The Budgetary Effects of Ending Drug Prohibition

Drug legalization affects various social outcomes. In the debate over marijuana legalization, academics and the media tend to focus on how legalization affects public health and criminal justice outcomes. But policymakers and scholars should also consider the fiscal effects of drug liberalization. Legalization can reduce government spending, which saves resources for other uses, and it generates tax revenue that transfers income from drug producers and consumers to public coffers.

Drawing on the most recent available data, this bulletin estimates the fiscal windfall that would be achieved through drug legalization. All told, drug legalization could generate up to $106.7 billion in annual budgetary gains for federal, state, and local governments. Those gains would come from two primary sources: decreases in drug enforcement spending and increases in tax revenue. This bulletin estimates that state and local governments spend $29 billion on drug prohibition annually, while the federal government spends an additional $18 billion. Meanwhile, full drug legalization would yield $19 billion in state and local tax revenue and $39 billion in federal tax revenue.

In addition, this bulletin briefly examines the budgetary effects of state marijuana legalizations that have already taken place in Colorado, Oregon, and Washington. This study finds that, so far, legalization in those states has generated more tax revenue than previously forecast but generated essentially no reductions in criminal justice expenditure. The bulletin offers possible explanations for those findings.

Estimates for Expenditures and Tax Revenue

Jeffrey Miron and Katherine Waldock previously estimated government anti‐​drug expenditures and potential tax revenue from legalization in a 2010 Cato Institute study.¹ That study examined the reduction in criminal justice spending ($41.3 billion per year) and the increase in tax revenue ($46.7 billion per year) that would result from legalizing all drugs in the United States at the federal level and in all states.

This study draws on more recent data and presents exciting new conclusions. These updates are useful because a number of states have legalized marijuana since the 2010 report; this provides some evidence on the validity of the 2010 estimates.² Moreover, additional states are considering marijuana legalization going forward, and updated estimates might be relevant to the legalization debates in those states.

This report uses the same analytic framework as the 2010 Cato study and therefore omits some details.³ The underlying data in this update are the latest numbers available and are sourced from 2015 or 2016 unless otherwise noted. Tables 1–6 in this report update the key information from Tables 2–7 in the 2010 report.

Table 1 shows estimated state and local criminal justice expenditures related to drug prohibition in 2016.⁴ In nearly all categories, estimated expenditure has risen, albeit slightly. For all drugs, the estimate is $29.4 billion; for marijuana, $6.0 billion; for heroin and cocaine, $12.8 billion; for synthetic drugs, $4.9 billion; and for all other drugs, $5.6 billion. Adjusted for inflation, these figures represent only about a 3 percent increase since 2010.

Table 2 provides a state‐​by‐​state breakdown of state and local expenditure on drug prohibition in 2016.

Table 3 presents estimated federal spending on drug prohibition enforcement for 2015 (in 2016 dollars). Real federal spending has risen by 4 percent since 2008. This report attributes that change to growth in the U.S. population.

Table 4 presents updated estimates of the tax revenue that federal, state, and local governments could collect if drugs were legal. Compared with the original study, these figures suggest that overall federal and state tax revenue would be higher than previously estimated after accounting for inflation; the federal government would take in $39.2 billion in federal tax revenues as a result of legalization today, compared with $31.2 billion in 2008. In real terms, this represents growth of about 12 percent. States would collect $19.6 billion today, compared with $15.6 billion in 2008. Beyond the effects of population growth and inflation, this upward trend reflects increasing use of marijuana, cocaine, and heroin—and therefore increasing consumer spending and potential tax revenue associated with those substances.

Table 1: State and local expenditures attributable to drug prohibition, billions of dollars, 2016

Source: Author’s calculations

Table 2: State and local expenditures attributable to drug prohibition, millions of dollars, 2016

Source: Author’s calculations.

Table 3: Federal expenditures attributable to drug prohibition, billions of dollars, 2015 (in 2016 dollars)

Source: Author’s calculations.

Table 4: State and federal tax revenues from drug legalization, billions of dollars, 2016

Source: Author’s calculations.

Table 5 estimates the tax revenue generated in each state by allocating the estimates from Table 4 to each state on the basis of population.

Table 6 summarizes the updated estimates for expenditure savings and additional revenues tied to drug legalization. Three aspects stand out. First, the total effect of drug legalization on government budgets would be approximately $106.7 billion in combined savings and additional revenue. In real terms, that marks an 8 percent increase from the estimates in the 2010 Cato study. Second, as in the previous report, nearly 60 percent of budgetary gains would come from legalizing heroin and cocaine. Third, the fiscal benefits of drug legalization would be roughly evenly shared between the states and the federal government.

Table 5: State tax revenues from drug legalization, distributed by population, millions of dollars, 2016

Source: Author’s calculations.

Table 6: Summary of expenditure savings and additional revenues from drug legalization, billions of dollars, 2016

Source: Author’s calculations.

Tax Revenue Comparisons

This section compares estimates from the 2010 Cato study with observed tax revenues following state‐​level marijuana legalization in a handful of U.S. states. Comparing those predictions to actual outcomes sheds light on the accuracy of the 2010 study.

Table 7 presents marijuana‐​related tax revenue from Colorado, Oregon, and Washington, the three states that had legalized and implemented recreational marijuana commerce at the time of writing. Although Colorado and Washington officially legalized marijuana in 2012 and Oregon followed suit in 2014, marijuana sales did not commence until regulatory frameworks were established. In each state, that process took several years. For example, Colorado did not begin issuing licenses to sell retail marijuana until the end of 2014. Table 7 also lists the tax revenue projections from the original report.

Washington collected nearly $70 million in marijuana tax revenues during the first year of legalization, almost exactly the estimate in the 2010 report once adjusted for inflation. In fiscal year 2016, however, Washington collected nearly triple that amount, and in fiscal year 2017 tax revenues reached nearly $320 million. Oregon collected only $20.6 million in fiscal year 2016, about half the 2010 estimate, but it collected $70.3 million in fiscal year 2017, well above the 2010 estimate. In Colorado, marijuana tax revenues have risen from $67.6 million in calendar year 2014 to $247.4 million in calendar year 2017. Even adjusting for inflation, those figures far outstrip the 2010 estimates as well as the updated estimates presented in this paper.

The discrepancy between the 2010 estimates and experience so far reflects, in part, an unexpectedly high amount of marijuana tourism in these three states; initial reports suggest that out‐​of‐​state marijuana shoppers account for a significant fraction of tax revenue. For example, a 2015 survey of adult tourists in Colorado found that 23 percent identified legal marijuana as a reason they traveled to the state.⁵

Alternatively, the discrepancy between the 2010 estimates and experience so far may indicate that some of the assumptions behind the original estimates were incorrect. For example, the 2010 report assumed that marijuana prices would fall by 50 percent in states that legalized; however, pricing data analyzed in a 2016 Cato Institute study by Angela Dills, Sietse Goffard, and Jeffrey Miron suggest that marijuana prices have not dropped that much.⁶ If true, this assertion would help explain the higher‐​than‐​expected tax revenue.

The implications of these initial data are therefore unclear. One caveat is that tax revenues are still in flux because of the recent nature of existing legalizations. Revenues may continue to increase over time as more stores open or if demand increases as a result of greater cultural acceptance of marijuana. This is a plausible explanation for the observed growth in tax revenue; Colorado, Oregon, and Washington have all seen gradually rising levels of marijuana use, according to survey data. But revenues in existing legalization states may also moderate if other states or the federal government legalize marijuana.⁷ Another consideration is that a nontrivial share of tax revenue in Colorado, Oregon, and Washington has been generated from collection of one‐​time application and licensing fees. To date, Colorado has collected $57.3 million in marijuana licensing and application fees, or about 8 percent of total state marijuana revenues since legalization.⁸ As recreational marijuana becomes a more established industry, states will likely see a decline in the number of new entrants and therefore a decline in licensing revenue.

Table 7: State tax revenue from marijuana, 2016 dollars

Sources: Colorado Department of Revenue; Washington State Liquor and Cannabis Board; Oregon Department of Revenue.

Further, if marijuana were legalized at the federal level, it would likely be taxed at both the state and federal level, similarly to how cigarettes are currently taxed. The addition of a federal tax would increase the price and drive down demand. States would then see less revenue as users reduced use in response to the price change. As California prepares to tax recreational marijuana sales statewide with a tax rate of 45 percent, we will be able, in the future, to observe consumer responses to widely varying tax rates.

Criminal Justice Expenditures in States with Marijuana Legalization

This section examines the effect of marijuana legalization on state‐​level criminal justice expenditures. Unlike tax revenues, no direct indicator shows how legalization affects state spending. Nevertheless, no evidence to date suggests that legalization generates a sharp decline in police, judicial, or correctional expenditures.

Table 8 shows marijuana arrests in Oregon and Washington and the percentage of total drug arrests and total statewide arrests attributed to marijuana offenses. Colorado does not publish comparable data for marijuana‐​specific offenses, so the table reports total arrests for any drug‐​related violation. The shaded box indicates the year in which marijuana legalization measures were passed.

Despite the sharp decline in marijuana arrests, criminal justice expenditures in Colorado, Oregon, and Washington have risen slightly, as shown in Table 9. One possible explanation is that marijuana offenses accounted for a small share of arrests and prosecutions even before legalization. In Table 2, the criminal justice expenditure attributable to marijuana represented only 15 percent of total expenditure in these three states. Another possible explanation is that states are shifting resources toward other types of drug and nondrug crimes.

Table 8: Marijuana arrests

Sources: Oregon Criminal Justice Information Services, https://​www​.ore​gon​.gov/​o​s​p​/​C​J​I​S​/​P​a​g​e​s​/​a​n​n​u​a​l​_​r​e​p​o​r​t​s​.aspx; Washington Association of Sheriffs and Police Chiefs, http://​www​.waspc​.org/​c​r​i​m​e​-​s​t​a​t​i​s​t​i​c​s​-​r​e​ports; Federal Bureau of Investigations Uniform Crime Reports.

Table 9: State and local expenditures by function (thousands of dollars)

Source: United States Census Bureau, http://​www​.cen​sus​.gov/​g​o​v​s​/​local.

Conclusion

At both the federal and state levels, government budgets would benefit enormously from drug legalization policies. This report estimates that $47.9 billion is spent annually on drug prohibition enforcement, whereas $58.8 billion could potentially be raised in tax revenue. Combined, these figures suggest that the fiscal windfall of drug legalization could be as high as $107 billion. Moreover, although media outlets and policymakers mostly focus on marijuana, the majority of budgetary gains would likely come from legalizing heroin and cocaine. In addition to providing new estimates of the revenue and expenditure implications of drug legalization, this report suggests two conclusions based on experience to date from state‐​level legalizations.

First, the tax revenues generated by legalization might significantly exceed the estimates in the 2010 Cato study. At the same time, several factors might moderate the tax revenues if legalization occurs widely at the state level or at the federal level.

Second, the reductions in criminal justice expenditures from legalization are likely to be modest in practice, even if the number of drug arrests falls substantially. Early experience suggests that governments will reallocate rather than reduce those expenditures. That reallocation may be beneficial, but it does not have a direct effect on the budget deficit.

Notes

¹ Jeffrey A. Miron and Katherine Waldock, “The Budgetary Impact of Ending Drug Prohibition,” Cato Institute white paper, September 27, 2010.

² Since the 2010 report, 30 states and Washington, DC, have decriminalized, medicalized, or fully legalized marijuana. Colorado (2012); Washington (2012); Alaska (2014); Oregon (2014); Washington, DC (2014); California (2016); Maine (2016); Massachusetts (2016); Nevada (2016); and Vermont (2018) have legalized recreational marijuana consumption. Connecticut (2011), Michigan (2012), Maryland (2014), Missouri (2014), Delaware (2015), Illinois (2016), and New Hampshire (2017) have decriminalized marijuana use. Arizona (2010), New Jersey (2010), Minnesota (2014), New York (2014), Utah (2014), Georgia (2015), Louisiana (2015), North Carolina (2015), Texas (2015), Arkansas (2016), Florida (2016), North Dakota (2016), Ohio (2016), and Pennsylvania (2016) have enacted medical marijuana laws.

³ An appendix available on request contains details of all the calculations.

⁴ This report ignores the difference made in the 2010 report between gross and net expenditures, which equals the value of asset seizures and fines collected in the process of prohibition enforcement. As noted in the 2010 report, those amounts are too small to have a meaningful effect on the overall estimates.

⁵ Jason Blevins, “Only 4% of Colorado Tourists Came for the Legal Weed in 2015, Survey Says,” Denver Post, July 20, 2016.

⁶ Angela Dills, Sietse Goffard, and Jeffrey Miron, “Dose of Reality: The Effect of State Marijuana Legalizations,” Cato Institute Policy Analysis no. 799, September 16, 2016.

⁷ Federal taxation would make it harder for states to impose significant taxation without recreating the black market. So even though federal legalization would generate federal revenue, it might imply lower state revenue in some cases.

⁸ Colorado Department of Revenue, “Marijuana Tax Data,” https://​www​.col​orado​.gov/​p​a​c​i​f​i​c​/​r​e​v​e​n​u​e​/​c​o​l​o​r​a​d​o​-​m​a​r​i​j​u​a​n​a​-​t​a​x​-data.