MORE Act revenue compared to state marijuana revenue

States with marijuana markets were collecting about $43 per man, woman, and child in marijuana excise taxes a couple of years ago.  Since then, that number has gone up.

The comparable number for the federal MORE Act, legalizing marijuana,is $2.15.

is CBO says the federal MORE Act, legalizing marijuana, would reduce the budget deficit by $1.258 billion in its most productive year, fiscal 2030.  Total tax increases amount to $2.256 billion (an ad valorem excise tax would collect $.762 billion; other revenues not from marijuana excises  – from income taxes, FICA, and the like, would go up $1.494 billion).  Spending, mainly on marijuana-related trust funds, would go up by $.998 billion.

So with a projected U.S. population of 355,100,000 in 2030, the gross tax increases are $6.35 per capita, and the deficit help is $3.54 per capita. The excise tax is $2.15.

Are there reiiable state revenue numbers for the increase in revenues not frommarijuana excise taxes?

MORE Act budget numbers, accepted

Earlier today, I wondered if the budget estimate for the MORE Act reflected enough savings.  Center for New Revenue Board Member Doug Berman, Professor at the Ohio State University Law School, explained on Twitter how the estimate could be right.

“Cost per inmate is often quoted much higher when total Bureau of Prison costs are divided by total number of inmates, but the “real” cost of one more or one less inmate is lower because removing, say, 50 inmates will not justify shutting a facility and reducing staff.

“Federal inmate population has dropped 60,000+ over last 8 years, and yet BOP budget has stayed stable … which means cost per prisoner calculation for whole system has increased and shows how reduction in prisoners does not perfectly convert to immediate cost savings.” 

And my friend Dr. Keith Humphreys of Stanford tweeted, “Given the strength of correctional unions, it’s not clear that releasing 10,000 inmates would either.”

I figured the average cost of housing an inmate would approximate the marginal cost of freeing one. But that figuring was not thought through.

MORE Act budget numbers, questioned

UNC Law Professor Emeritus Bill Turnier points out that the Federal Register indicates that “the fee to cover the average cost of incarceration for Federal inmates was $34,704.12 ($94.82 per day) in FY 2016 and $36,299.25 ($99.45 per day) in FY 2017.”

The official budget estimate for the MORE Act (posted here yesterday projects) savings from less incarceration:  “reducing both the number of inmates in federal facilities and the aggregate time they serve would result in net savings of about $1 billion over the 2021-2030 period.”  “CBO estimates that H.R. 3884 would reduce time served by 73,000 person-years, among existing and future inmates.”  Calculating, that’s a cost of about $13,700 per person year.

So the actual savings using the Federal Register numbers would seem to be nearly three times the $1 billion stated. Usually, CBO is pretty thorough about this kind of budget estimate, so Professor Turnier and I are curious here.

Official MORE Act budget numbers

The last few posts here have been about the MORE Act, legalizing marijuana, and taxes. Here and below are official budget numbers for the MORE Act, which passed the House yesterday:

A quick tax, taxes first:

“CBO and the staff of the Joint Committee on Taxation estimate that H.R. 3884 would increase revenues, on net, by about $13.7 billion over the 2021-2030 period by creating business income, compliance, and occupational taxes; those increases would be partially offset by allowing certain deductions for business expenses associated with trafficking controlled substances.”

That $13.7 billion comes from two sources:  an ad valorem excise tax, picking up $5.677 billion, and general revenues like income taxes that come from opening up the marijuana market to legality and the light of day, and picking up $8.013 billion.  The latter number would be greater if 280E, disallowing those business expense deduction, were in place.  A 2017 estimate of 280E repeal was that it lost $5 billion over 10 years – for a 10-year period when legalization was less widespread.  That number should be higher than $5 billion now. Keeping 280E (or reinstituting it) and foregoing the excise tax might yield more revenue.

The bill would reduce the deficit by about $ 1 billion over 5 years, and by about $7 billion over 10 years.  Most of that impact comes from the $8 billion in “other” taxes like post-280E income taxes; the excise taxes are spent, via trust fund operations.

Other stuff:

Other budget impacts include savings from less incarceration:  “reducing both the number of inmates in federal facilities and the aggregate time they serve would result in net savings of about $1 billion over the 2021-2030 period.”  “CBO estimates that H.R. 3884 would reduce time served by 73,000 person-years, among existing and future inmates.”  Calculating, that’s a cost of about $13,700 per person year.  I thought incarceration cost more than that.  Anyway, prisoners aren’t eligible for federal programs; lessening the prison population and freeing folks up for programs is estimated to cost $636 over the period.

MORE Act revenue effect — superseded

So the MORE Act legalizing marijuana just passed the House.  It loses revenue, by eliminating the 280E tax on advertising and marketing (which are now non-deductible, but would become deductible under the Act).  It gains revenue, by imposing a 5% (rising to 8% in year 5) ad valorem grower tax (converted to THC and product weight after year 5).  

But what is the net revenue effect?  Is the Act a net gainer or loser?  I haven’t seen.

Update: Here are the numbers, issued December 4:

Tax questions about the MORE Act legalizing marijuana

1.  The tax rate now inches up over time, landing at 8 percent ad valorem in year 5.  The tax then switches from ad valorem to weight and THC after year 5, but still works with an ANNUAL look-back to price.  If you keep looking back, why bother switching?  This switch mimics the old Blumenauer–Wyden bill switch to an annual lookback, which I thought so weird that I couldn’t believe the old bill did it, so I misread the old bill. Isn’t the main point of switching to weight and THC to keep the tax burden steady, rather than fluctuating with price?

2.  280E repeal effective date:  Say date of enactment is July 1, and marijuana is descheduled then.  Taxpayer is on a calendar year, or say a June 1 fiscal year.  280E is the law for part of the taxpayer’s tax year.  What happens?  There may be an easy answer.

3. The new ad valorem tax is effective “after 180 days after the date of the enactment of this Act.”  Are no sales contemplated before those 180 days expire?

New tax provisions in MORE Act — some improvement

The MORE Act, legalizing marijuana, has been amended before a House vote this week, with a little improvement in the tax provisions.  Here is the new text:

Here are some quick reactions:

1.  The tax rate inches up over time, instead of being frozen at a permanently low rate.  This increase follows the old Blumenauer–Wyden bill,, but with lower rates, capped at 8 percent ad valorem, instead of 25 percent in Blumenauer–Wyden.  The tax then switches from ad valorem to weight after year five, but still works with a look-back to price.  If you keep looking back, why bother?  That’s a switch that makes no sense to me. But increasing rates have been suggested since at least RAND-Vermont,

2.  The tax kicks in right away, as near as I can tell, instead of being delayed.

Old MORE Act section 4(c) :

Effective Date.—(1) IN GENERAL.—Except as otherwise provided in this subsection, the amendments made by this section shall apply to articles manufactured or imported in calendar quarters beginning more than one year after the date of the enactment of this Act.


(f) Effective date.— 

(1) In general.— Except as otherwise provided in this subsection, the amendments made by this section shall apply to removals, and applications for permits under section 5922 of the Internal Revenue Code of 1986 (as added by subsection (b)), after 180 days after the date of the enactment of this Act. 

The bill doesn’t necessarily contemplate sales before 180 days after the date of the enactment of this Act.  I’m not sure, but it’s an improvement over the one-year-plus tax holiday in the original MORE Act.


Most of the criticism of the original bill here ( still obtains. For instance, the Act still favors advertising and disfavors Mom & Pops by repeal of 280E, and it still requires transfer pricing adjustments, but there has been a little improvement.  So good.


On a non-tax issue, I think this is new:

Packages.— All cannabis products shall, before removal, be put up in such packages as the Secretary shall by regulation prescribe. 

Calaveras’s smart marijuana tax move?

On Election Day, Calaveras County, California, reportedly repealed its weight-based tax on marijuana and replaced it with a square footage tax.

Now inferior jurisdictions’ taxes on production make no sense if the consumption level is the wellspring of any negative externalities (DUI, kids) that might justify taxation, so that taxing production would only drive production to lower-tax jurisdictions, competing for industry.  (So states tax cigarette consumption by the pound, but place no special taxes on growing or manufacture, excise or otherwise.)

But for production, some say there is a negative externality, odor, which is so difficult to measure that proxy taxes are necessary.  Doesn’t Calvaras’s shift from weight-based tax to area create a better proxy for smell?

Mj podcast, and questions

Duke Law School Professor Kim Krawiec and I talk marijuana legalization on a 62- minute podcast.

After we finished, she forwarded some questions from her students.  Here are the questions, and some quick reactions.

— Is it true that the tax revenue from legalized marijuana has gone towards improving public schools in states like Colorado?

Continue reading “Mj podcast, and questions”

Kevin Sabet’s Decrim Bill

Kevin Sabet, America’s most prominent marijuana prohibitionist, suggests a decriminalization option – possession of a very small amount would not be a crime, but it would involve penalties.  His plan is included in a bill in New Jersey.

The penalties seem to be $150 for the first offense, $200 for the second, $500 for the third — the first penalty waivable upon completion of “a substance abuse assessment by a professional.” 

Here’s the bill:

2018 Kevin Sabet Rice bill 1926_I2

Illegal weed market? It’s the impunity, stupid.

It’s the impunity, stupid.

The marijuana industry says taxes keep the illegal market alive.  At the margin, maybe taxes drive some transactions away from legal sellers.  

But analogy to illegal logging in the rain forest of the Brazilian Amazon offers another answer, explained by the Wall Street Journal.  

The success of legal, licensed loggers, operating “concessions” with constraints designed for sustainability, “depends on the government’s ability to crack down on illegal logging.  Since they pay no taxes and make no effort to protect certain species or invest in restoration, illegal loggers can charge $431 per square meter of lumber, compared with $1,511 per square meter of legally logged timber, concession operators said.  ‘It is like having a regular, taxpaying shop competing with lots of tax-free peddlers right in front of your door,’ said Jonas Perutti, owner of Lumbering Industrial Madeflona Ltda., which also operates concessions in the Amazon.”

. . . 

“Illegal logging is thriving in part because the Bolsonaro administration has cut environmental protection budgets. Here in Rondônia, a state the size of Michigan, there is only about one patrol agent from the Ibama environmental agency per 540 square miles. ‘What makes these loggers continue in the forest? It’s the certainty of impunity,’ said a government official.” 

California’s illegal market in retail marijuana is not due to taxes, whose burden is lower than that in Washington and maybe comparable to that in Colorado.  A lack of licensesand an overhang of supply from entrenched growers are not helping.  But a key factor is the open and flagrant operation of illegal storefronts, whose operators keep dodging what law enforcement there is. Legitimate sellers are frustrated by lack of enforcement.

Look, you’ll never eliminate the illegal market.  You’ll only marginalize it.  The optimal amount of crime is not zero.  But a knee-jerk blaming of taxes doesn’t complete the analysis.

How much revenue can law enforcement bring?  Here’s an old take:

“The President’s budget for fiscal year 1991, for example, proposes adding 3,600 staff to examine more tax returns, collect more unpaid taxes, and do other tax enforcement jobs. IRS estimated that with this additional staff, it would generate about $500 million of additional revenue in 1991 and about $6.5 billion by the end of fiscal year 1995.” (GAO report questioning those numbers and suggesting other methodologies).

That old take gives little about quantification today — how much enforcement could help with marijuana tax collections, or with protecting the Amazonian rain forest.  But enforcement should not be overlooked.

Growers pay federal cannabis tax under the MORE Act

Who would pay the federal cannabis tax under the proposed MORE Act?  I was talking today with a sophisticated private sector marijuana guy, who thought it was a retail tax. But no. It’s a grower tax.

The MORE Act says:

(1) IN GENERAL.—Section 5701 of the Internal Revenue Code of 1986 is amended by redesignating subsection (h) as subsection (i) and by inserting after subsection (g) the following new subsection:

“(h) CANNABIS PRODUCTS.—On cannabis products, manufactured in or imported into the United States, there shall be imposed a tax equal to 5 percent of the price for which sold.”.



“(1) IN GENERAL.—Any person who plants, cultivates, harvests, produces, manufactures, compounds, converts, processes, prepares, or packages any cannabis product shall be treated as a manufacturer of cannabis products (and as manufacturing such cannabis product).


Section 5703 identifies the taxpayer under 5701 as amended. It’s the manufacturer. Since the MORE Act says the grower is a manufacturer, the first one in the supply chain, the grower pays the new tax.

(a)Liability for tax

(1)Original liability

The manufacturer or importer of tobacco products and cigarette papers and tubes shall be liable for the taxes imposed thereon by section 5701.

Auctioning marijuana licenses — for a share of future revenue

Lotteries and “on the merits” selection of marijuana license holders are not working well. A 2015 RAND report, Considering Marijuana Legalization: Insights for Vermont and Other Jurisdictions, suggested auctions.  

But there’s a killer problem:  “Auctions ordinarily favor well-capitalized bidders (e.g., existing businesses), which can pay in all events, without depending on results. This favors favor bidders with wealth—with risk capital. So auctions are regressive in their allocation of a state-owned intangible asset.”

That problem could be addressed by having bidders bid not in dollar amounts, but in percentage of revenue they will pay as a license fee – an ad valorem tax by another name.  The fee would be paid after a year’s business is done, so undercapitalized and social equity licensees would not have to come up with money up front.

India is implementing this kind of “revenue share” auction for state-owned coal assets::

“Coal mines will be put for auction on the basis of revenue share instead of the regime of fixed rupee/tonne.

“The Cabinet Committee on Economic Affairs, chaired by Prime Minister Narendra Modi, has approved the methodology, the Coal Ministry said in a release.

“‘This methodology provides that bid parameter will be revenue share. The bidders would be required to bid for a percentage share of revenue payable to the Government. The floor price shall be 4 per cent of the revenue share.

“”Bids would be accepted in multiples of 0.5 per cent of the revenue share till the percentage of revenue share is up to 10 per cent and thereafter bids would be accepted in multiples of 0.25 per cent of the revenue share,’ the release said.”

For cannabis licenses, a lot of thinking would have to be done.  Maybe government would allocate licenses to the high bidders in percentage terms.  No one would be disqualified, but government could decide when to stop issuing licenses on the basis of predicting how much various bidders could in fact sell.

Here are excerpts from RAND-Vermont:

Bidding for permanent licenses is likely, from the state’s perspective, to leave money on the table. As the nascent marijuana industry struggles to outperform the black market, entrants are unlikely to amass and risk the capital necessary to pay for the present value of future rents from a permanent marijuana license. That is, given uncertainty, firms would deeply discount the prospective profits from the out-years. Taxation can make up for that shortcoming by taking up economic rents as they materialize. But proceeds of a one-time auction for permanent rights would bring in a disproportionate amount of revenue up front and could help pay for the cost of setting up legalization.

Instead of permanent licenses, annual licenses could be auctioned. (Auctions could sell off licenses for any number of years; we consider one year as an example.) A series of annual auctions might, like nimble taxes, yield increasing revenue over time. As the cost of producing marijuana drops, the privilege of selling it legally in a restricted market increases in value. With annual auctions, the state might take at least some of the increasing value of that privilege. If so, winners’ payments to the state would increase their costs of production, which might help prevent an unfortunate price collapse. 

The state could set a year’s quantity target for quotas based on prior volume (or, for the first year, on an estimate of likely first-year consumer demand). Several types of auctions are available (Klemperer, 1999). If the state seeks to avoid market concentration, it could limit producers to a certain percentage of the total quota. Related-party rules would be needed to enforce that kind of limitation. This gets back to the issues surrounding supply architectures in Chapter Four. 

The winner of an annual auction has no guarantee of renewal. That uncertainty might keep bids low and might keep potential entrants out of the market. Indoor marijuana production, in particular, would seem to be benefit from long-term investment. Typical investments in real property and in plant and equipment with a useful life beyond one year would be thrown into turmoil by the prospect of nonrenewal. Outdoor growers, whose operations are less capital-intensive, might be less daunted by the risk of nonrenewal. Annual licenses, perhaps more than any other allocation mechanism, could work against incumbent marijuana businesses. That would be a positive feature if government wants to disfavor entrenched incumbency in the marijuana industry. As Caulkins, Hawken, Kilmer, and Kleiman (2012) noted, “[O]nce business interests get entrenched in an industry, toughening the laws related to that industry becomes more difficult. . . . Private interests prioritize profit, not public health or public safety” (p. 245). 

2020 Initiative Marijuana Tax Ranking — Arizona Comes in Last

Who wins the marijuana tax booby prize this year?

In 2012, State Tax Notes published a ranking of marijuana tax proposals, with Washington and Colorado, which were enacted, ranked 1 and 2, ahead of Oregon, which lost – the booby prize winner of the proposals that made it to the ballot (but fourth out of six total).

About Oregon’s proposal, I wrote, “The only hope for Measure 80 is that Oregon law allows the legislature to fix it.”  Measure 80 failed, but Oregon came back with a better plan in 2014, which passed.

About Washington’s I-502, I wrote, “It’s not just this year’s best. It’s the strongest marijuana revenue plan American voters have ever seen.”

No such luck this year. 

This year, four proposals are on the table:  Arizona, Montana, New Jersey, and South Dakota.

All impose only primitive ad valorem (percentage of price) taxes.  All will see tax per unit of THC decline, as prices decline over time. (Revenue will go up so long as sales increase in dollar terms, but price taxes leave money on the table.) New Jersey’s seems weakest on the surface, with the tax at only the standard state sales tax rate, while the others vary between 15 and 20 percent.  But New Jersey contemplates new taxes, to be added by the Legislature, so it’s hard to rank low.

Little variations in rates aren’t much to hang your hat on.  This year’s weakest structure is in Arizona, which ties the hands of future Legislatures at rates below Washington’s black market-beating 43.5 percent (37 percent excise plus 6.5 percent standard sales tax)  —  plus local taxes.

Here’s some of the Arizona proposal:



Not only that, the proposal caps future taxes in light of possible federal legalization:


Looks like industry prevailed in Arizona.  It’s not clear how the Arizona tax cap would work if the federal marijuana tax is based on weight or THC.  

The other three states don’t have tax structures to brag about, but they leave Arizona in the desert dust. Maybe Arizona is desperate enough for legalization that this flimsy tax structure will bind it for years.

Problems with the MORE Act

UPDATE 10 February 2021:

And in a sign of both the maturation of cannabis as a serious business—and in a clear indication of how the cannabis industry is using the same tactics employed by Silicon Valley, alcohol, and other big business sectors to secure favorable regulation—Altria is also covering its bases politically, at both the state and federal levels.

“To woo members of Congress, Altria in 2020 hired Denver-based Brownstein Hyatt Farber Shreck, one of the nation’s top cannabis and hemp law firms, on policies related to CBD and “non-tobacco excise taxes,” recent federal disclosure filings show.”


The MORE Act legalizing marijuana is supposed to be voted on by the U.S. House in the lame duck session. It has lots of issues, political and substantive.

Three non-tax issues, first; then three tax issues.

Without nationwide testing and packaging standards, states will race to the bottom, as they have with hemp, and whichever states have the loosest standards will see their industry flourish.  Some states don’t want to be in that race to the bottom.

2. Interstate commerce will will wipe out growers in most states other than California and Oregon.  The once-prominent STATES Act would go the other way, and let states protect growers from imports.  Interstate commerce is a killer defect for some growers. Maybe all, if low-cost imports from South America come in.   

3. interstate commerce will bring a quick end to experimentation with regulatory models, experimentation we desperately need.  Even Canada is struggling with getting the details right.  Take taxation, for instance.  There is nowhere near a consensus on what to tax (the tax base:  weight? THC? ad valorem?), let alone where to collect and how much to tax. I.  280E repeal

Now some tax issues:

Continue reading “Problems with the MORE Act”

Marijuana Grow Quotas

At the moment when marijuana is legalized, demand for growing licenses or quotas will exceed supply, right?

At some point, someone will have to say who gets them.  Who will say?  And how?  

And once the decider decides, what appeals will be allowed?  If appeals are allowed, to whom?


FDR’s wealth sharing in the tobacco program said government agents decided, and no appeals were allowed.

It’s not a model to follow, but a proof of concept. 

Allocating grow quotas on the basis of claimed prior acreage might not withstand the modern version of Due Process, and anyway, we don’t have a class to favor on that basis..

“Prosperity Road: the New Deal, tobacco, and North Carolina,” by Anthony J. Badger, describes the process.


Late 1933 into 1934:

Growers had to vote the program in every year.  It based quotas on a rolling average production history. County agents took farmers’ reports of historic production of tobacco.

So the government’s agents had to evaluate whether claims of historic growing were exaggerated.

“[M]ost growers overestimated.” P91 County agents negotiated and adjusted amounts allowed. “Few were able to bring the production records that might have solved the problem.” 91

One practice, by at least one agent, was to list allotments publicly in local stores “and inviting neighbors secretly to inform him in their neighbors were overestimating.”

Once quotas were allocated, “It was a troublesome and time consuming task to measure some 500,000 fields in the cotton and tobacco counties of North Carolina. In a tobacco county like Nash or Pitt as many as ninety compliance supervisors were employed.” 92


Production during the years 1931-33 was to be the base for tobacco allotments.

Nonsigning farmers got no payments, and paid a tax of 25 percent on their tobacco sales. That rate was chosen by the U.S. Secretary of Agriculture. p 73


Digital technology ought to facilitate sharing the wealth — instead of among farmers, among registered voters, say.

NC Senate candidate’s marijuana-biz texting partner irritates California activist

Being in the marijuana revenue policy business, I subscribe to a California marijuana newsletter put out by Brett Stone, an activist who has provided me (and others) a lot of useful information over the years.

Here’s an opinion piece he just posted about the North Carolina Senate race, with the title “DPFCA_NEWS: California cannabis strategist Arlene Guzman Todd caught up in NC Democratic Senate candidate sexting and sexual relationship scandal – claims ‘He knows (that I) can tank his campaign.’”:

“Passing along an item that might be of interest to some readers on the list. News reports are that a LA based cannabis marketing and public relations executive, Arlene Guzman Todd who works for Potnt (website @ ) where she is listed as a management team member is involved in a sexting and sexual relationship scandal with the Democratic candidate for US Senator in North Carolina. She works with a few well known cannabis businesses. Their clients include HelloMD and Jim McAlpine of 420 Games.

“I wonder how a cannabis business can continue to do business with Potnt knowing that one of their employees, Arlene Guzman Todd, is threatening to take down the campaign of the democratic challenger for US Senator in North Carolina, one of the tightest and most watched races currently with the democrats having a real shot at replacing a republican senator unless this scandal creates problems so close to the election. Has anyone ever heard of this woman or done business with her? 

“Story is making a lot of noise in North Carolina but not a peep here in California or Los Angeles where she is based and lives so I put together a few articles for those who may be interested in reading about it. IMO she is a snake in the grass, no pun intended. 

“Some of the articles have been updated with some headlines changed but the information is still there. They say the greatest disinfectant is sunlight so here’s some sunshine for you Ms. Todd.”

[He attached easy-to-find articles from News & Observer, and AP.]

But I wonder if a boycott might not drive her to sabotage the campaign (as she has reportedly threatened to do).  That is, if supporters of the Democratic candidate in North Carolina go against her, she might retaliate.  Just wondering – I have no idea.   But then how could she retaliate and harm Cunningham further? Posting photos? That would make her seem small, I think. She can hardly retaliate against Cunningham at this point, I think.

A typical boycott aims to change the behavior of the boycottee going forward. What change in Ms. Todd’s behavior might this boycott encourage? I’m not sure what she could do at this point to make it right.