The Cannabis Industry's 16-Million-Pound Phantom: Why Everyone Gets Consumption Wrong

A data-driven investigation into cannabis consumption patterns across seven North American markets


The Problem

If you've read a cannabis market analysis in the past five years, you've probably seen the same magic number repeated over and over: 1.5 grams per day. Industry consultants cite it. Investors model with it. Cultivation applicants build business plans around it.

There's just one problem: it's completely wrong.

When you actually look at government sales data—the real numbers from dispensaries across North America—average cannabis consumption is closer to 0.5-0.7 grams per day of flower. That's about half what the industry assumes.

This isn't a rounding error. At the national level, this overestimate creates approximately 16-20 million pounds of phantom annual demand—cannabis that doesn't exist, won't be consumed, and is driving billions in misallocated capital.

The Data

I analyzed publicly available sales data from seven jurisdictions with transparent reporting:

U.S. States

  • Florida: 929,000 medical patients, 437 million mg THC dispensed
  • Oregon: Mature recreational market, $925 sold annually
  • Michigan: $3.0B annual sales, 835M grams flower
  • Colorado: $1.3B annual sales, mature market since 2014
  • Illinois: $2.0B annual sales, high-tax environment
  • Massachusetts: $1.8B annual sales, low tax-high competition market
  • Washington: $1.1B annual sales, established market

Canada

  • Quebec: State monopoly (SQDC), 50,652 kg sold to 1.26M users

These markets represent fundamentally different regulatory models—from Florida's medical-only program to Quebec's government-controlled retail monopoly to Oregon's wide-open commercial market. Yet they all tell the same story.

What We Found

Finding #1: Consistent Consumption Across Markets

Despite massive differences in regulation, pricing, and market structure, consumption patterns are remarkably consistent:

JurisdictionConsumption (g/day)Market TypePrice/gram
Florida0.51Medical only$6.00
Oregon0.63Mature rec$4.00
Quebec0.55State monopoly$8.00
Michigan1.5Export hub$6.35
Colorado0.58Mature rec$5.50
Illinois0.56High-tax rec$7.80
Washington0.62Established rec$6.30

Range: 0.51-0.63g/day flower consumption

This includes only flower weight. When you add concentrates and edibles on a THC-equivalent basis, total consumption rises to about 1.0 gram per day flower-equivalent—still well below the 1.5 g/day industry standard.

Note: Michigan's higher consumption (0.75 g/day) reflects substantial cross-border sales to neighboring prohibition states (Ohio, Indiana, Wisconsin). Export-adjusted consumption likely matches other states at 0.60-0.65 g/day.

Finding #2: Price Doesn't Matter (Much)

Look at that table again. Oregon's retail prices are around $5.00 per gram with low taxes. Quebec's state monopoly charges $8.00 per gram—60% higher.

Consumption? Virtually identical.

  • Oregon: 0.60 g/day
  • Quebec: 0.55 g/day

This tells us something important: cannabis consumption is primarily driven by biological and behavioral factors, not economic ones. People consume what they consume, regardless of whether it costs $5 or $8 per gram.

This has massive implications for tax policy. States worried that high taxes will drive consumers to the illicit market are missing the point—consumers don't reduce consumption when prices rise. Instead, high prices drive consumers to the black market where they maintain the same consumption at lower prices. This is why Illinois (42% legal capture) has persistent black market activity despite five years of operation, while Oregon (~100% legal capture) has eliminated it.

Finding #3: Regulatory Model Doesn't Matter Either

Quebec operates a government monopoly—all cannabis sold through state-run SQDC stores, limited hours, no private retail, no advertising. It's the polar opposite of Oregon's free-market free-for-all.

Result? Same consumption levels.

  • Quebec: 0.55 g/day
  • Oregon: 0.60 g/day

Whether you're buying from a government bureaucrat or a private dispensary with 50 product options, users consume roughly the same amount. This suggests that consumer behavior is the constant, not the regulatory variable.

Finding #4: Legal Market Share Varies Dramatically

While consumption per user is consistent, the percentage of total cannabis demand captured by legal channels varies enormously—from 21% to over 100%.

To calculate legal market share, we compare actual sales to Total Addressable Market (TAM):

TAM = (Adult Population × Participation Rate × 1.0 g/day × 365 × Pre-tax Price)

Critical point: TAM must use pre-tax pricing because black markets don't collect taxes. Using post-tax prices artificially deflates legal market share by 15-25%.

Example: Illinois

  • Legal sales: $1.72B (2024)
  • TAM: 1.8M users × 1.0 g/day × 365 × $6.25/gram pre-tax = $4.1B
  • Legal market share: 42%

Despite five years of legal adult-use sales, Illinois captures less than half the market due to high taxes (24-37% total burden).

Legal Market Share by Jurisdiction:

JurisdictionLegal ShareKey Factor
Oregon~100%Mature (10+ yrs), low tax, complete black market displacement
Colorado~104%Mature + cross-border sales from neighboring prohibition states
Quebec50%State monopoly, documented in government surveys
Illinois30%High taxes limit legal market capture
Florida21%Medical-only program ($200+ annual fees, by design)

Tax policy matters: The 58 percentage point difference between Oregon's ~100% and Illinois's 30% demonstrates that tax burden is the primary driver of black market persistence in adult-use programs.

Florida's 21% isn't a failure—it's characteristic of medical-only programs, which by design serve 900,000 qualified patients rather than the 2.6 million total cannabis users in the state.

(Full state-by-state legal market share analysis forthcoming)

Why This Matters

1. Cultivation Oversupply

When you think demand is 50% higher than reality, you issue too many cultivation licenses. Michigan and Oregon are exhibit A:

  • Oregon: Unlimited cultivation licensing led to massive oversupply, wholesale collapse
  • Michigan: Similar pattern, wholesale flower at $2.18/gram (production cost floor), retail at $6.35/gram. Michigan's consumption appears 25% higher than other states (0.75 vs 0.60 g/day), likely explained by substantial cross-border sales to neighboring prohibition states (Ohio, Indiana, Wisconsin) where black market prices reach $10-15/gram.

Result: Industry consolidation, bankruptcies, capital destruction

States still planning their markets (looking at you, Texas, Pennsylvania) should take note: use real consumption data, not industry fantasy numbers.

2. Tax Revenue Projections

If you're modeling tax revenue based on 1.5 g/day consumption when reality is 1.0 g/day, you're overestimating revenue by 50%.

Policymakers: when that revenue doesn't materialize, don't blame consumers or the illicit market—blame your spreadsheet.

3. Black Market Displacement

Legal market share analysis reveals that tax policy drives black market persistence, not market maturity or regulatory structure.

  • Low-tax mature markets (Oregon): ~100% legal capture, black market eliminated
  • High-tax markets (Illinois): 42% legal capture, persistent black market despite 5 years of operation
  • Medical-only (Florida): 21% legal capture, by design rather than failure

If a state wants to eliminate the black market, the solution is clear: keep taxes reasonable. Market maturity alone won't fix high-tax problems.

4. Investment Decisions

How much capital has been deployed based on inflated demand assumptions? Billions, easily. Cultivation facilities built for markets that don't exist. Retail chains expanded for customer bases that won't show up.

The phantom 20 million pounds isn't being consumed—it's sitting in vaults, getting destroyed, or sold at a loss.

5. Market Sizing

Every market analysis, every investor deck, every legislative fiscal note uses consumption assumptions as the foundation for total addressable market (TAM) calculations.

If consumption is wrong, everything downstream is wrong:

  • TAM estimates
  • Cultivation capacity planning
  • Retail store density models
  • Tax revenue forecasts
  • Jobs creation projections

The Florida Surprise

Here's something that shocked me: Florida's medical program shows the cleanest consumption data of any market studied.

Why? Because Florida requires:

  • Annual patient registration ($75 state fee)
  • Annual physician recertification ($200-300)
  • Mandatory electronic reporting of all dispensing

This creates a high-quality dataset: we know exactly how many active patients exist (people paying $275-375 annually aren't ghost accounts), and we know exactly what they're consuming (every transaction is tracked).

Result: 0.51 grams per day—the lowest in our sample.

This suggests that many recreational market calculations may be overstating participation rates (counting occasional users as regular consumers), which would push per-user consumption even lower than reported.

What About Concentrates and Edibles?

Sharp readers will note this analysis focuses on flower consumption. What about the vape pens and gummies?

In Florida, which provides the most granular product data:

  • Flower: 55% of total sales
  • Vapes/concentrates: 37% of sales
  • Edibles: 8% of sales

When converted to flower-equivalent (using THC content), total consumption rises to approximately 1.0 gram per day flower-equivalent. Still well below the 1.5 g/day industry standard.

The Biomass Reality Check

Here's a tangible way to understand these numbers. Florida dispensed 437 million mg of THC through vapes and concentrates in our study period.

To produce that requires:

  • 1.84 metric tons of flower (at 75% THC extraction efficiency)
  • About 4,060 pounds of raw plant material

That's the physical reality of 437 million mg. Not an abstract number—actual tons of cannabis that had to be grown, harvested, extracted, and processed.

This kind of conversion helps policymakers and regulators understand cultivation capacity needs in real agricultural terms rather than milligram abstractions.

Methodology Notes

For transparency, here's how we calculated these numbers:

Tier 1 Data (Direct Weight Measurement):

  • Florida: OMMU weekly reports, actual mg dispensed ÷ active patients ÷ 365
  • Oregon: OLCC monthly reports, actual grams sold ÷ estimated users ÷ 365
  • Quebec: SQDC annual report, kg sold ÷ surveyed users ÷ 365

Tier 1.5 Data (Direct Weight with Export Adjustment):

  • Michigan: CRA monthly reports, actual pounds sold ÷ estimated users ÷ 365, with 60% export adjustment for cross-border sales to OH/IN/WI prohibition states

Tier 2 Data (Revenue-Based Calculation):

User Estimates:

  • Medical markets: Actual registered patient counts
  • Recreational markets: 15-20% of adult population (NSDUH national data, Quebec provincial survey)

Legal Market Share Calculation:

  • TAM = Adult population × Participation rate × 1.0 g/day × 365 × Pre-tax price
  • Legal market share = Legal sales ÷ TAM
  • Uses pre-tax pricing to avoid 15-25% systematic underestimation

All data sources are publicly available government reports. No proprietary data, no industry-funded surveys, no self-reported consumption estimates.

Where the 1.5 g/day Number Came From

I genuinely don't know.

It appears in industry reports without attribution. When you trace it back, the methodology disappears. It's one of those numbers that gets cited so often it becomes accepted truth.

My best guess: it's probably mixing flower weight with total product weight, or using cultivation input rather than consumer output, or counting occasional users in the denominator.

But here's the thing: it doesn't matter where it came from. What matters is it's wrong, and it's causing real economic damage.

What Should States Do?

If you're a policymaker designing a cannabis program:

  1. 1. Use 0.6-0.7 g/day flower consumption for market modeling (about 1.0 g/day flower-equivalent including concentrates)
  2. 2. Calculate cultivation capacity based on in-state population × participation rate × consumption rate—not industry consultant projections
  3. Keep taxes reasonable if you want to displace the black market. The 58 percentage point difference between Oregon (~100% legal capture) and Illinois (42% legal capture) demonstrates tax policy is the primary driver of black market persistence.
  4. Don't panic about price elasticity—consumption is relatively inelastic to price changes
  5. Plan for consistency—your market won't be radically different from Oregon, Michigan, or Quebec regardless of regulatory model

6. Require transparent reporting—Florida's mandatory tracking creates better data than any other state

The Bottom Line

Cannabis consumption across legal North American markets is 0.5-0.7 grams per day of flower, translating to roughly 1.0 gram per day flower-equivalent when concentrates and edibles are included.

This is consistent across:

  • Medical and recreational markets
  • State monopolies and private retail
  • High-tax and low-tax jurisdictions
  • Mature and emerging markets
  • $4-8/gram retail price points

However, legal market share varies dramatically (21-100%), driven primarily by tax policy rather than market maturity or regulatory structure.

The widely-cited industry estimate of 1.5 g/day overstates consumption by approximately 50%, creating roughly 16-20 million pounds of phantom annual demand at the national level.

This phantom demand is driving:

  • Cultivation oversupply
  • Capital misallocation
  • Inflated tax revenue projections
  • Unrealistic market size estimates

Time to update the spreadsheets.


Data Availability

All data and calculations are available through Harvard Dataverse.

Includes:

  • State-by-state consumption calculations
  • Legal market share analysis by jurisdiction
  • Methodology documentation
  • Source URLs for all government reports
  • Price data and market structure variables

About This Research

This analysis is part of ongoing cannabis policy research published at dankreports.com. This research uses only publicly available government data. No industry funding, no survey self-reports, no proprietary datasets. Just actual sales numbers from actual dispensaries.

The empirical methodology and cross-market validation framework supporting this analysis are published as an academic working paper: "Cannabis Consumption in North American Legal Markets: Evidence from Seven Jurisdictions Across Regulatory Structures" available at SSRN: https://papers.ssrn.com/abstract=5889482


This research uses only publicly available government data. No industry funding, no survey self-reports, no proprietary datasets. Just actual sales numbers from actual dispensaries.

Read more