 1 million are grouped into this category. Counties in this category are usually smaller, lower density, and less well-off than their counterparts in larger metropolitan areas. As such, this group is significantly more Republican.
Rural: This category compromises all micropolitan areas and noncore counties. This population lives almost exclusively in small towns or rurals, and are consequently strongly Republican.
Utilizing these categories, it is possible to view how the geography of legalization referendums have changed.
2016
2016 saw five states hold referendums on legalization. By this point in time, public opinion had already swung sharply in its favor. Pew Research found that 66% of Democrats supported legalization, while just 41% of Republicans did. It also found support from 59% of both white and Black voters, while hovering around 46% for Hispanics.
This depolarization is reflected in 2016’s election results. Overall, the Large Central counties voted for Hillary Clinton by 31 percentage points, but legalization won by just 14. By contrast, the Rural counties voted for Trump by 7 percentage points, but opposed legalization by a narrower margin.
2018
As support increased for marijuana nationwide, the performance gap between heavily Democratic and heavily Republican areas began to shift. Pew Research found Democratic support for legalization rising to 78% by 2019, with Republican support climbing — by a lesser extent — to 44%. Similarly, support rose to 67% among Black voters and 63% among Hispanics. In 2018, these shifts were seen in Michigan and North Dakota. Legalization only underperformed Clinton’s margin by 7 percentage points in the Large Central counties, while outrunning her by double digits in every other category.
2020
By 2020, support for legalization had consolidated most Democratic voters while Republicans remained divided on the issue. This is best shown by legalization’s string of massive overperformances relative to Joe Biden. Only in New Jersey’s heavily Democratic Large Central counties did legalization underperform Biden. In largely rural, Republican states like Montana and South Dakota, legalization significantly outran Biden nearly everywhere.
2022/23
In the Biden era, support for legalization has plateaued around 70%. Gallup found that 81% of Democrats and a majority of Republicans now support legalization. Most blue states had already legalized marijuana, either through legislation or referendum, by this point. Consequently, with the sole exception of Maryland, the referendums in this era take place in red states. As such, a greater portion of these referendums failed, although strikingly legalization has begun to more consistently outperform Biden in all categories.
There are a few limitations to this approach. First, very few states have had multiple referendums on the issue in the time period selected (2016 to present). As such, direct comparisons usually cannot be made, and the geographic results of each year should be considered separately.
Second, since each year features a set of different states and different counties, it is more important to view the referendum performances in comparison to each year’s unique batch of presidential results, rather than directly comparing each referendum performance year-to-year.
Finally, while geography is being used as a proxy for partisanship and demographics, there are notable differences from county to county. For example, Baltimore City and Maricopa County are both classified as Large Central counties, but Maricopa is far more suburban and Republican than the nearly entirely urban Baltimore.
The past decade has seen a marked shift in opinions toward marijuana legalization. In referendums, it was never unusual to see legalization underperform traditional Democratic margins, especially in heavily nonwhite areas, while outperforming them in rural areas. This reflected divisions along multiple demographic lines, such as age and race. Nevertheless, as support surged across the board, underperformances relative to Democratic margins have noticeably decreased, especially in metropolitan areas. Meanwhile, overperformances in Republican rurals have increased.
Nevertheless, support among Republican legislators has remained low. Very few states currently controlled by Republicans have pursued legalization, while nearly all Democratic states have. Despite increasing support among Republican voters, legalization has still failed in recent referendums in very red states like Arkansas or the Dakotas. Consequently, most of the proverbial “low-hanging fruit” has been picked. It may be that legalization efforts have plateaued – for now.
I am an analyst specializing in elections and demography, as well as a student studying political science, sociology, and data science at Vanderbilt University. I use election data to make maps and graphics. In my spare time, you can usually find me somewhere on the Chesapeake Bay. You can find me at @maxtmcc on Twitter.