cannabis: $75b opportunity; category …...source: state reports, nsduh and cowen and company...

Equity ResearchCannabis

Alcoholic Beverages

CANNABIS: $75B OPPORTUNITY; CATEGORY CROSS-CURRENTS KEEP US CAUTIOUS ON BOOZEAPRIL 4, 2018

We expect cannabis can generate $75 bn in sales by 2030 (assuming federal legalization), driven primarily by consumers ages 26+ (given that 55+ is the fastest growing cannabis cohort).

Access to legal cannabis drives lower levels of binge drinking (as measured by our proprietary analysis on binge intensity and number of binges per month).

Adult-use cannabis states binge drink 13% fewer times per month than non-cannabis states, and 9% below the national average.

We remain constructive on our Outperform-rated cannabis stocks: KSHB, WEED and LEAF, and cautious on big beer (TAP).

Vivien Azer 646 562 1351 [email protected]

Brian Nicholas Velez 646 562 1353 [email protected]

Gerald Pascarelli, CFA 646 562 1362 [email protected] COWENRESEARCH COWEN.COM

This page left blank intentionally.

COWEN.COM2

COWENEQUITY RESEARCH April 4, 2018

EQUITY RESEARCH MULTI-SECTOR

April 4, 2018

■ Alcoholic Beverages■ Cannabis

$75B CANNABIS OPPORTUNITY KEEPS USCAUTIOUS ON BOOZE - AHEAD OF THE CURVE

Vivien Azer646 562 [email protected]

Brian Nicholas Velez646 562 [email protected]

Gerald Pascarelli, CFA646 562 [email protected]

THE COWEN INSIGHT

We expect cannabis sales of ~$75 bn by 2030, supported by new analysis on categoryretention gaps and population-weighted, state-based analysis on cannabis first use andbinge drinking (where cannabis access points to more cannabis trial and use, and less bingedrinking, and an opioid solution). Maintain Outperform on KSHB, WEED, & LEAF; and remaincautious on big beer, including TAP.

Bottom Line

Ahead of our 2nd Annual Cannabis Summit, we revise up our cannabis market target by $25bn, as we now look for the industry to generate $75 bn in sales by 2030 (vs. $50 bn by 2026previously). Underpinning this estimate is proprietary analysis on binge drinking behavior,as well as the role that legal cannabis access has in terms of driving trial (measured bycannabis first use rates for 18+ consumers). This work builds on our prior assertions thatcannabis acts as a substitute social lubricant for consumers. We believe this is the first timedetailed state-level binge drinking statistics have been analyzed and juxtaposed againstcannabis use, where we found that legal cannabis states (as of 2016) binge drink 13% fewertimes per month than non-cannabis states.

COWEN.COMPlease see pages 59 to 62 of this report for important disclosures.3

What's New

We now expect the category to generate ~$75 billion in gross sales by 2030 (up from $50bn by 2026). This new forecast reflects:

■ Binge Drinking Intensity Variances. States for which adult use cannabis is available (as of2016) already had binge drinking intensity rates that fall ~9% below the national average,and ~11% below non-cannabis states. Newly added states (like CA, NV) currently havehigher rates of binge drinking intensity, and lower levels of cannabis consumption, thoughwe expect mean reversion for these states too, given the historical precedent. As such,we believe it’s reasonable to assume that as more legal cannabis states come on-line,alcohol binge drinking intensity should fall.

■ Binge Drinking Volumes: Binges Per Month. While the CDC estimates ~17% of the U.S.population reports binge drinking, only 1 in 6 report doing so 4+ times per month. In legaladult use cannabis states, the number binge drinking sessions per month (for states legalthrough 2016) was ~9% below the national average.

■ Category Spending. While we had forecast a 2026 total cannabis market of $50 bn by2026, new forecasts suggest that the market is already that size. As such, we are revisingour forecasting methodology, including a higher per capita spending estimate. With areported 35 mm annual cannabis consumers, current estimates imply annual spendingof $1,500 per year (well ahead of our initial $1,000 forecast). This likely reflects theresponse bias in the number of cannabis users (given its Schedule I status). As such, weraise our estimate by 25% as an offset.

■ Category Retention. Lots of consumers have tried both alcohol and cannabis in theirlifetime: ~209 mm for alcohol and ~115 mm for cannabis (as of 2016, according togovernment forecasts, for which cannabis is subject to a response bias as a Schedule Inarcotic). But, while consumers come back to alcohol at consistently high rates, the sameis not true for cannabis (at least as reported). However, the gap has been closing overthe last decade and we expect that trend to persist. And, despite the convergence inconsumer consumption, our ~$75 bn estimate still puts cannabis category retention ~34pts behind alcohol on a past year basis.

■ First Use Cannabis Benchmarking. Our new analysis introduces cannabis “first use”data (the % of 18+ consumers who have tried cannabis in the last year). To be clear, thisdoes not reinforce the notion that cannabis is a gateway drug; we address that in the“Unintended Consequences” section of the note. But, among consumers over the ageof 18, in legal cannabis states, they show a greater propensity for cannabis trial, whichultimately looks to translate to higher levels of past month cannabis incidence.

■ Expanded Opioid Research. According to a November 2017 White House report,the opioid epidemic costs our country ~$500 bn per year (and that cost is growingexponentially). Newly published research reinforces the work that we have already done,showing that for some, cannabis is an effective opioid substitute (in particular in treatingchronic pain).

COWEN.COM4


Stock Implications

We remain Outperform rated on the three cannabis companies under our coverage: KSHB,WEED, and LEAF. While this report focuses on the U.S. and has direct implications forKSHB, the trends that we've historically seen for cannabis in the U.S. tend to be mirrored inCanada, building on our conviction around the cannabis opportunity for WEED and LEAF asthe Canadian market expands to adult use later in 2018. Meanwhile, our extensive analysison the impact that legal cannabis access can have on overall drinking patterns reinforcesour conviction that cannabis and alcohol are substitute products. We believe that lower-end beer is most exposed to this substitution risk, such that we remain cautious on MarketPerform-rated TAP.

What to Watch

Growing consumer support for legal cannabis, and increased access, looks to be drivingmeaningful change at the state level. We expect to see ballot initiatives to legalize cannabisfor adult use in both Michigan and Illinois in the upcoming November election. As well, anumber of states are looking at expanding cannabis access (both medical and adult use) viathe legislative and executive branches (with New Jersey the most likely in 2019). Expandedaccess to Maine, Massachusetts and New Jersey could prompt other states in the Northeastto make changes.

COWEN.COM 5



COWEN.COM6


Cannabis: $75B Opportunity; Category Cross-Currents Keep Us Cautious On Booze

Next week, we head to Los Angeles for Cowen’s 2nd Annual Cannabis Summit. We

launched on the sector with a collaborative analysis spanning 10 Cowen Analysts in

September 2016 (here). A lot has changed since then.

18 months ago, we estimated that the legal cannabis market would generate $50 bn in

legal sales by 2026. Today, we think the direct and indirect revenue opportunities (and

risks) are far greater. Assuming federal legalization, we believe cannabis can generate

gross sales of ~$75 bn by 2030 (and ~$17.5 bn in tax revenue), though still far less than

the over $210 bn in sales that the alcohol industry generates today.

Our new proprietary analysis supports our view that cannabis is a substitute social

lubricant for alcohol, and will likely continue to present an incremental headwind to

alcohol sales, in particular beer. Our revised outlook reflects the following:

1. Legal Bud Weighing on Binge Drinking. Herein, we build on our work around alcohol

and cannabis substitution by introducing unique analysis around the evolution of

binge drinking trends across the U.S., segmented by cannabis states. We found that

states with access to adult use cannabis binge drink ~5% less frequently per month

than medical cannabis states (using 2016 legal status) and ~13% less frequently

than non-cannabis states. As cannabis access expands, we expect further pressure

on alcohol sales, given this notable divide in consumer consumption patterns.

2. Extended Outlook: $75 bn Opportunity by 2030. We extend our outlook for the

cannabis market to 2030, while also revising our underlying methodology to reflect

both the market’s evolution and a better understanding of those underlying drivers.

In addition to refining our user evolution analysis, we also adjust our spending

estimates to account for the response bias in government user numbers. Initially,

we had estimated per capita annual spend of ~$1,000 (with no meaningful increase

in the next decade). However, with the market now estimated to be $50 mm, today

this calls into question that rate of spend, given the estimated 35 mm consumers in

the category. As such, we revise up our targeted spend by ~25%, which partially

mitigates this disparity, as we continue to believe that underlying spending is likely

~$1,000 per year.

Figure 1 Per Capita Spend Estimates Reflect a Notable Response Bias Among Consumers

Source: State Reports, NSDUH and Cowen and Company

COWEN.COM 7


https://cowen.bluematrix.com/docs/pdf/d5e219f6-a2d1-41e3-862d-9e8e23d74c54.pdf

Stock Implications

This analysis focuses on the U.S. cannabis market, and thus has implications for

Outperform rated Kush Bottles (KSHB), as well as mixed implications for our publicly

traded alcohol companies, as we continue to build our body of work showing that

alcohol and cannabis are substitute social lubricants.

As well, while we do not cover the Canadian cannabis market in this note, we have

found in our previously published research, that consumer trends in the U.S. and Canada

largely mirror one another when it comes to both cannabis category engagement as

well as the interaction between alcohol and cannabis (in particular among younger adult

consumers). As such, we also reiterate our Outperform ratings on Canadian cannabis

stocks Canopy Growth (WEED) and MedReleaf (LEAF).

Lastly, while we broadly view cannabis and alcohol as substitute social lubricants, we

believe our Top Pick Constellation Brands (STZ) is best positioned to weather this

evolving consumer landscape, as their alcohol portfolio is the most defensible from

cannabis substitution (given the company’s outsized exposure to more affluent, multi-

cultural and gender-diverse consumers), while also having a notable first mover

advantage as the first (and thus far only) publicly traded alcohol company with exposure

to the legal cannabis market (via their 10% stake in medical market share leader WEED

in Canada).

Figure 2 Alcoholic Beverage Stock Implications

Ticker Rating Implications Reason

STZ Outperform Constructive

Positive: First (and thus far only) alcohol company with legal cannabis exposure

Positive: Defensible

Positive: Continues to consolidate share of total beer

Negative: More responsible drinking behavior

BREW Outperform Contructive

Positive: ABI qualified offer optionality

Positive: Diversification with Kona

Positive: Growing in home markets

Negative: Overindexes to West Coast

Negative: Cannabis substitution risk


BF/B Market Perform Neutral

Positive: Over-indexes to premium bourbon, which continues to gain share



TAP Market Perform Cautious

Positive: High degree of confidence in ability to generate targeted cost savings

Negative: Cannabis risk

Negative: Overindexes to lower-income consumers

Negative: Spirits subsitiution risk


SAM Market Perform Cautious

Negative: Share donor within craft


Negative: More responsible drinking behavior Source: Cowen and Company

COWEN.COM8


Legal Bud Weighs on Binge Drinking

We have consistently argued that cannabis and alcohol are substitute social lubricants.

To be sure, we do not dispute that alcohol will continue to be quite popular in the U.S.

(generating over $210 bn in annual retail sales today). We are, however, focused on the

marginal alcohol unit, which given the cannabis category’s much smaller size, creates a

sizable opportunity for the cannabis industry.

Herein, we examine the relationship between binge drinking statistics across adult use,

medical use and non-cannabis states, by evaluating (1) binge intensity, and (2) number

of binge drinking sessions per month.

We found that adult use and medical use cannabis states have:

Lower levels of binge intensity, and

Fewer binge drinking sessions per month

We believe this furthers our argument that cannabis and alcohol are substitute social

lubricants. Our assertion (better detailed in our Appendix) is further evidenced by our

previously published analysis that shows:

Outsized switching among younger consumers

Shifting risk perceptions among 18-25 year olds

Decelerating per capita alcohol tax revenue generation pressure in legal

cannabis states

Consumer survey work on alcohol consumption among cannabis consumers

Academic research that concludes medical cannabis weighs on alcohol

purchases

Figure 3 Young Adult Alcohol Incidence on the Decline

Figure 4 As Cannabis Is on the Rise, Beer Continues to Fall

Source: NSDUH and Cowen and Company Note: NSDUH lifetime data is Ages 12+, NIAAA per capita consumption is Ages 14+

Source: NIAAA, NSDUH and Cowen and Company

60.9% 60.9% 60.7%

61.5% 61.6%61.2%

61.5% 61.6%61.0%

60.5%59.9%

59.6%59.0%

57.8%

57.0%

58.0%

59.0%

60.0%

61.0%

62.0%

Alcohol Past Month Incidence (Ages 18-25)

1.05

1.15

1.25

1.35

1.45

25.0%

30.0%

35.0%

40.0%

45.0%

19

79

19

85

19

91

19

93

19

95

19

97

19

99

20

01

20

03

20

05

20

07

20

09

20

11

20

13

20

15 P

er

Ca

pit

a C

on

su

mp

tio

n (in

ga

llo

ns)

Lif

eti

me

In

cid

en

ce R

ate

Lifetime Cannabis Use vs. Beer Per Capita Consumption

Cannabis Beer

COWEN.COM 9


According to the CDC, 1 in 6 adults in the U.S. binge drink four times per month

(consuming seven drinks per binge). A binge drinking session is defined as five or more

drinks for men in about two hours (or four or more drinks for women). According to the

National Institute on Alcohol Abuse and Alcoholism (NIAAA) this is the most common

pattern of excessive alcohol consumption, and unsurprisingly, binge drinking tends to

over-index to younger adult consumers, with over 25% of consumers ages 18-34 having

engaged in binge drinking in 2015. These cohorts, however, have also seen the largest

increases in past month cannabis incidence in the last decade (in percentage points).

Figure 5 Binge Drinking Skews to Younger Adults…

Figure 6 …Where We’re Seeing the Biggest Cannabis Increases

Note: 2015 Youth Risk Behavior Survey conducted through the CDC

Source: CDC and Cowen and Company

Source: NSDUH and Cowen and Company

With increased access to legal cannabis (in particular adult use), we expect to see

pressure on binge drinking behavior. Indeed, adult use states (both from 2016, as well

as new markets that have or will come online by year-end) have binge drinking intensity

rates that fall well below medical and non-cannabis states, while also showing higher

rates of past month cannabis incidence. As well, our analysis found that adult use

cannabis states also have consumers that report meaningfully lower quantities of

alcohol consumed per binge drinking session, relative to medical and non-cannabis

states.

Figure 7 Adult States as of 2016 Report Fewer Binge Drinking Sessions…

Figure 8 …And Lower Levels of Binge Drinking Intensity

Source: CDC, U.S. Census, and Cowen and Company Source: CDC, U.S. Census, and Cowen and Company

17.7%

25.1% 25.7%

19.6%

13.7%

4.6%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

14-17 18-24 25-34 35-44 45-64 65+

Binge Drinking Prevalence by Age

9.3%

16.8%

8.8%

5.5%3.2%

0.2%

9.0%

21.2%

14.9%

8.1%6.3%

2.3%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

14-17 18-24 25-34 35-44 45-64 65+

Past Month Cannabis Incidence by Age

2006 2016

3.7

3.9

4.1

4.3

4.5

4.7

4.9

5% 6% 7% 8% 9% 10% 11% 12% 13% 14% 15% 16%

# o

f B

ing

e S

ess

ion

s P

er M

on

th

Past Month Cannabis Incidence

Legacy (2016) States - Past Month Cannabis

Incidence vs. # of Binge Drinking Sessions

Adult-Use Medical Use Non-Cannabis

6.2

6.4

6.6

6.8

7.0

7.2

7.4

7.6

5% 6% 7% 8% 9% 10% 11% 12% 13% 14% 15% 16%# o

f D

rin

ks D

uri

ng

A B

rin

g

Se

ssio

n


Legacy (2016 ) States - Past Month Cannabis

Incidence vs. Binge Intensity


COWEN.COM10


Indeed, as we look to the expanded legal states that have or will come online between

2016 and 2018, we can see that the relationship still holds, as adult use states have

lower levels of binge intensity and fewer binge drinking sessions than states with more

restrictive cannabis laws. The key difference is that newly opened states have lower

levels of past month cannabis incidence, on a population-weighted basis, which leads us

to believe that past month cannabis incidence will likely climb as cannabis is more

widely available.

Figure 9 A Similar Trend Holds If We Add Current Legal States…

Figure 10 …Which Show Lower Binge Drinking With Higher Incidence

Source: CDC, U.S. Census, and Cowen and Company Source: CDC, U.S. Census, and Cowen and Company

The evolution of these trends suggests that while excessive drinking is curbed by access

to legal cannabis, those states that don’t have access, in particular non-cannabis states,

have rising rates of binge drinking sessions. Meanwhile, we see medical use states

report 17% more binge drinking sessions per month relative to adult use states (that will

be opened by the end of 2018), while non-cannabis states binge 21% more frequently

per month.

Figure 11 Legacy (2016) Cannabis States…

Figure 12 …And, New (2018) Have Lower Levels of Binge Drinking

4.1

4.4

4.7

3.5

4.0

4.5

5.0

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Legacy (2016) States - Avg # of Binge Drinking

Sessions Per Month (Population Weighted)


% Gap Above Adult Use (2016):

Medical Use = +6%

Non-Cannabis = +15%

3.9

4.6

4.8

3.5

4.0

4.5

5.0

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

New (2018) States - Avg # of Binge Drinking

Sessions Per Month (Population Weighted)



Medical Use = +17%

Non-Cannabis = +21%

Source: CDC, U.S. Census and Cowen and Company Source: CDC, U.S. Census, and Cowen and Company

As well, these legal adult use cannabis states also show that their consumers report

fewer drinks consumed during a binge drinking session. Specifically, non-cannabis

states averaged 7.4 drinks per binge, ~12% higher than the 6.6 drinks per binge seen in

3.7

3.9

4.1

4.3

4.5

4.7

4.9

5% 6% 7% 8% 9% 10% 11% 12%

# o

f B

ing

e S

ess

ion

s P

er M

on

th


New (2018) States - Past Month Cannabis

Incidence vs. # of Binge Drinking Sessions


6.6

6.8

7.0

7.2

7.4

7.6

5% 6% 7% 8% 9% 10% 11% 12%# o

f D

rin

ks D

uri

ng

A B

rin

g

Se

ssio

nPast Month Cannabis Incidence

New (2018) States - Past Month Cannabis

Incidence vs. Binge Intensity


COWEN.COM 11


adult use cannabis states (open as of 2016). Meanwhile, when we add new states for

2017 and 2018, we can see that the average moves up to 7.1. This suggests to us as

these new cannabis states mature there is a strong possibility that the number of drinks

per binge session will fall to the levels we have seen for legacy states.

Figure 13 Legacy (2016)

Figure 14 New (2018)

6.6

7.2

7.4

6.0

6.2

6.4

6.6

6.8

7.0

7.2

7.4

7.6

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Legacy (2016) States - Avg # of Drinks During a

Binge Drinking Session (Population Weighted)



Medical Use = +9%

Non-Cannabis = +12%

7.1

7.3

7.4

6.0

6.2

6.4

6.6

6.8

7.0

7.2

7.4

7.6

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

New (2018) States - Avg # of Drinks During a

Binge Drinking Session (Population Weighted)



Medical Use = +3%

Non-Cannabis = +5%

Source: CDC, U.S. Census and Cowen and Company Source: CDC, U.S. Census, and Cowen and Company

COWEN.COM12


Consumption Convergence

As cannabis access has broadened, the makeup of the admitted cannabis user base has

evolved considerably. Over the last decade, as discussed below, the category is

becoming more balanced. For instance, 55+ consumers now make up 17% of past

month users (vs. 5% a decade ago).

New User Outlook. More consumers (all age cohorts) are using cannabis more

frequently (see Fig. 48/49). As such, we re-evaluate our market estimates through an

alcohol lens by juxtaposing category retention rates across these two categories.

Figure 15 Cannabis Trails Alcohol Meaningfully, but We Expect the Category Retention Gap to Close over Time

Note: Ages 18+


New Spending Estimates. We agree with STZ’s estimate that the current cannabis

market is likely $50 bn in total sales (legal and illicit). That, however, creates a modeling

disconnect as we rely on government estimates in terms of the number of users

(currently ~35 mm annual, ~22 mm monthly). The government’s user estimate would

imply weighted average spend to be ~$1,450 / person / year. While this is likely high, it

reflects the under-reporting that we believe occurs in these government surveys, as not

all cannabis users will admit to consuming a Schedule I narcotic in a government survey.

While we believe that annual spend of $1,000 is likely more reasonable, we gross that

estimate up by ~25% to account for this response bias.

Figure 16 First Use of Marijuana Gaps Expanded Considerably Over the Last Decade

Source: NSDUH, U.S. Census and Cowen and Company

U.S. Cannabis TAM Assumptions & Key Metrics Population Drivers

5-Yr Avg 2017 2017 2030 2016 2030

Incidence Rates Past Year Retention (18+) 30.9% 43.1% Past Month Users (in mm)

Change in Lifetime Cannabis (18-25) 0.0% 0.0% Past Month Retention (18+) 64.9% 66.5% 18 to 25 7.2 8.1

Change in Lifetime Cannabis (26+) 0.6% 0.2% 26+ 15.2 30.5

Past Month Spend $1,725 $1,725 18+ 65% 22.4 38.6

Change in Past Month Cannabis (18-25) 0.4% 0.2% Past Year, Not PM Spend $400 $400

Change in Past Month Cannabis (26+) 0.5% 0.4% PM Consumption Growth 3.0% Past Year Users (in mm)

PY Consumption Growth 3.0% 18 to 25 11.4 11.8

Change in Past Year Cannabis (18-25) 0.4% 0.1% Inflation 2.0% 26+ 23.2 46.2

Change in Past Year Cannabis (26+) 0.6% 0.6% Cannabis Price Deflation -5.0% 18+ 34.6 58.0

Total Spend $1,259 1,281.3

Change in Past Month Alcohol (18+) -0.2% -0.2% Past Month Can/Alc Gap 16.6% 27.0%

Change in Past Year Alcohol (18+) -0.2% -0.2% Total Cannabis Industry Spend $74.3 Past Year Can/Alc Gap 20.4% 32.2%

COWEN.COM 13


Annual Spending

In calendar 2017, legal cannabis sales (both adult and medical use) totaled $4 bn in CO,

WA, OR, and NV (which opened on July 1, 2017), up 35% YoY. Annualized for a full year

of NV sales, we believe total cannabis sales would be $4.2 bn (+42% YoY). These four

states account for 6.2% of the population, and 6.4% of GDP. Of note, we exclude Alaska

given its small size, and D.C., where formal retail sales are not permitted. If we assume

no illicit sales, and no cross border movement of product, that would imply annual per

capita spend of $1,375 for all admitted past year cannabis users.

Figure 17 Sales Growth Remains Robust…

Figure 18 …As Legal Cannabis Sales Continue to Climb

Note: Total Growth includes Nevada sales in 2017

Source: State Reports and Cowen and Company


Notably, the $4 bn in sales from these four states accounts for only a fraction of total

cannabis sales in the U.S., as it excludes: (1) legal medical sales, which are not reported,

but we suspect are growing robustly, given the strong patient growth that we are

seeing in the U.S., and (2) illicit sales. Recently, STZ estimated that the total U.S.

cannabis market (legal and illicit) likely generates $50 bn today.

Figure 19 STZ Estimates Current Cannabis Market Is $50 bn

Note: Includes estimated legal and illegal markets in the U.S.

Source: Company Reports, Marijuana Business Daily Factbook 2017, and Cowen and Company

15.3%

53.5%

35.4% 35.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

Cannabis Gross Sales 2017 YoY Growth

$0

$1,000

$2,000

$3,000

$4,000

$5,000

2014 2015 2016 2017

Cannabis Gross Retail Sales (in $ mm)

Colorado Nevada Oregon Washington

$1,659

$765

$2,976

$4,018

$110

$77

$60$50

$0

$20

$40

$60

$80

$100

$120

Beer Cigarettes Wine Total Cannabis

STZ's Estimated Cannabis Sales vs. Other U.S. CPG Categories (in $ bn)

COWEN.COM14


STZ’s current market estimate of $50 bn seems reasonable to us, but does call into

question our previous assumptions around per capita annual spend. Specifically, we had

been using an estimate for ~$1,000 per year for all annual users (with higher spend for

past month users, and lower spend for only annual users). However, that methodology

fails to take into account the response bias.

Our analysis uses government reported data for the number of admitted past month

users. If we assumed that there was no response bias, the 35 mm users would imply a

national per capita annual cannabis spend of ~$1,450. If, however, our assumption for

$1,000 annual spend is correct, that would imply that the number of users in the U.S.

would be closer to 50 mm (~43% ahead of the government’s estimate). As well, we can

also use sales from the biggest legal cannabis states as an additional measure.

Specifically, annualized sales in CO, WA, OR, NV imply annual per capita spend of

$1,375, which assuming a $50 bn total category would imply 36 mm past year cannabis

users in the U.S.

Figure 20 Per Capita Spend Estimates Reflect a Notable Response Bias Among Consumers

Source: State Reports, NSDUH and Cowen and Company

COWEN.COM 15


Lifetime Use.

Over the last decade, lifetime cannabis incidence (trial) has increased roughly 500 bps,

to 47% (which rate has generally held steady over the last three years). The driver of

the increase in overall cannabis incidence looks to be older consumers (26+), where

lifetime trial steadily approaches the rates seen for 18-25 year olds (between 50-55%).

Figure 21 Steady Increases in Lifetime Cannabis Trial…

Figure 22 …Driven by 26+ Year Old Consumers

42.7% 42.7%42.3%

43.7%

44.7%

45.5%

47.1% 47.0%

40.0%

41.0%

42.0%

43.0%

44.0%

45.0%

46.0%

47.0%

48.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Lifetime Cannabis Use (Ages 18+)

53.8%

51.8%

40.8%

46.2%

40.0%

43.0%

46.0%

49.0%

52.0%

55.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Lifetime Cannabis Incidence

18-25 26+

Source: NSDUH and Cowen and Company Source: NSDUH and Cowen and Company

Our analysis assumes no increase in lifetime cannabis incidence among 18-25 year olds,

as that measure has held relatively constant between 50-53% since 2008. However, we

do assume a continued convergence in lifetime trial among 26+ consumers, a

continuation of the historical trend. In 2002, the gap in cannabis trial between these

two age cohorts was 1300 bps; it narrowed to 560 bps by 2016 (with the gap closing an

average of 20 bps per year). Looking ahead, we conservatively estimate no further

increases in lifetime cannabis trial among 18-25 year olds (though it remains ~30 pts

below alcohol) and assume the gap holds. As such, we expect the number of 18-25

lifetime users to be unchanged through 2030, while we expect the 26+ cohort to grow

at a 1.3% CAGR.

Figure 23 Although Lifetime Growth Will

Slow…

Figure 24 …We Expect Lifetime Trial to Converge

0.3%

2.3%

0.0%

1.3%

-0.5%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

18 to 25 26 or Older

Lifetime User CAGRs

2009A-2016A 2017E-2030E

51.8% 51.8%

46.2%

49.0%

42.0%

44.0%

46.0%

48.0%

50.0%

52.0%

54.0%

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

20

17

E

20

18

E

20

19

E

20

20

E

20

21

E

20

22

E

20

23

E

20

24

E

20

25

E

20

26

E

20

27

E

20

28

E

20

29

E

20

30

ELifetime Incidence (Trial) by Age Group



COWEN.COM16


Lifetime use is a key area of conservatism in our forecasting, as expanded access to legal

cannabis should drive more trial than we are anticipating (across all age cohorts).

This is evidenced by the disparities in state-level cannabis “first use” trends among the

18+ cohort nationally (the percentage of the population that has tried cannabis for the

first time, in any given year). Based on state legal status in 2016 (adult, medical, or non-

use), we can see that there are widening gaps, favoring states with legal access.

When we readjust these state weightings to account for states that have or will change

status through year-end 2018 (CA, NV, MA, ME, VT), we can see that “first use” is

depressed. This shows that newly legalized states lag those already legal two years

ago. So, increased access should drive trial. Ultimately, that should also drive category

retention, as there looks to be a positive correlation between first use trends and past

month incidence.

Figure 25 Legacy (2016) Adult Use States Try Cannabis for the First Time

More Often…

Figure 26 …We Expect New States (2018) to Catch Up

Note: Ages 18+


Note: Ages 18+


Figure 27 First Use of Marijuana Increasing Across All States…

Figure 28 …Correlates to Cannabis Access, and Drives Past Month

Incidence

Note: Ages 18+


Note: Ages 18+


1.9%

1.5%

1.3%

0.8%

1.0%

1.2%

1.4%

1.6%

1.8%

2.0%

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

First Use of Marijauna (Population Weighted)


% Gap Below Adult Use (2016):

Medical Use = -25%

Non-Cannabis = -34%

1.7%

1.4%

1.3%

0.8%

1.0%

1.2%

1.4%

1.6%

1.8%

2.0%

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

First Use of Marijauna (Population Weighted)


% Gap Below Adult Use (2016):

Medical Use = -19%

Non-Cannabis = -26%

0.8%

1.0%

1.2%

1.4%

1.6%

1.8%

2.0%

2.2%

5% 6% 7% 8% 9% 10% 11% 12% 13% 14% 15% 16%

Ra

te o

f F

irst

Use

of

Ma

riju

ana


Past Month Cannabis Incidence vs. First Use of Marijuna (18+)


0.8%

1.0%

1.2%

1.4%

1.6%

1.8%

5% 6% 7% 8% 9% 10% 11% 12%

Ra

te o

f F

irst

Use

of

Ma

riju

ana


Past Month Cannabis Incidence vs. First Use of

Marijuna (18+)


COWEN.COM 17


Past Year Incidence.

Since 2009, we have seen steady increases in past year cannabis incidence among all

age cohorts. Of the nearly 35 mm adults that admitted to past year cannabis use in the

U.S. in 2016, ~23 mm were 26+. This reflects continued mix shift to an older user

cohort, as 26+ now accounts for 67% of past year cannabis users, up from 59% in 2009.

Figure 29 User Mix Moving Toward 26+…

Figure 30 …As Older Past Year Use Continues to Grow


Looking ahead, we expect past year cannabis incidence to grow roughly 50% from our

2017E through 2030. A vast majority of that growth will come from the 26+ cohort.

And, despite that outsized growth, we still expect that 26+ incidence will still be 15.0 pts

below 18-25 year olds (vs. the current 22.0 pt gap).

Figure 31 Alcohol and Cannabis Incidence Have Inverse Relationship

Figure 32 Older Consumers Are Narrowing the Incidence Gap

69.8%

69.2%

10.4%

14.1%

8.0%

10.0%

12.0%

14.0%

16.0%

67.0%

68.0%

69.0%

70.0%

71.0%

72.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Ca

nn

ab

is In

cid

en

ce

Alc

oh

ol I

nci

de

nce

Past Year Incidence (Ages 18+)

Alcohol Cannabis

33.0%34.4%

11.0%

19.4%

6.0%

11.0%

16.0%

21.0%

26.0%

31.0%

20

09

20

11

20

13

20

15

20

17

E

20

19

E

20

21

E

20

23

E

20

25

E

20

27

E

20

29

E

Cannabis Past Year Incidence (in %)


2016:

22.0 pt Gap

2030:

15.0 pt Gap


41% 33%

59% 67%

0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

2009 2016

Past Year User Mix by Age


29.8%

33.0%

7.0%

11.0%

6.0%

7.0%

8.0%

9.0%

10.0%

11.0%

12.0%

26.0%

28.0%

30.0%

32.0%

34.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Ag

es

26

+

Ag

es

18

-25

Cannabis Past Year Incidence (in %)

18-25 26+

COWEN.COM18


Past Year Retention.

These incidence estimates are supported by the evolution in cannabis category

retention. On a past year basis, this is measured by past year incidence divided by

lifetime trial. Overall, among the 18+ cohort, close to 31% report consuming on a past

year basis, which rate of category engagement or retention has risen ~700 bps over the

last decade. However, when segmented by age, we can see that past year category

retention among 18-25 year olds (63.7% in 2016) notably exceeds past year retention

among 26+ consumers (23.8%).

Figure 33 Past Year Category Retention on the Rise

24.4%23.4% 23.7% 23.6% 23.6%

22.9% 22.9%

25.2% 25.3% 25.4%26.4%

26.8%

28.2%29.0%

30.6%

20.0%

22.0%

24.0%

26.0%

28.0%

30.0%

32.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Past Year Category Retention (Ages 18+)

Note: Past Year Retention = Past Year Incidence/Lifetime Incidence


While we expect past year cannabis retention to reach ~66% and ~40% for 18-25 year

olds and 26+ consumers, respectively, by 2030, it’s important to view these numbers in

the context of alcohol. In 2016, ~92% and ~79% of 18-25 year olds and 26+ consumers,

respectively, who tried alcohol became past year users. Our numbers reflect a

conservative view on conversion of lifetime (trial) cannabis users, as we expect the

retention gap between 18-25 year olds and those older will only narrow to 26.8 pts by

2030 vs. the current 12.3 pt gap in alcohol.

Figure 34 Cannabis in 2030 Still Below Alcohol

Retention Levels…

Figure 35 …As the Gap Between Older Consumers Is Expected To Narrow Over Time

66.4%

91.5%

39.6%

79.2%

0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

Cannabis

(2030E)

Alcohol (2016)

Past Year Retention

18-25 26+

63.7%66.4%

23.8%

39.6%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

20

02

20

04

20

06

20

08

20

10

20

12

20

14

20

16

20

18

E

20

20

E

20

22

E

20

24

E

20

26

E

20

28

E

20

30

E

Past Year Category Retention

18-25 26+

2016:

39.9 pt Gap

2030:

26.8 pt Gap


COWEN.COM 19


Past Month Incidence.

Over the last decade, past month cannabis incidence has climbed over 50%, to 9.1% in

2016. As is true of past year incidence, age cohort gaps remain wide. With 100% of

population growth coming from 26+, we expect this population segment to drive a

majority (93%) of past month cannabis user growth. In absolute terms, we expect the

number of consumers that report using cannabis on a past month basis to increase ~16

mm, or grow at a 3.9% CAGR for the next 13 years.

Figure 36 Cannabis Past Month Users by Age

6.6 6.8 6.9 7.2 7.2 7.3 7.3 7.3 7.4 7.5 7.6 7.7 7.7 7.8 7.9 7.9 8.0 8.1

11.4 13.5 13.6 15.2 16.2 17.2 18.2 19.2 20.3 21.4 22.4 23.6 24.7 25.8 27.0 28.1 29.3 30.5

0

10

20

30

40

20

13

20

14

20

15

20

16

20

17

E

20

18

E

20

19

E

20

20

E

20

21

E

20

22

E

20

23

E

20

24

E

20

25

E

20

26

E

20

27

E

20

28

E

20

29

E

20

30

E

Past Month Cannabis Users (in mm)

18 to 25 26 or Older Source: NSDUH and Cowen and Company

This outsized growth in older users 26+ is a function of faster population growth rate

relative to 18-25 year olds in the U.S., as well as higher expected changes in incidence

rates. While past month incidence for 18-25 year olds averaged ~+50 bps per year

between 2009 and 2016, we expect this to moderate to +~20 bps per year through

2030, reflecting roughly flat population growth for the cohort and given incidence is

already ~2x the rate of users 26 or older. Meanwhile, users 26 or older have seen

changes in past month incidence average +~40 bps in the prior 7-year period, a trend

that we expect to continue through 2030, given much of the conversion to cannabis

usage will occur in this cohort and the population is expected to grow ~90 bps per year

over the next 13 years.

Figure 37 Cowen Model Expects 18-25 Change in Incidence To Moderate, While 26+ Will Continue Its Trend

18.2%

4.6%

20.8%

7.2%

23.6%

12.8%

0%

5%

10%

15%

20%

25%



2009 2010 2011 2012 2013 2014 2015 2016 2017E 2018E 2019E

2020E 2021E 2022E 2023E 2024E 2025E 2026E 2027E 2028E 2029E 2030E

Avg. Incidence Change

2009-16 = +50 bps

2017E-30E = +20 bpsAvg. Incidence Change

2009-16 = +40 bps

2017E-30E = +40 bps

Note: Ages 18+


COWEN.COM20


Past Month Retention.

The growing rate of category engagement is also evident when we examine the

percentage of past year users that also report consuming cannabis on a past month

basis. Among 18+ year old consumers, past month category engagement has also been

on the rise, having increased 560 bps since 2007. Interestingly, however, these age-

specific gaps are far less evident when we look at the percentage of past year users that

also report cannabis use on a past month basis. Indeed, past month cannabis category

retention does not differ meaningfully between 18-25 year olds and 26+, as there have

been steady increases in category engagement for both age cohorts.

Figure 38 Past Month Category Retention Also on the Rise…

Figure 39 …With Similar Rates of Category Engagement

57.7%58.4%

59.0%60.0% 60.2%

60.8%

63.9%63.2%

54.0%

56.0%

58.0%

60.0%

62.0%

64.0%

66.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Past Month Category Retention (Ages 18+)

57.1%

65.5%

58.1% 63.0%

55.0%

57.0%

59.0%

61.0%

63.0%

65.0%

67.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Past Month Category Retention

18-25 26+

Note: Past Month Retention = Past Month Incidence/Past Year Incidence


Note: Past Month Retention = Past Month Incidence/Past Year Incidence


The latter trend, in terms of consumers transitioning from past year to past month use

makes sense to us as per capita consumption data from the U.S. government has also

shown a trend toward daily use as consumers migrate along the per capita consumption

curve.

Figure 40 Cannabis Frequency Among Past Month Users

40.4%

11.4%6.8%

9.4%

32.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

1-6 7-12 13-18 19-24 25-30

Cannabis Frequency Among Past Month Users (# of Days Per Month)

2002 2003 2004 2005 2006 2007 2008 2009

2010 2011 2012 2013 2014 2015 2016


COWEN.COM 21


Consumer Support at Record Levels

The growth in consumer support for full cannabis legalization for adult use continues to

be readily apparent. Across the U.S., 64% of adults support legalization. Much of the

change in sentiment has occurred over the last 20 years, as support has roughly

doubled. This improved view of adult use cannabis legalization coincides with the initial

introduction of medical cannabis (beginning in CA if 1996), and has accelerated as a

growing number of states have legalized cannabis for medical use (now 29, and D.C.),

and adult use access (9 states, and D.C.).

Figure 41 National Support Has Increased Meaningfully in the Last 20 Years

12%

28%23% 25%

31%36%

46%

58%64%

0%

10%

20%

30%

40%

50%

60%

70%

19

69

19

72

19

75

19

78

19

81

19

84

19

87

19

90

19

93

19

96

19

99

20

02

20

05

20

08

20

11

20

14

20

17

Do you think the use of marijuana should be made legal, or not?

Source: Gallup and Cowen and Company

Encouragingly, this support looks to be increasingly diverse. While Republicans favor

legal access less than Democrats and Independents as of October 2017, a majority (51%)

of Republicans also support adult use. We expect this support to continue to grow.

While as 77% of adults aged 18-34 in 2016 supported legalization of marijuana (up 33

pts since 2003/05), the 35-54 and 55+ age cohorts also saw 26 pt and 16 pt increases,

respectively, over the same time period.

Figure 42 Marijuana Support Across All Political Parties…

Figure 43 …With Growing Support Across All Age Groups

20

51

35

72

4767

0

20

40

60

80

20

03

20

05

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

20

17

Support for Legal Marijuana by Political Party

Republicans Independents Democrats

35%44%

35%29%

60%

77%

61%

45%

0%

20%

40%

60%

80%

100%

National

Adults

18-34 35-54 55+

Support for Legalization of Marijuana, by Age

2003 and 2005 2016

Source: Gallup and Cowen and Company Note: Based on aggregated data from 7/10/13, 7/8/15, 7/13/16, and 7/5/17

Source: Gallup and Cowen and Company

COWEN.COM22


Shifts in Consumer Demographics

Cannabis Use by Age. The growing support for cannabis legalization, as discussed

above, not only reflects the increased access to cannabis, but also the coincident growth

in admitted cannabis use. The data referenced below sources for the U.S. government’s

NSDUH, which likely under-reports actual usage, as not all consumers are likely willing

to admitting to use of a Schedule I controlled substance. Despite this response bias,

there have been steady increases in admitted cannabis use in the U.S. over the last

decade. Among consumers 18+, ~22 mm report past month cannabis use (at 9.1%

incidence), while a greater ~35 mm report annual cannabis use (at 14.1% incidence).

Figure 44 Diverging Trends on a Past Month…

Figure 45 …and Past Year Basis, As Alcohol Falls

54.9%

55.0%

6.0%

9.1%

5.0%

6.0%

7.0%

8.0%

9.0%

10.0%

52.0%

53.0%

54.0%

55.0%

56.0%

57.0%

58.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Ca

nn

ab

is In

cid

en

ce

Alc

oh

ol I

nci

de

nce

Past Month Incidence (Ages 18+)

Alcohol Cannabis

69.8%

69.2%

10.4%

14.1%

8.0%

10.0%

12.0%

14.0%

16.0%

67.0%

68.0%

69.0%

70.0%

71.0%

72.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Ca

nn

ab

is In

cid

en

ce

Alc

oh

ol I

nci

de

nce

Past Year Incidence (Ages 18+)

Alcohol Cannabis

Note: Ages 18+


Note: Ages 18+


Unsurprisingly, admitted cannabis incidence materially overindexes with younger adult

consumers, with past month incidence among consumers under the age of 40 exceeding

the national average. However, it is interesting to see that universally cannabis

incidence has been on the rise for all age cohorts above the age of 18 over the last

decade.

Figure 46 Past Month Use on The Rise with Most Age Cohorts (2006 v. 2016)

16%

10%

7%6% 5% 5% 4%

1% 2%0%

21%

17%

13%

10%

6%8%

6%7%

4%2%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

18-25 26-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+

Past Month Cannabis Incidence by Age

2006 2016


COWEN.COM 23


What is less apparent, but perhaps far more telling, is the growth rates in the absolute

number of admitted past month cannabis users of the last 5 and 10 years. Indeed, while

the number of reported past month cannabis users has grown with every cohort,

growth has been most apparent among older consumers, where we have seen double-

digit growth with every cohort age 55 and above.

Figure 47 Older Cohorts Posting Outsized Growth in Cannabis Use

6.5%

11.2% 10.1%6.4%

8.3%

2.3%

17.8%

12.9%16.4%

30.0%

5.5%7.8%

5.2%

0.5%3.8% 3.9%

21.5% 21.4%

15.3%

31.3%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

26-29 30-34 35-39 40-44 45-49 50-54 55+ 55-59 60-64 65+

Past Month Cannabis Users by Age Cohort

5-Yr CAGR 10-Yr CAGR


This outsized user growth among older consumers has driven a notable shift in

underlying past month cannabis user demographics. Consumers under the age of 30 are

still a large block (38% of admitted consumers in 2016, vs. 45% in 2006). However, the

far more notable change has been the proportion of past month users over the age of

55, who now account for 17% of past month users, up from only 5% 10 years ago.

Figure 48 21-25 Year Olds Remain the Largest Cohort…

Figure 49 …But, 55+ is the Fastest Growing

21-25

29%

26-29

16%30-34

12%

35-39

11%

40-44

11%

45-49

9%

50-54

7%

55+

5%

Past Month User Mix (2006)

21-25

23%

26-29

15%

30-34

14%

35-39

10%

40-44

7%

45-49

7%

50-54

7%

55+

17%

Past Month User Mix (2016)


COWEN.COM24


Cannabis Use by Education Level. The shift in cannabis consumption by age cohort is

mirrored by changes in other consumer segments. From an education perspective, we

continue to see outsized admitted past month incidence led by consumers with lower

levels or educational attainment. However, as was true with the age segmentation, the

most robust growth is coming from more highly educated consumers, that have

historically under-indexed to cannabis consumption.

Figure 50 Cannabis Growing in Popularity with Most Cohorts

Figure 51 Strong Cannabis Uptake Among Highly Educated Consumers

6.6%

11.3%

10.0%

8.3%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16


< High School High School GraduateSome College College Grad

-1.4%

4.1%

10.0%11.6%

1.4%3.1%

8.6%7.0%

-5.0%

0.0%

5.0%

10.0%

15.0%

< High School High School

Graduate

Some College College Grad

Past Month Cannabis Users by Education


Note: Ages 18+


Note: Ages 18+


Interestingly, these past month user rates don’t seem to be a function of lifetime trial.

Indeed, lifetime trial is in fact higher among consumers with higher levels of educational

attainment. This helps to explain the outsized growth that we are seeing from more

highly educated consumers, as these adults had already tried cannabis, and simply look

to be re-engaging with the category, as the groups’ past month use to lifetime trial ratio

looks to be approaching levels that we have seen for consumers with lower levels of

educational attainment.

Figure 52 While Incidence Varies by Education…Trial Is Generally High

8.3%

32.6%

10.0%

44.5%

11.3%

53.5%

6.6%

48.5%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

Past Month Lifetime

Cannabis Use by Education

< High School High School Graduate Some College College Graduate

Note: Based on 2016 data


COWEN.COM 25


Cannabis Use by Income. This normalization in consumer cannabis trends by age and

educational cohort is also mirrored by income-specific data. Growth in the absolute

number of users has been fairly consistent by income cohort. What’s perhaps more

interesting is the consistency in lifetime trial, which is over 40% regardless of income

bracket. As such, while past month use has historically been inversely correlated with

household income, the normalization of cannabis consumption would point to an

outsized opportunity to re-engage lapsed, higher-income consumers.

Figure 53 Lower-Income Consumers Over-Index to Cannabis

Figure 54 Cannabis Trial Consistent Across Income Groups

3.1%

2.3%2.7%

2.3%

3.9%

2.8%

3.7%3.1%

0.0%

2.0%

4.0%

6.0%

< $20K $20K-$49.9K $50K-$74.9K > $75K

10-Yr Change in Cannabis Incidence by Income (2005-2015, in %)

Past Month Past Year

11.3%

41.0%

8.9%

41.0%

7.8%

45.2%

6.6%

47.3%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

Past Month Lifetime

Cannabis Use by Income

< $20K $20K-$49.9K $50K-$74.9K > $75K

Note: Ages 12+


Note: Based on 2015 data


Cannabis Use by Gender. While the gender gap in past month and past year cannabis

use remains wide (with male incidence higher than for women), the gap has been

closing. Specifically, in 2016 male cannabis incidence was 11.7%, while female cannabis

incidence was 6.7%, both on a past month basis. However, over the last decade, male

past month cannabis incidence has risen 3.5 pts, while female cannabis incidence has

increased 2.9 pts.

Figure 55 While Male Incidence Surpasses Female Incidence…

Figure 56 …Female Usage Has Been Accelerating Faster

13.3%

17.3%

7.7%

11.2%

6.0%

8.0%

10.0%

12.0%

14.0%

16.0%

18.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Cannabis Past Year Cannabis Incidence

Male Female

8.0%

11.7%

4.1%

6.7%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Cannabis Past Month Cannabis Incidence

Male Female


COWEN.COM26


Political Pivot

As we think ahead to expanded cannabis access, increased consumer support should

continue to drive a political pivot. Seemingly, the robust support that was broadly seen

during the 2016 election cycle was a good earlier indicator that consumer demands and

political action likely needed to converge further.

Since the 2016 election, there is already ample evidence that political momentum

around cannabis (both medical and adult use) continues to build. Indeed, we have

already seen:

1) West Virginia. Governor Jim Justice (R) signed a medical cannabis bill in April 2017.

In a deep red Republican state, this move is telling, given that West Virginia has one

of the highest rates of opioid morbidity in the country (as discussed below).

Implementation won’t occur sooner than July 2019.

2) Vermont. Vermont became the first state to legalize cannabis for adult use through

their state legislature. While the measure does not include taxation or legal retail, it

is an interesting shift from the 100% ballot-led moves we have typically seen.

3) Utah. In mid-March, Governor Gary Hubert (R) signed House Bill 195, which will

give restrictive medical marijuana access to terminally-ill patients.

2018 Elections

1) Michigan. MI is close to approving a ballot measure for adult use cannabis in the

upcoming 2018 election, which requires supporters to obtain ~250,000 verified

signatures before moving on. After approving medical cannabis in 2008, MI is one

of the largest medical cannabis states in the country, with almost 270,000 qualified

patients as of September 2017 (representing over 20% growth since the year-ago

period).

While the election is still seven months away, it is already becoming a relevant issue

in the upcoming primaries. Among Democratic candidates for Attorney General,

Dana Nessel has been a vocal supporter; her Democratic opponent Patrick Miles

recently flipped into the “yes” column, after not having commented.

2) Illinois. After state senators voted 37-13 in March 2018 to put adult-use

legalization on the November 2018 ballot, it’s already become a hot-button topic in

the upcoming Gubernatorial race. Incumbent Bruce Rauner (R) opposes marijuana

legalization (though he did decriminalize small possessions and extended the state’s

medical cannabis program through 2020). His Democratic opponent J.B. Pritzker,

who won a 6-candidate primary with 45% of the vote (~20 pts ahead of the runner-

up), has made legalizing adult use cannabis as a key campaign point.

Of note, the primary election that determined these candidates in Cook County

included a non-binding measure on adult use legalization. Cook County, which

covers Chicago, the second largest municipality in the U.S., voted overwhelmingly

for adult use cannabis (at 63%), which we find to be a noteworthy straw poll.

COWEN.COM 27


3) Oklahoma. After Jeff Session announced the repeal of the Cole Memo, Oklahoma’s

Governor Mary Fallin (R) announced the inclusion of a medical marijuana ballot

measure for the state’s June 26, 2018 primary. This move comes as a result of a

2016 petition that gathered ~66,000 signatures from state residents in support of

adding the measure to the ballot.

If successfully passed, the measure would allow those aged 18 or older to carry

“small amounts” of cannabis and grow a “restricted number” of plants, with both

requiring a doctor’s permission.

Adult-Use States on the Horizon

1) New Jersey. NJ has been contemplating the legalization of adult use cannabis since

2016. Then Senate President Stephen Sweeney (D) noted that legalization would

be a “game-changer” for the state and indicated his intention to pursue legalization

in 2018 (upon the election of a new governor). Newly-elected Governor Phil

Murphy (D) unveiled his budget in late March, which included a provision for

legalization by January 2019 (with an estimate for $60 mm in tax revenue

generation). The State Assembly has until June to ratify the budget.

2) Connecticut. While exiting Governor Dannel Malloy (D) has historically been against

cannabis legalization, in the Governor’s FY19 Budget, while not in the formal

budget proposal, his administration did include an alternative to the recommended

revisions, which included the possibility for “legalizing and taxing the recreational

use of marijuana.”

CT’s Gubernatorial primary, which has crowded fields on both sides, will be held on

August 14, 2018 (ahead of the general election in November). In the meantime, the

General Assembly has the opportunity to consider legalization as well, with several

bills in Committee, though the measure continues to see Republican opposition.

3) Delaware. The Adult Use Cannabis Task Force studied the issue for six months,

with the committee concluding their work in March 2018. While the vote was

close, the committee will release their report, which findings can be utilized by the

General Assembly in crafting potential legislation.

4) Rhode Island. In April 2017, Governor Gina Raimondo (D) endorsed the formation

of an exploratory committee around the legalization of adult use cannabis.

5) New Mexico. In February 2018, the Senate Rules Committee voted four to three to

include a ballot in the upcoming November election to vote on adult use cannabis

legalization. As well, the state legislature is considering adding opioid use disorder

as a qualifying condition for medical cannabis patients.

6) Minnesota. House File 2714 and House File 926 were submitted in May and

February 2017, respectively, in order to have the question of adult use cannabis

legalization appear on the November 2018 ballot. Out of the Democratic-Farmer-

Labor candidates, five out of six of them, including the retiring DFL Governor Mark

Dayton, have said they favor legalizing cannabis for adult use.

COWEN.COM28


Medical States on the Horizon

1) Nebraska. In January 2018, Legislative Resolution 293CA was introduced by

Senator Anna Wishart (D) to add a constitutional amendment for medical cannabis

use, subject to regulation. Governor Pete Ricketts (R), a known cannabis-opposer,

will face opposition from cannabis-enthusiast Krystal Gabel (R) in the upcoming

election.

2) Missouri. New Approach Missouri, the leading campaign trying to bring legal

medical cannabis to Missouri recently collected 250,000 signatures to get medical

cannabis on the November 2018 ballot. The petition has until May 2018 to collect

signatures before filing with the secretary of state for verification.

3) Utah. In addition to approving restrictive medical marijuana use to terminally ill

patients, Governor Gary Hubert (R) also approved House Bill 197 and Senate Bill

130, which would allow the state Department of Agriculture and Food to contract

the production and storage of approved marijuana products, and to approve the

sale of CBD oil, respectively.

This mounting focus on cannabis is not a surprise given the broad-based support for

legalizing cannabis evident in the November 2016 elections, where 8 out of 9 cannabis

ballot measures passed (and Arizona losing by a small 130 bps margin). As a result,

~60% of the U.S. population now resides in a state that has (or will have) access to

medical cannabis. And, by year-end 2018, assuming MA and ME open on time, adult-use

cannabis will cover ~25% of U.S. GDP.

Figure 57 Clear Support For Cannabis Across Most Voting States in 2016

Election 2016 - Cannabis Ballot Results

State Ballot Number Approval Disapprove

Legalization Ballot Initatives

California Proposition 64 57.1% 42.9%

Maine Question 1 50.3% 49.7%

Massachusetts Question 4 53.7% 46.3%

Nevada Question 2 54.5% 45.5%

Arizona Proposition 205 48.7% 51.3%

Medical Ballot Initiatives

Arkansas Issue 6 53.1% 46.8%

Florida Amendment 2 71.3% 28.7%

Montana I-182 57.9% 43.7%

North Dakota Measure 5 63.8% 36.2%

Source: Ballotpedia and Cowen and Company

COWEN.COM 29


Attractive Tax Revenue Opportunity

One reason that we believe cannabis legalization will continue to gain momentum is that

in CO, WA and OR collectively, tax revenue generation has meaningfully exceeded

expectations. According to the Cato Institute: “One area where legal marijuana has

reaped unexpectedly large benefits is state tax revenue.”

In total, CO, WA and OR are now expected to generate $1.4 billion in cannabis tax

revenue in the upcoming biennial fiscal year, 24% higher than was expected 12 months

ago. Moreover, each of these three states has consistently raised their tax revenue

forecasts on a 12-month basis (e.g., 12 months before the preliminary data was

released), in each of the full fiscal years that adult use cannabis has been legal.

Impressively, cannabis is expected to account for 1.3% of GDP for these three states

between 2017 and 2019 (according to state budget forecasts), with the contribution

from cannabis taxes to GDP having doubled in 2016, and having increased another 50%

in 2017 in CO, WA and OR combined.

Taking our 2030 estimate for total gross cannabis sales of $75 bn, and assuming all of

that is through legal channels, we estimate that the cannabis category could ultimately

generate ~$17.5 bn in tax revenues, including:

1) ~$3 bn in Federal Taxes (assuming a 5% federal excise tax),

2) $11.5 bn in States Taxes (assuming a 20% state excise tax),

3) ~$3 bn in Local Taxes (assuming a 5% tax rate).

Figure 58 Tax Revenues Generally Ahead of Expectations

Figure 59 CO Has Been a Consistent Positive Revision Story

1.7%

1.1%

0.3%

1.0%

2.4%

1.0%0.8%

1.3%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

CO WA OR CO+WA+OR

Cannabis % Contribution to Overall Revisions

(2017-2019 - Biennium)

12 Month Ago Expectations Current Expectations

0.3%

0.6%

0.9%

1.3%

0.0%

0.2%

0.4%

0.6%

0.8%

1.0%

1.2%

1.4%

2015 2016 2017 2017-

2019E

% Cannabis Tax Contribution (CO+WA+OR)

Note: CO and WA based on Sept Expectations and OR based on May expectations for current year, June expectations for past year; Not adjusted for changes in tax structure

Source: Colorado Legislative Council Staff, Washington Economic and Revenue Forecast Council, Oregon Office of Economic Analysis, and Cowen and Company

Source: Colorado Legislative Council Staff, Washington Economic and Revenue Forecast Council and Cowen and Company

COWEN.COM30


Colorado

FY 16/17 tax receipts of $211 million, ~75% higher than expected two years

ago (Dec. 2015).

~50% increase in taxes YoY, in the 4th year of adult use.

Figure 60 CO Steadily Raised Outlook…

Figure 61 …As Tax Receipts Grow 50% in Year 4

$0

$50

$100

$150

$200

$250

Total Proposition

AA Taxes

Total 2.90% Sales

Tax

Total Taxes on

Marijuana

CO Projected Cannabis Taxes (FY 16/17)

Dec-14 Dec-15 Dec-16 Dec-17

+75%+83%

+50%

2%6% 6%

11%15%

11%14%

19%22%

33%32%

44%49%49%

0%

10%

20%

30%

40%

50%

60%

Se

p-1

4

De

c-1

4

Mar

-15

Jun

-15

Se

p-1

5

De

c-1

5

Mar

-16

Jun

-16

Se

p-1

6

De

c-1

6

Mar

-17

Jun

-17

Se

p-1

7

De

c-1

7

CO Projected Growth for FY 16/17 Cannabis

Taxes

Source: Colorado Legislative Council Staff, Washington Economic and Revenue Forecast Council and Cowen and Company

Source: Colorado Legislative Council Staff and Cowen and Company

Oregon

FY17 tax receipts came in over 2x what was forecast in the prior year, at $70

million (up from $33 million in May 2016).

The biennium tax outlook for OR looks equally encouraging, with an expected

$160 million in receipts as of December 2017, up 156% since May 2016.

Figure 62 Tax Revenue Generation in Oregon…

Figure 63 …is Expected to Accelerate

$11$12

$28$22

$33

$21

$70

$0

$20

$40

$60

$80

FY 2016 FY 2017

Gross Marijuana Revenue Projections (not incl. licenses and fees), USD in mm

HB2041(2015) SB 1511 (2016) May-16 Actual

$47$62 $62 $62

$157 $160 $160

$0

$50

$100

$150

$200

Gross Marijuana Revenues Porjections (FY 2017-

19 Biennium), USD in mm

Source: Oregon Legislative Revenue Office, Oregon Office of Economic Analysis and Cowen and Company

Source: Oregon Legislative Revenue Office, Oregon Office of Economic Analysis and Cowen and Company

COWEN.COM 31


Nevada

While Nevada is a relatively newer state, trends look encouraging. In the first seven

months of legalization (after the market opened on July 1, 2017), tax revenue collection

has totaled ~$36 mm. What is more, in the last four months, tax revenue generation

has exceeded $5 mm (which is the targeted collection that the state budgeted for). On

its current trajectory, tax revenue generation looks poised to at least come in $11 mm

ahead of the state’s two-year estimate (for $120 mm in tax collection).

Figure 64 After Seven Months, Nevada’s Run-Rate Is On-Track to Meet Its 2-Yr Target

$36

$131

$0

$50

$100

$150

Jul-

17

Au

g-1

7

Se

p-1

7

Oct

-17

No

v-1

7

De

c-1

7

Jan

-18

Feb

-18

Mar

-18

Ap

r-1

8

May

-…

Jun

-18

Jul-

18

Au

g-1

8

Se

p-1

8

Oct

-18

No

v-1

8

De

c-1

8

Jan

-19

Feb

-19

Mar

-19

Ap

r-1

9

May

-…

Jun

-19

Nevada Cumulative Cannabis Taxes at T3M Run-Rate ($ in mm)

Note: Assumes a $5.6 million run-rate beginning in February 2018 (which is the trailing 3-month average)

Source: Nevada Department of Taxation and Cowen and Company

Washington

WA is a cautionary tale on excessive taxation, in particular when competing with a large

and established illicit market.

Biennial total cannabis excise taxes and license fees for 2015 – 2017 came in

at $469 million, 250% higher than the February 2014 projected cannabis

excise taxes and license fees.

Washington originally set their cannabis tax as a 25% tax rate on the producer,

processor and retailer, but changed it to a single 37% rate charged at retail in

July 2015. This change coincided with a change to tax modeling methodology,

which has improved the accuracy of Washington’s forecasting.

Figure 65 WA Made a Notable Tax Change…

Figure 66 …Though Tax Growth Rates Remained Impressive

$749

$0

$200

$400

$600

$800

Feb-

14

Sep-

14

Nov-

14

Feb-

15

Jun-

15

Sep-

15

Nov-

15

Feb-

16

Jun-

16

Sep-

16

Nov-

16

Mar

-17

Jun-

17

Sep-

17

Nov-

17

Feb-

18

WA Projected Cannabis Excise and License Taxes (FY 2017-2019 Biennium), USD in mm

96%

75%64%

86%

67%63%60%58%55%55%47%46%

58%58%60%

0%

20%

40%

60%

80%

100%

120%

Sep-

14

Nov

-14

Feb-

15

Jun-

15

Sep-

15

Nov

-15

Feb-

16

Jun-

16

Sep-

16

Nov

-16

Mar

-17

Jun-

17

Sep-

17

Nov

-17

Feb-

18

WA Projected Growth (FY 2017-2019 Biennium)

Note: Legislation was passed changing the tax structure between the Feb 2015 and June 2015

Source: Washington Economic and Revenue Forecast Council and Cowen and Company

Note: Legislation was passed changing the tax structure between the Feb 2015 and June 2015

Source: Washington Economic and Revenue Forecast Council and Cowen and Company

COWEN.COM32


Criminal Justice Cost Savings: $3bn + Opportunity

According to the ACLU, over 8 mm individuals in the U.S. were arrested for cannabis

related charges between 2001 and 2010 (with 88% arrested for simple possession).

Indeed, in 2010, cannabis accounted for 52% of total drug-related arrests. And, among

those arrested, African-Americans are four times more likely to be arrested, even

though use rates between African Americans and Whites are fairly similar (with even

higher proportions of arrests in select states). While cannabis clearly represents an

important social justice issue, legalization would also offer meaningful cost savings. The

same ACLU study estimated that states spend $3.6 billion annually enforcing marijuana

laws.

Figure 67 Cannabis Arrests…

Figure 68 …Disproportionately Target African Americans

4.7x

4.8x

5.2x

6.0x

6.0x

7.6x

7.8x

8.1x

8.3x

4.0x 5.0x 6.0x 7.0x 8.0x 9.0x

NY

NE

SD

PA

KY

WI

IL

MN

D.C.

Times More Likely Blacks Are Arrested Than Whites For Marijuana

Source: ACLU, FBI/Uniform Crime Reporting Program Data, U.S. Census Source: ACLU, FBI/Uniform Crime Reporting Program Data, U.S. Census and Cowen

and Company

Growing Approval Among Law Enforcement

In terms of police opinions on cannabis legalization, a Pew Research survey from

January 2017 of nearly 8,000 police officers found that a majority (69%) support some

form of legalization (at least medical). Meanwhile, 32% support full adult use

legalization. As a point of comparison, this survey also studied the broader civilian

population and found 49% approval for adult use cannabis (below the Gallup readings

from October 2017 of 64%). 30% of police officers do not think cannabis should be legal

at all, compared to only 15% among all Americans in this same survey. For police

officers under the age of 35, support for full legalization was 37%. Meanwhile, for those

over the age of 50, support was a lower 27%.

Figure 69 32% of Police Support Legalizing Cannabis for Adult Use

Source: Note: Survey of law enforcement conducted 5/16-8/16, survey of U.S. adults conducted 8/16-9/16

Source: Pew Research and Cowen and Company

32%49%

37%

35%

0%

20%

40%

60%

80%

100%

Police Officers General Public

Views on Marijuana Use (% Approval)

Medical + Adult-Use Medical Use Only

COWEN.COM 33


Robust Medical Demand, When It’s Done Right (& Until Adult Use)

In medically legal, non-adult use states, cannabis patient counts have been posting

robust growth. This consistent momentum underscores the increasing demand for

cannabis and herein we look at 3 such states (FL, AZ, and NY), which continue to benefit

from actions / enhancements to their respective medical cannabis programs.

Figure 70 Medical Cannabis Patient Counts in Non-Adult Use States Have Been Growing at a Torrid Pace

Source: State Reports and Cowen and Company; Note: Dates below state acronyms represent the time-periods used to calculate growth rates

Momentum Markets:

Florida. FL benefitted from expanded medical cannabis access as a result of the

election, as the program now allows for access to high THC products (as opposed to only

low-THC / high-CBD). While early days as the market expands for broader access to

medical cannabis (as the bill also included increased licenses), the patient count in FL

(where the population is 16 mm) has increased 5x since May 2017. And, just since the

end of 2017, we have seen the number of patients grow 41% in the first three months

of 2018, matched by 41% growth in the number of qualified physicians (at 1,246 as of

March 23).

The much broader access to medical cannabis should continue to drive growth in

patients and prescribing physicians, given the severity of the opioid crisis in FL.

Specifically, the Florida Department of Law Enforcement found that in 2016 the state

saw a 97% increase in fentanyl-related deaths (to 1,390 deaths). Meanwhile, opioid-

related deaths were up 35% (to 5,725).

050,000

100,000150,000200,000250,000300,000

MI AZ FL NM NY IL CT MT HI RI MN NH

9/16-

9/17

2/17-

2/18

5/17-

3/18

2/17-

2/18

3/17-

3/18

6/17-

1/18

5/17-

3/18

2/17-

2/18

2/17-

2/18

12/16-

12/17

12/16-

12/17

12/16-

12/17

Medical Cannabis Patients (Non-Adult Use States)

+23% +33% +440% +45% +219% +44% +34% +110% +31% +17% +102% +128%

COWEN.COM34


Figure 71 With a New Framework…

Figure 72 …Medical Cannabis Is Off to the Races in FL

Source: State Reports and Cowen and Company Source: State Reports and Cowen and Company

Arizona. Since becoming legal, Arizona’s medical cannabis market has expanded

considerably, ending December with ~153,000 patients. This marks a ~34% increase

relative to 2016, which grew 30% YoY. Through the first 2 months of 2018, Arizona has

added an additional ~5,500 patients. The annual growth in medical patients is notable

when considering Arizona is bordered by 3 adult use states (CA, NV and CO).

Lawmakers recently introduced SB 1420, which main purpose is to protect patients

through increased regulation by requiring dispensaries to have their product tested by

state-licensed laboratories. This bill also originally included a provision to lower the cost

of registration cards from $150 to $50, with $25 renewals, though the provision was

pulled from the bill.

Figure 73 Arizona’s Medical Market Has Been Steadily Increasing

Note: YTD18 reflects data through 2/28/18


New York. The medical cannabis market in NY has been steadily increasing as

evidenced by the number of certified patients partaking in the program. The increase in

patient count has been amplified through a series of enhancements, which include

adding chronic pain and PTSD as qualifying conditions. Indeed, in early August, the NY

Department of Health indicated that NY’s certified patient count was up 77% since

adding chronic pain as a qualifying condition roughly four months earlier in late March

2017. PTSD is a more recent enhancement, taking effect in late December 2017. As of

March 20, 2018, NY had over 48,000 certified patients, representing a 25% increase

since early December 2017.

17 2027

36 3847

5764

72

8491

0

20

40

60

80

100

May

-17

Jun

-17

Jul-

17

Aug

-17

Sep

-17

Oct

-17

No

v-1

7

Dec

-17

Jan-

18

Feb-

18

Mar

-18

Florida Medical Patients (in thousands)

~5x increase in 10 months

02004006008001,0001,2001,400

-

20,000

40,000

60,000

80,000

100,000

12/

29/2

017

1/5

/20

18

1/1

2/2

018

1/1

8/2

018

1/2

6/2

018

2/2

/20

18

2/9

/20

18

2/1

6/2

018

2/2

3/2

018

3/2

/20

18

3/9

/20

18

3/1

6/2

018

3/2

3/2

018 Q

ualif

ied

Phy

sici

ans

# o

f R

egis

tere

d P

atie

nts

Florida Medical Cannabis Market

Qualified Physicians Registered Patients

17.9

34.743.1

61.3

87.9

114.4

153.0 158.5

0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0

180.0

2011 2012 2013 2014 2015 2016 2017 YTD18

AZ Medical Cannabis Patients (in thousands)

COWEN.COM 35


Figure 74 The NY Medical Market Is Approaching 50,000 Users


Impact of Adult Use on Medical

Colorado. Colorado had been increasing its medical patient count consistently prior to

the legalization of adult use, which took place in late 2014. We can see that post adult-

use legalization, the number of medical cannabis patients has declined for the last 3

consecutive years, and is down almost 20% from its peak of over 115,000 patients in

2014.

Figure 75 Medical Patients Had Been Growing Consistently in Colorado Until Adult Use Legalization

82

109111

115

108

95 93

60

70

80

90

100

110

120

2011 2012 2013 2014 2015 2016 2017

CO Medical Cannabis Patients (in Thousands)


The decline in patient count post legalization makes good sense to us, as certain medical

participants would seemingly transition to adult use in order to avoid having their name

in a registered patient database, as well as to avoid potential co-pays on office visits.

We can see below that the adult use market has more than offset the headwind from

lower patient counts. Adult use cannabis revenues continue to post strong DD growth

(averaging almost 30% in 2017), while the sales mix shift of adult use has increased 20

pts since January 2015, making up three quarters of the total Colorado cannabis market

as of December 2017.

11 12 14 15 17

22 24 27 29

31

39

46 48

0

10

20

30

40

50

60 NY Medical Cannabis Patients (in Thousands)

COWEN.COM36


Figure 76 Adult Use Sales Notably Outpacing Medical…

Figure 77 …Resulting in Increased Mix Shift

Source: State Reports and Cowen and Company Source: State Reports and Cowen and Company

The impact of adult use cannabis can be seen using Oregon as a case study as well. Over

70,000 people were medical patients in 2014, which number grew 10% in 2015, topping

out at over 77,000 patients. Oregon began selling adult use cannabis in early 2016,

which resulted in a 13% YoY decline. Most recently in 2017, the amount of medical

cannabis patients decreased notably, down 25% YoY.

Figure 78 Oregon Cannabis Patients Continue to Fall Post Adult Use Legalization


-50.0%

0.0%

50.0%

100.0%

150.0%

200.0%

Jan

-15

Ap

r-1

5

Jul-

15

Oct

-15

Jan

-16

Ap

r-1

6

Jul-

16

Oct

-16

Jan

-17

Ap

r-1

7

Jul-

17

Oct

-17

YoY Change in Sales - CO

Medical Recreational

$0

$50,000,000

$100,000,000

$150,000,000

Jan

-15

Ap

r-1

5

Jul-

15

Oct

-15

Jan

-16

Ap

r-1

6

Jul-

16

Oct

-16

Jan

-17

Ap

r-1

7

Jul-

17

Oct

-17

CO Sales Mix

Medical Recreational

70

77

67

50

0

10

20

30

40

50

60

70

80

90

2014 2015 2016 2017

OR Medical Cannabis Patients (in Thousands)

COWEN.COM 37


OG Kush to Offset Opioids?

According to the White House Council of Economic Advisors November 2017 report,

opioids have a total annual cost to the U.S. of $500 bn.

To frame the scope of this problem, the Department of Health and Human Services

reported that in 2016:

o 11,500,000 individuals misused opioids,

o 2,100,000 people had an opioid use disorder,

o 2,100,000 people misused prescription opioids for the first time,

o 42,249 individuals died from an opioid overdose,

o An additional 19,413 people died from a synthetic opioid overdose, and

o 15,469 overdoses on heroin.

While legalizing cannabis will certainly not be a comprehensive solution, as discussed

herein, there is growing compelling evidence that it could help significantly. Indeed,

should the U.S. experience the same 6% decline in opioid deaths seen in CO in the first

two years of legalization that would reduce the cost of the opioid crisis by $30 bn

(assuming 2016’s total $504 bn opioid burden nationally).

Figure 79 Opioid Deaths Are Skyrocketing

Source: CDC

COWEN.COM38


https://www.google.com/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=2ahUKEwiBpNWy74faAhVHc98KHdFSBb8QjRx6BAgAEAU&url=https://www.cdc.gov/drugoverdose/data/analysis.html&psig=AOvVaw1w3q7F7JsBjHURW0Cak1BI&ust=1522080615035009

New Research Is Increasingly Encouraging

The body of compelling research supporting our assertion that cannabis is an

underappreciated policy solution for the opioid crisis continues to grow, with two

noteworthy studies out already in 2018 (see the Appendix for additional peer-reviewed

research).

Minnesota. A 2018 study published by Minnesota’s Department of Health on the state’s

medical marijuana program was particularly encouraging. Looking at medical cannabis

patients (N=2,174) that were being treated for “intractable pain” from August –

December 2016, and purchased medical cannabis during the survey period. Patients

reported that:

o 63% were able to reduce or eliminate opioids after six months

o 62% reported a reduction in pain severity

o 54% indicated marijuana provided “a high level of benefit”

These results were validated by the prescribing physicians, who reported that 58% of

patients were able to reduce or eliminate their opioid reliance.

Israel. A 2018 Israeli Study published in the European Journal of Internal Medicine

evaluated the use of medical cannabis for cancer patients, which has been approved

since 2007. The nearly 3,000 cancer patients tracked between 2015 and 2017 were of

an average age of 60, and reported the following symptoms:

o 78% sleep problems

o 73% weakness

o 65% nausea

o 49% lack of appetite

After six months of follow-up:

o 25% had died

o 19% had stopped treatment

o 4% reported no change

Of the remaining close to 60%: 96% reported an improvement of condition

The study concluded: “Cannabis as a palliative treatment for cancer patients seems to

be well tolerated, effective and safe option to help patients cope with the malignancy

related symptoms.”

COWEN.COM 39


Cowen Proprietary Analysis

Below, we expand on the prior research using data on opioid overdose mortality from

the CDC spanning from 2000 to 2016. In evaluating states where cannabis has been

legal as of 2000 (CA, AK, OR, WA, ME, and HI), we can see that over time while opioid

overdose mortality contintues to climb, overdose incidence consistently trails in states

where medical cannabis is available. Indeed, over the last 10 years, growth in opioid

mortality in legal medical cannabis states has trailed non-legal states by ~30%.

Figure 80 Opioid Overdose Mortality Has Increased In All States But Slower in Medical Cannabis States

18.4

14.6

4.0

7.0

10.0

13.0

16.0

19.0

20

00

20

01

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Opioid Overdose Mortality Per 100,000 Population

States Legal as of 2000 Non-Legal States

Note: Non-legal states exclude states that legalized medical cannabis between 2001 and 2016. Data is based on crude rates.


Looking at more recent data, given that 8 states have introduced legal medical cannabis

between 2001 and 2010, we can see that these trends impressively still hold.

Figure 81 Including States That Have Legalized Cannabis More

Recently…

Figure 82 …This Trend Holds True

1.6% 1.8%

5.6%

4.1%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%


Opioid Overdose Mortality Per 100,000


4.2%

5.6%

3.0%

3.5%

4.0%

4.5%

5.0%

5.5%

6.0%


Opioid Overdose Mortality Per 100,000 Population (2011-2016 CAGR)

Note: Non-legal states exclude states that legalized medical cannabis between 2006 and 2016. Growth is based on crude rates.


Note: Non-legal states exclude states that legalized medical cannabis between 2011 and 2016. Growth is based on crude rates.


COWEN.COM40


Colorado Case Study: $30 Billion Opportunity in 2 Years

While much of the research on opioid and cannabis use focuses on medical marijuana, a

recent study published in the American Journal of Public Health examined the impact of

adult use cannabis on opioid overdoses in Colorado.

After steady increases in opioid overdoses, the study found that overdoses declined 6%

in CO in the first two years post legalization of adult use cannabis (2014/2015).

Adjusting for changes to the state’s opioid program, the researchers estimated the

decline would have been 6.5%.

If we extrapolate that decline nationally, we would estimate that national cannabis

legalization could save the country $30 bn in the first two years of legalization

(assuming the current $504 bn in annual costs, which will likely only rise from here

without effective policy solutions, and a 6% two-year decline in opioid deaths).

Figure 83 Legal Cannabis Driving Declines in Opioid Deaths in Colorado

Source: American Journal of Public Health

California Case Study

As the first legal cannabis state in the U.S. (1996), California serves as the most robust

case study in terms of the interaction between opioids and legal cannabis (in addition to

being the country’s most populous state). Impressively, CA now has one of the lowest

levels of opioid morbidity in the nation, at 13.7 deaths per 100,000. And, relative to the

national average, CA has consistently trailed below the national average opioid

mortality rate for 12 consecutive years. Ironically, the LA Times recently published an

article noting this phenomenon, also indicating however that experts cannot explain

why. Reader comments consistently pointed to cannabis access.

COWEN.COM 41


Figure 84 Opioid Deaths in CA Have Trailed that of the U.S…

Figure 85 …For 12 Straight Years

4.0

9.0

14.0

19.0

24.0

19

99

20

00

20

01

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Opioid Overdose Mortality Per 100,000

California U.S.

0.50

0.70

0.90

1.10

1.30

1.50

19

99

20

00

20

01

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

CA vs. U.S.: Relative Opioid Overdose Mortality Per 100,000 Population

Note: Data is based on crude rates.


Note: Data is based on crude rates.


Arizona Case Study

Looking at a more recent medical cannabis state, Arizona (which legalized medical

cannabis in 2011) has also shown signs of progress in combatting the state’s above

average opioid mortality. In Arizona, where ~75% of medical cannabis patients report

severe chronic pain as an indication, we have seen AZ’s mortality rate relative to the

U.S. decrease over the past 4 years, while patient count has grown to ~153,000

(representing ~34% YoY growth).

Figure 86 In Arizona, Relative Overdose Mortality Began to Fall…

Figure 87 …As the Medical Cannabis Program Continued to Grow

0.80

1.00

1.20

1.40

1.60

1.80

19

99

20

00

20

01

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

AZ vs. U.S.: Relative Opioid Overdose Mortality Per 100,000 Population

17.934.7

43.1

61.3

87.9

114.4

153.0

0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0

180.0

2011 2012 2013 2014 2015 2016 2017

AZ Medical Cannabis Patients (in thousands)

Note: Data is based on crude rates


Source: Arizona Department of Health Services and Cowen and Company

COWEN.COM42


Unintended Consequences

Gateway Drug/Addiction

In an attempt to define clearly what is meant by the term “addiction” and which drugs

are most addictive, Dr. Jack E. Henningfield of the National Institute on Drug Abuse and

Dr. Neal Benowitz of the University of California at San Francisco offered the rankings

on six different substances. Among them, marijuana broadly ranks as the least

addictive, with particularly low rankings for dependence and addiction potential,

relative to nicotine, heroin, cocaine, and alcohol. Indeed, it is perhaps most telling that

for ~67% of these addiction measures cannabis in fact scored below caffeine.

Figure 88 Cannabis is Less Addictive Than Other Drugs

0

1

2

3

4

5

6

Nicotine

[LEGAL]

Heroin

[ILLEGAL]

Cocaine

[ILLEGAL]

Alcohol [LEGAL] Caffeine

[LEGAL]

Marijuana

[ILLEGAL]

Substances Compared

Withdrawal Reinforcement ToleranceDependence Intoxication Addiction Potential

Source: NY Times and Cowen and Company

COWEN.COM 43


Impaired Driving: Alcohol Is A Bigger Problem

The legalization of cannabis raises the issue of intoxicated driving. The Governors

Highway Safety Association’s Drug Impaired Driving Guide for States (April 2017) offers

some relevant insights. Among the drug tested drivers that were fatally injured, ~12%

tested positive for marijuana in some form. Meanwhile, ~37% of fatally injured drivers

tested positive for alcohol. This was consistent with findings from Canada.

Figure 89 In U.S. Fatal Crashes, Alcohol Is a Big Factor…

Figure 90 …As Is True in Canada

44.6%37.3%

18.9%12.2%

7.4%3.2% 2.9%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

All

Dru

gs

Alc

oho

l

Oth

er

Dru

gs

Mar

ijua

na

Dru

g N

ot

on

FA

RS

Lis

t

Am

ph

eta

min

e

Un

kno

wn

U.S. % of Fatal Crashes

39.1%

34.2%

21.6%

13.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

Alcohol All Drugs Other Drugs Marijuana

Canada % of Fatal Crashes

Source: GHSA and Cowen and Company Source: GHSA and Cowen and Company

A secondary analysis the GHSA relied upon was roadside testing, where results varied

notably between the U.S. and Canada. In the U.S., marijuana was the most often

detected, though alcohol closes the gap significantly on the weekends. In Canada, like

the fatality reports in both markets suggest, alcohol is far more prevalent than

marijuana in roadside tests. We find the data from Canada to be more compelling,

given:

1) Absence of reliable cannabis testing technology (as opposed to alcohol)

2) The reduced stigma around cannabis in Canada, vs. the U.S., which could eliminate

some of the subjectivity found in road-side cannabis testing.

Figure 91 U.S. Screens High for Weed in Roadside Stops…

Figure 92 …Canada Data Is More Consistent with Fatality Records

10.3%

7.3%

11.7%12.6%

1.1%

8.3%

0.0%

3.0%

6.0%

9.0%

12.0%

15.0%

Weekday Weekend

U.S. Roadside Suvey (2013-2014)

Medication Marijuana Alcohol

7.4%

6.5%

3.3%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

Any Drug Alcohol Marijuana

Canada Roadside Survey (2012)

Source: GHSA and Cowen and Company Source: GHSA and Cowen and Company

COWEN.COM44


Clearly, testing methodology needs to be improved, and that innovation will likely

accelerate as California and Canada both legalize cannabis in 2018. In the meantime, it

has been encouraging to see that in CO, the total number of DUI/DUID citations has

fallen steadily since 2014, with marijuana’s contribution down through September

2017.

Canada Testing. Ahead of 2H18 legalization, Canada had begun roadside

testing, which the federal government deemed successful. Police officers from

seven jurisdictions administered roadside testing via random stops and

roadblocks, collecting over 1140 saliva samples using 2 kinds of devices

between mid-December and early March. These portable screening devices

can be used to detect the recent presence of several drugs including THC. Of

all the roadside tests, 80% were taken from drivers and 20% from passengers

and approximately 15% of the tests registered a positive drug reading. Testing

remains a work in progress, however, as questions remain on timing of

ingestion (as a positive reading could indicate someone smoked the day or

week before).

Colorado. After legalizing recreational cannabis in an attempt to documents

trends, the Chief of the Colorado State Patrol ordered the specific tracking of

marijuana-related citations beginning in January 2014. While overall

DUI/DUID citations have been declining YOY, marijuana’s share as a

percentage of total DUI/DUID citations has increased over the past 2 years,

contributing just over 17% in 2016. This trend looks to have reversed through

September 2017, however, as marijuana’s contribution to total DUI/DUID

citations has decreased to just over 14%.

Figure 93 DUI/DUID Citations Falling Steadily in CO…

Figure 94 …While Also Showing Improvements Relative Other Citations

in 2017

5,546

4,546 4,458

3,657

0

1,000

2,000

3,000

4,000

5,000

6,000

2014 2015 2016 2017E

Total DUI/DUID Citations - Colorado

12.15%

14.63%

17.21%

14.38%

0.00%

5.00%

10.00%

15.00%

20.00%

2014 2015 2016 2017E

Marijuana as a % Total DUI/DUID Citations

Note: 2017 data is through October 4, 2017

Source: Colorado Department of Transportation and Cowen and Company

Note: 2017 data is through October 4, 2017

Source: Colorado Department of Transportation and Cowen and Company

COWEN.COM 45


Overdoses

Unlike other Schedule I narcotics, the risk from overdosing on cannabis is non-material.

“No death from overdose of marijuana has been reported.” – Drugs of Abuse |

A DEA Resource Guide: 2017 Edition

“Available evidence suggests that cannabis use is not associated with an

increased risk of all-cause mortality.” – National Academy of Sciences

Cannabis Review (2017)

“People have tried to ingest as many plant cannabinoids as possible and there

is no fatal overdose that has ever been documented in human beings.” – Johan

Marcu, Chief Science Officer, Americans for Safe Access

Underage Use

Research suggests that the earlier the age of cannabis initiation (among other factors),

the higher the likelihood of cannabis dependence; that is true of nicotine and alcohol as

well, as developing brains do not benefit from exposure to any regulated substances.

However, there is little evidence to suggest legalization increases cannabis use among

youth. In fact, data from Colorado points to the exact opposite:

NSDUH data through 2016 shows that in the two years since legalization

cannabis incidence rates among 12-17 year olds has fallen more sharply than

the national average (and interestingly, so has alcohol, tobacco and heroin use).

Figure 95 Outsized Declines in Youth Incidence in CO

12.6%

9.1%

7.2% 6.8%

6.0%

8.0%

10.0%

12.0%

14.0%

20

02

-20

03

20

03

-20

04

20

04

-20

05

20

05

-20

06

20

06

-20

07

20

07

-20

08

20

08

-20

09

20

09

-20

10

20

10

-20

11

20

11

-20

12

20

12

-20

13

20

13

-20

14

20

14

-20

15

20

15

-20

16

Past Month Cannabis Use (Ages 12-17)

Colorado United States

Dotted line denotes

adult-use legalization


COWEN.COM46


The Colorado Department of Public Health and Environment found of 17,000

middle and high school students in CO in 2015, past month incidence was 21%,

slightly below the national average (and down from 25% in 2009). And, the

percentage of teens that had tried cannabis remained well below 2009 levels

within two years of legalization.

Figure 96 CO Teen Use Below National Trends…

Figure 97 “Never Tried” Teen Trends in CO Improved Since 2009

17.0%

19.0%

21.0%

23.0%

25.0%

27.0%

2005 2007 2009 2011 2013 2015

Past Month Use Incidence

Colorado National

57.6% 57.7% 57.4%

60.5%

63.1%

62.0%

54.0%

56.0%

58.0%

60.0%

62.0%

64.0%

2005 2007 2009 2011 2013 2015

CO Youth - Never Tried Cannabis

Source: CDC, State Reports and Cowen and Company Source: CDC, State Reports and Cowen and Company

One of the reasons that we believe legalization of cannabis has little impact on teen

incidence, is that access does not seem to be a problem. According to the University of

Michigan’s Monitoring the Future Survey for 2017, 80% of 12th graders indicated it was

easy or fairly easy to obtain cannabis, compared to 78% for cigarettes (which generally

has a lower threshold age for purchase of 18, vs. 21 for legal cannabis).

Figure 98 Cannabis Is Generally as Easy to Obtain as Cigarettes and Alcohol

46.2 44.135.2

53.262.5 66.3 64.6

71.577.9 78.2 79.8

87.1

0

20

40

60

80

100

Cigarettes Vaping Device Cannabis Alcohol

% of Students Saying It Is Faily Easy or Very Easy to Obtain (2017)

8th Grade 10th Grade 12th Grade

Source: Monitoring the Future and Cowen and Company

COWEN.COM 47


APPENDIX

The Consumer Waterfall: Underlying Population Estimates

The benefits of aging cannabis consumer base are amplified by expected shifts in

underlying consumer demographics. Our absolute user estimates are all based on cohort

incidence, underpinned by Census estimates.

According to the U.S. Census Bureau’s 2017 national population projections through

2030, the overall 18-25 year old cohort is expected to be essentially unchanged (vs.

+0.5% CAGR in the prior seven years). Population growth among 26+ is also expected to

slow 40 bps: +1.3% 2009-2016 vs. 0.9% 2017-2030. In other words, 26+ is expected to

drive 100% of the population growth through 2030.

Figure 99 The U.S.’s Aging Population…

Figure 100 …Will Shift Age Distribution in the U.S.

Source: U.S. Census, U.S. Department of Commerce Source: U.S. Census, U.S. Department of Commerce

COWEN.COM48


First Use Underlying Analysis





Legacy (2016) States

First Use of Marijuana 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Adult-Use 1.1% 1.2% 1.3% 1.3% 1.4% 1.5% 1.5% 1.7% 1.7% 1.9% 1.9%

Medical Use 1.0% 1.0% 1.1% 1.2% 1.2% 1.3% 1.4% 1.4% 1.4% 1.5% 1.5%

Non-Cannabis 0.9% 0.9% 0.9% 1.0% 1.1% 1.1% 1.1% 1.1% 1.2% 1.3% 1.3%

Total United States 0.9% 1.0% 1.0% 1.1% 1.2% 1.2% 1.3% 1.3% 1.3% 1.4% 1.4%

Relative to Total U.S.

Adult-Use 19.3% 23.6% 24.2% 21.7% 21.8% 23.7% 23.1% 28.3% 28.1% 36.8% 36.8%

Medical Use 0.7% 2.3% 4.8% 4.2% 4.7% 5.3% 7.4% 6.1% 5.1% 3.3% 3.3%

Non-Cannabis -3.4% -5.9% -9.0% -7.8% -8.3% -9.4% -12.1% -11.5% -10.3% -10.0% -10.0%


Adult Use/Non-Cannabis Delta -22.6% -29.4% -33.2% -29.4% -30.1% -33.1% -35.2% -39.8% -38.4% -46.7% -46.8%

New (2018) States

First Use of Marijuana 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Adult-Use 1.0% 1.1% 1.2% 1.3% 1.3% 1.4% 1.5% 1.6% 1.6% 1.7% 1.7%

Medical Use 1.0% 1.0% 1.0% 1.1% 1.2% 1.2% 1.2% 1.3% 1.3% 1.4% 1.4%

Non-Cannabis 0.9% 0.9% 0.9% 1.0% 1.1% 1.1% 1.1% 1.1% 1.2% 1.3% 1.3%



Adult-Use 5.8% 10.6% 17.6% 15.5% 12.9% 18.3% 22.0% 21.0% 19.3% 19.4% 19.4%

Medical Use 0.1% 0.7% -2.3% -2.0% 0.6% -2.4% -2.4% -1.8% -2.3% -2.8% -2.8%

Non-Cannabis -3.7% -8.2% -10.5% -8.9% -10.0% -11.0% -14.9% -14.7% -12.3% -11.6% -11.6%


Adult Use/Non-Cannabis Delta -9.5% -18.7% -28.1% -24.4% -22.8% -29.3% -36.9% -35.7% -31.5% -31.0% -31.0%

COWEN.COM 49


Cannabis’ Impact on Alcohol Consumption

As we have detailed in our previous work, we believe that cannabis has, and will

continue to have, a negative impact on alcohol consumption.

Outsized switching among younger consumers. Among consumers ages 21-25

(as well as 18-25) we have seen steady increases in reported past month and

past year cannabis incidence since 2009. Within a year of that inflection, we

began to see steady declines in reported past year and past month alcohol

incidence.

Figure 103 Declines in Alcohol Use Accelerated…

Figure 104 ...While Cannabis Use Continued To Rise

Note: Ages 21-25


Note: Ages 21-25


Shifting Risk Perceptions. One of the key drivers of this category switching, we

believe, is the evolving risk perceptions among these younger consumers (18-

25) around alcohol and cannabis. Over the last decade, while risk perceptions

around alcohol consumption have been on the rise, they’ve fallen by roughly

50% for cannabis.

Figure 105 Young Adults View Alcohol Increasingly More Harmful Than Cannabis

Note: Ages 18-25


66.8%

67.6%

15.8%

20.2%

14.0%

16.0%

18.0%

20.0%

66.0%

67.0%

68.0%

69.0%

70.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Ca

nn

ab

is In

cid

en

ce

Alc

oh

ol I

nci

de

nce

Past Month Incidence (Ages 21-25)

Alcohol Cannabis

82.0%

82.6%

27.4%

31.9%

25.0%

27.0%

29.0%

31.0%

33.0%

81.0%

82.0%

83.0%

84.0%

85.0%

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Ca

nn

ab

is In

cid

en

ce

Alc

oh

ol I

nci

de

nce

Past Year Incidence (Ages 21-25)

Alcohol Cannabis

32.7%

37.0%33.7%

27.0%

36.7%

19.1%

10.0%

20.0%

30.0%

40.0%

5+ Drinks 1x-2x / Week Smoke 1x or 2x / Week

Perception of Greater Risk (in %)

2005 2010 2015

COWEN.COM50


National Substitution (All Age Cohorts). This pressure on alcohol is reinforced

by our multidecade analysis on per capita alcohol consumption in the U.S.

Looking at data since 2009, when this youth incidence divergence began, we

can see that on a national basis (for all age cohorts) we have seen a consistent

decline in per capita alcohol consumption, and an increase in reported cannabis

incidence. As well, it would also seem that cannabis category retention is on

the rise (as measured by past year incidence divided by lifetime trial).

Figure 106 Cannabis Retention Increasing…

Figure 107 …As Alcohol Looks to be In Another Contraction Cycle

26.0%

27.0%

28.0%

29.0%

30.0%

31.0%

32.0%

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

Past Year Retention (2009-2016)

-5.0% -3.0% -1.0% 1.0% 3.0% 5.0% 7.0% 9.0%

2009

2010

2011

2012

2013

2014

2015Past Year Cannabis Use vs. Ethanol Per Capita Consumption

Ethanol Cannabis

Alcohol

Down 4

Years

Cannabis

Up 6

Years

Note: Past Year Retention = Past Year Incidence / Lifetime Incidence


Note: Ages 12+ for incidence

Source: NIAAA, NSDUH, WHO and Cowen and Company

Survey Says: Consumer Substitute. Our own research in North America yielded

similar results in terms of consumer behavior. In the U.S. we find 64% of

consumers report a reduction in their alcohol consumption, while a greater 72%

of Canadians report cutting back. In both instances, over 30% of cannabis

consumers report “drinking a lot less.”

Figure 108 Cannabis Consumers Tend to Drink Less

Source: Cowen Proprietary Survey

28%36%

21% 17%

18%16%

32%31%

0%

20%

40%

60%

80%

Cowen Canada Survey Cowen U.S. Survey Cowen Canada Survey Cowen U.S. Survey

How Does Cannabis Impact Alcohol Consumption?

Drink the Same Amount Drink a Little Less Drink Moderately Less Drink A Lot Less

72%64%

Reduced Alcohol Consumption

COWEN.COM 51


Number of Binge Sessions per Month: Underlying Analysis

Figure 109 # of Binge Sessions Per Month

Source: CDC, U.S. Census and Cowen and Company

Figure 110 # of Binge Sessions Per Month



Weighted # of Binge Sessions Per Month 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Adult-Use 3.8 3.9 4.1 3.9 4.0 4.0 4.1 4.2 4.2 4.2 4.1

Medical Use 3.9 4.2 4.2 4.2 4.1 4.0 4.2 4.3 4.2 4.1 4.4

Non-Cannabis 4.2 4.2 4.4 4.3 4.2 4.3 4.4 4.4 4.4 4.5 4.7

Total United States 4.0 4.2 4.3 4.3 4.2 4.1 4.3 4.3 4.3 4.3 4.5


Adult-Use -6.3% -5.8% -5.6% -7.6% -3.9% -3.9% -5.6% -1.9% -3.4% -2.9% -9.1%

Medical Use -3.0% 0.9% -2.3% -1.0% -1.7% -3.1% -2.3% -1.4% -2.7% -5.6% -3.8%

Non-Cannabis 3.4% -0.3% 2.7% 1.7% 2.0% 3.1% 2.6% 1.5% 2.8% 5.1% 4.3%


Adult Use/Non-Cannabis Delta 9.7% 5.5% 8.2% 9.3% 5.8% 7.1% 8.3% 3.4% 6.2% 8.0% 13.3%

New (2018) States

Weighted # of Binge Sessions Per Month 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Adult-Use 3.9 4.2 4.0 4.0 3.9 3.8 4.0 4.0 4.2 4.0 3.9

Medical Use 4.0 4.2 4.3 4.4 4.2 4.1 4.3 4.4 4.2 4.2 4.6

Non-Cannabis 4.1 4.1 4.4 4.3 4.3 4.3 4.4 4.4 4.4 4.6 4.8



Adult-Use -4.5% -0.2% -7.0% -7.3% -6.7% -7.2% -7.0% -7.7% -3.4% -8.3% -13.4%

Medical Use -0.2% 1.7% 0.3% 2.8% 0.3% 0.4% 0.3% 2.7% -1.4% -2.4% 1.4%

Non-Cannabis 2.6% -1.9% 3.3% 0.6% 3.2% 3.3% 3.3% 1.0% 3.4% 6.6% 4.9%


Adult Use/Non-Cannabis Delta 7.1% -1.8% 10.3% 7.9% 9.9% 10.5% 10.3% 8.6% 6.8% 14.9% 18.3%

COWEN.COM52


Number of Drinks per Binge Session: Underlying Analysis

Figure 111 Binge Intensity (# of Drinks Per Binge)


Figure 112 Binge Intensity (# of Drinks Per Binge)



Weighted Binge Intensity 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Adult-Use 6.6 6.6 6.8 6.4 6.5 6.6 6.8 6.8 6.6 6.6 6.6

Medical Use 7.0 7.2 7.1 7.0 6.8 7.1 7.1 7.1 7.1 7.0 7.2

Non-Cannabis 7.4 7.3 7.4 7.1 7.1 7.4 7.4 7.5 7.4 7.3 7.4



Adult-Use -7.3% -8.8% -6.6% -8.1% -6.4% -8.4% -6.6% -6.1% -8.3% -7.7% -9.3%

Medical Use -2.0% -0.5% -2.0% -0.5% -1.7% -1.7% -2.0% -2.1% -1.9% -2.0% -0.9%

Non-Cannabis 2.6% 1.3% 2.5% 1.3% 2.2% 2.4% 2.5% 2.6% 2.6% 2.6% 1.8%



New (2018) States

Weighted Binge Intensity 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Adult-Use 6.9 7.0 7.0 6.8 6.7 6.9 7.0 7.0 7.1 7.0 7.1

Medical Use 7.1 7.3 7.2 7.1 6.8 7.2 7.2 7.3 7.1 7.0 7.3

Non-Cannabis 7.4 7.3 7.4 7.0 7.1 7.4 7.4 7.5 7.5 7.4 7.4



Adult-Use -4.2% -3.1% -3.6% -3.2% -2.9% -4.9% -3.6% -4.4% -1.7% -2.3% -3.1%

Medical Use -0.6% 0.4% -0.7% 1.1% -1.2% -0.3% -0.7% 0.0% -1.8% -1.9% -0.1%

Non-Cannabis 3.0% 1.2% 2.7% 0.6% 2.9% 3.0% 2.7% 2.4% 2.9% 3.3% 1.8%



COWEN.COM 53


Survey Says: Consumer Substitute. Cannabis’ negative impact on alcohol sales

was also evident in research from the University of Connecticut and Georgia

State University published in October 2017. The study found that alcohol sales

in counties that bordered counties with access to medical cannabis between

2006 and 2015 showed an average 13% decline in alcohol purchases.

Figure 113 Alcohol Sales Under Pressure in Legal States…

Figure 114 …With Falling Alcohol Sales Increasingly Evident over Time

Source: University of Connecticut and Cowen and Company Source: University of Connecticut and Cowen and Company

COWEN.COM54


Cannabis & Opioids: Additional Published Research

A 2017 study published in the American Journal of Public Health found that

access to adult use cannabis led to a “reversal” in opioid overdoses in 2015.

A 2017 study from the University of New Mexico showed a 47% reduction in

daily opioid dosages among consumers participating in their medical marijuana

program.

A 2017 study from the University of California, San Diego found that opioid

painkiller dependence and abuse fell an average 23% after medical cannabis

was legalized, and opioid-related hospitalization rates fell 13%.

A 2016 study from Columbia University (looking at 18 medical states from

1999-2013) concluded "we would expect the adverse consequences of opioid

use to decrease over time in states where medical cannabis is legal.”

A 2016 study from the University of Michigan found that patients who used

medical cannabis to control chronic pain reported a 64% decrease in their use

of Vicodin, OxyContin and other prescription opioids.

A 2016 study from the University of Georgia found that the average physician

wrote 12% fewer painkiller prescriptions for Medicare D patients than in states

where cannabis was still prohibited.

A 2015 study from the Rand Corporation suggested a 20% decrease in

admissions to treatment programs for opioid addiction in states that allowed

medical cannabis dispensaries.

A 2014 study published in JAMA Internal Medicine found that death rates from

opioid overdoses were 25% lower in states that had approved medical

cannabis.

COWEN.COM 55



COWEN.COM56



COWEN.COM 57



COWEN.COM58


Ticker Rating Price* Price Target

SAM Market Perform $194.35 $195.00WEED Outperform C$27.99 C$44.00BREW Outperform $18.60 $23.00TAP Market Perform $73.64 $85.00

Ticker Rating Price* Price Target

BF/B Market Perform $53.97 $57.00STZ Outperform $225.09 $265.00KSHB Outperform $5.02 $9.50LEAF.CN Outperform C$16.05 C$26.00

*As of 04/03/2018VALUATION METHODOLOGY AND RISKS

Valuation Methodology

Our valuation methodology is primarily based on Price-to-Earnings (P/E), followed byRelative Price-to-Earnings (vs. the S&P 500), Price-to-Sales (P/S) as well as Enterprise Valueto EBITDA (EV/EBITDA). In cases where GAAP net income includes large, non-cash items(e.g., restructuring charges), we may use non-GAAP EPS.

Cannabis:

Our valuation methodology is primarily based on DCF, followed by Price-to-Sales. In caseswhere GAAP net income includes large, non-cash items (e.g., restructuring charges), we mayuse non-GAAP EPS.

Investment Risks

The global beverage industry is subject to a number of potential headwinds. For alcoholicbeverages, heightened regulation and taxation are key risks, as is emerging access to legalcannabis. For non-alcoholic beverages, declining consumption, in particular for carbonatedsoft drinks, represents a key headwind, as do health and wellness concerns around artificialsweeteners. What is more, energy drinks have also come under scrutiny, which has resultedin softer demand, as well as litigation and regulatory risks.

Cannabis: Cannabis is an emerging industry and is subject to regulatory headwinds. Whileover 50% of the population is in favor of legalization, only a few states have thus farlegalized cannabis for recreational use and the product remains illegal at the federal level.Looking forward, much work and change still needs to occur in order for this industry torealize its full potential.

Risks Pertaining to U.S. Cannabis-Related Companies: If you are considering investing in aU.S. company that is connected to the cannabis industry, be aware that cannabis-relatedcompanies may be at risk of federal and/or state criminal prosecution. The Department ofTreasury has issued guidance that The Controlled Substances Act (“CSA”) makes it illegalunder U.S. federal law to manufacture, distribute, or dispense cannabis and cannabis-relatedproducts. Many states impose and enforce similar prohibitions. Notwithstanding the federalban, however, many U.S. states and the District of Columbia have legalized certain cannabis-related activities.

Risks Pertaining to Canadian Cannabis-Related Companies: In Canada, cannabis is anemerging industry and is subject to regulatory headwinds. While medical cannabis is legalin Canada, legislation has also been introduced to legalize adult-use sales no later than July2018. An initial regulatory framework has been laid out for the adult-use market, lookingahead, the category will be subject a number of potential headwinds, including taxes andrestrictions on from factors and packaging.

COWEN.COM 59


ADDENDUM

Analyst CertificationEach author of this research report hereby certifies that (i) the views expressed in the research report accurately reflect his or her personal views about any and all of the subjectsecurities or issuers, and (ii) no part of his or her compensation was, is, or will be related, directly or indirectly, to the specific recommendations or views expressed in this report.

Important DisclosuresThis report constitutes a compendium report (covers six or more subject companies). As such, Cowen and Company, LLC chooses to provide specific disclosures for the companiesmentioned by reference. To access current disclosures for the all companies in this report, clients should refer to https://cowen.bluematrix.com/sellside/Disclosures.action or contactyour Cowen and Company, LLC representative for additional information.Cowen and Company, LLC compensates research analysts for activities and services intended to benefit the firm's investor clients. Individual compensation determinations for researchanalysts, including the author(s) of this report, are based on a variety of factors, including the overall profitability of the firm and the total revenue derived from all sources, includingrevenues from investment banking, sales and trading or principal trading revenues. Cowen and Company, LLC does not compensate research analysts based on specific investmentbanking transactions or specific sales and trading or principal trading revenues.The Nielsen material contained in this report represent Nielsen’s estimates and do not represents facts. Nielsen has neither reviewed nor approved this report and/or any of thestatements made herein.

DisclaimerOur research reports are simultaneously available to all clients are on our client website. Research reports are for our clients only. Not all research reports are disseminated, e-mailedor made available to third-party aggregators. Cowen and Company, LLC is not responsible for the redistribution of research by third party aggregators. Selected research reports areavailable in printed form in addition to an electronic form. All published research reports can be obtained on the firm’s client website, https://cowenlibrary.bluematrix.com/client/library.jsp.

The information, opinions, estimates and forecasts are as of the date of this report and subject to change without prior notification. We seek to update our research as appropriate, butvarious regulations may prevent us from doing so. Research reports are published at irregular intervals as appropriate in the analyst’s judgement.

Further information on subject securities may be obtained from our offices. This research report is published solely for information purposes, and is not to be construed as an offerto sell or the solicitation of an offer to buy any security in any state where such an offer or solicitation would be illegal. Other than disclosures relating to Cowen and Company, LLC,the information herein is based on sources we believe to be reliable but is not guaranteed by us and does not purport to be a complete statement or summary of the available data.Any opinions expressed herein are statements of our judgment on this date and are subject to change without notice. The opinions and recommendations herein do not take intoaccount individual client circumstances, objectives or needs and are not intended as recommendations of investment strategy. The recipients of this report must make their ownindependent decisions regarding any securities subject to this research report. In some cases, securities and other financial instruments may be difficult to value or sell and reliableinformation about the value or risks related to the security or financial instrument may be difficult to obtain. To the extent that this report discusses any legal proceedings or issues,it has not been prepared to express or intended to express any legal conclusion, opinion or advice. Our salespeople, traders and other professionals may provide oral or writtenmarket commentary or trading strategies to our clients that reflect opinions that are contrary to the opinions expressed in our research. Our principal trading area and investingbusinesses may make investment decisions that are inconsistent with recommendations or views expressed in our research. Cowen and Company, LLC maintains physical, electronicand procedural information barriers to address the flow of information between and among departments within Cowen and Company, LLC in order to prevent and avoid conflicts ofinterest with respect to analyst recommendations.For important disclosures regarding the companies that are the subject of this research report, please contact Compliance Department, Cowen and Company, LLC, 599 LexingtonAvenue, 20th Floor, New York, NY 10022. In addition, the same important disclosures, with the exception of the valuation methods and risks, are available on the Firm's disclosurewebsite at https://cowen.bluematrix.com/sellside/Disclosures.action.

Equity Research Price Targets: Cowen and Company, LLC assigns price targets on all companies covered in equity research unless noted otherwise. The equity research price targetfor an issuer's stock represents the value that the analyst reasonably expects the stock to reach over a performance period of twelve months. Any price targets in equity securitiesin this report should be considered in the context of all prior published Cowen and Company, LLC equity research reports (including the disclosures in any such equity report or onthe Firm's disclosure website), which may or may not include equity research price targets, as well as developments relating to the issuer, its industry and the financial markets. Forequity research price target valuation methodology and risks associated with the achievement of any given equity research price target, please see the analyst's equity research reportpublishing such targets.

Cowen Credit Research and Trading: Due to the nature of the fixed income market, the issuers or debt securities of the issuers discussed in “Cowen Credit Research and Trading”research reports do not assign ratings and price targets and may not be continuously followed. Accordingly, investors must regard such branded reports as providing stand-aloneanalysis and reflecting the analyst’s opinion as of the date of the report and should not expect continuing analysis or additional reports relating to such issuers or debt securities of theissuers.

From time to time “Cowen Credit Research and Trading” research analysts provide investment recommendations on securities that are the subject of this report. Theserecommendations are intended only as of the time and date of publication and only within the parameters specified in each individual report. “Cowen Credit Research and Trading”investment recommendations are made strictly on a case-by-case basis, and no recommendation is provided as part of an overarching rating system or other set of consistentlyapplied benchmarks. The views expressed in this report may differ from the views offered in the firm’s equity research reports prepared for our clients.

Notice to UK Investors: This publication is produced by Cowen and Company, LLC which is regulated in the United States by FINRA. It is to be communicated only to persons of a kinddescribed in Articles 19 and 49 of the Financial Services and Markets Act 2000 (Financial Promotion) Order 2005. It must not be further transmitted to any other person without ourconsent.Notice to European Union Investors: Individuals producing recommendations are required to obtain certain licenses by the Financial Regulatory Authority (FINRA). You can review theauthor’s current licensing status and history, employment history and, if any, reported regulatory, customer dispute, criminal and other matters via “Brokercheck by FINRA” at http://brokercheck.finra.org/. An individual’s licensing status with FINRA should not be construed as an endorsement by FINRA. General biographical information is also available for eachResearch Analyst at www.cowen.com.

Additionally, the complete preceding 12-month recommendations history related to recommendation in this research report is available at https://cowen.bluematrix.com/sellside/Disclosures.action

The recommendation contained in this report was produced at April 03, 2018, 17:30 ET. and disseminated at April 04, 2018, 05:28 ET.

COWEN.COM60


https://cowen.bluematrix.com/sellside/Disclosures.action

https://cowenlibrary.bluematrix.com/client/library.jsp

https://cowenlibrary.bluematrix.com/client/library.jsp


http://brokercheck.finra.org/

http://brokercheck.finra.org/



Copyright, User Agreement and other general information related to this report© 2018 Cowen and Company, LLC. . All rights reserved. Member NYSE, FINRA and SIPC. This research report is prepared for the exclusive use of Cowen clients and may not bereproduced, displayed, modified, distributed, transmitted or disclosed, in whole or in part, or in any form or manner, to others outside your organization without the express priorwritten consent of Cowen. Cowen research reports are distributed simultaneously to all clients eligible to receive such research reports. Any unauthorized use or disclosure isprohibited. Receipt and/or review of this research constitutes your agreement not to reproduce, display, modify, distribute, transmit, or disclose to others outside your organization.All Cowen trademarks displayed in this report are owned by Cowen and may not be used without its prior written consent.

Cowen and Company, LLC. New York (646) 562-1010 Boston (617) 946-3700 San Francisco (415) 646-7200 Chicago (312) 577-2240 Cleveland (440) 331-3531 Atlanta (866)544-7009 Stamford (646) 616-3000 Washington D.C. (202) 868-5300 London (affiliate) 44-207-071-7500

COWEN AND COMPANY EQUITY RESEARCH RATING DEFINITIONS

Outperform (1): The stock is expected to achieve a total positive return of at least 15% over the next 12 months

Market Perform (2): The stock is expected to have a total return that falls between the parameters of an Outperform and Underperform over the next 12 months

Underperform (3): Stock is expected to achieve a total negative return of at least 10% over the next 12 months

Assumption: The expected total return calculation includes anticipated dividend yield

Cowen and Company Equity Research Rating DistributionDistribution of Ratings/Investment Banking Services (IB) as of 03/31/18Rating Count Ratings Distribution Count IB Services/Past 12 MonthsBuy (a) 460 62.33% 111 24.13%Hold (b) 266 36.04% 13 4.89%Sell (c) 12 1.63% 0 0.00%(a) Corresponds to "Outperform" rated stocks as defined in Cowen and Company, LLC's equity research rating definitions. (b) Corresponds to "Market Perform" as defined in Cowenand Company, LLC's equity research ratings definitions. (c) Corresponds to "Underperform" as defined in Cowen and Company, LLC's equity research ratings definitions. Cowen andCompany Equity Research Rating Distribution Table does not include any company for which the equity research rating is currently suspended or any debt security followed by CowenCredit Research and Trading.

Note: "Buy", "Hold" and "Sell" are not terms that Cowen and Company, LLC uses in its ratings system and should not be construed as investment options. Rather, these ratings termsare used illustratively to comply with FINRA regulation.

COWEN.COM 61


POINTS OF CONTACT

Analyst Profiles

Vivien Azer

New York

646 562 1351

[email protected]

Vivien Azer is a senior analyst coveringbeverages and tobacco. She joined Cowen in2014.

Brian Nicholas Velez

New York

646 562 1353

[email protected]

Brian Nicholas Velez is an associatecovering beverages. He joined Cowenin 2017 and is a graduate of ColumbiaUniversity.

Gerald Pascarelli, CFA

New York

646 562 1362

[email protected]

Gerald Pascarelli is an associate coveringbeverages. He joined Cowen in 2014.

Reaching Cowen

Main U.S. Locations

New York

599 Lexington AvenueNew York, NY 10022646.562.1010800.221.5616

Atlanta

3399 Peachtree Road NESuite 417Atlanta, GA 30326866.544.7009

Boston

Two International PlaceBoston, MA 02110617.946.3700800.343.7068

Chicago

181 West Madison StreetSuite 3135Chicago, IL 60602312.577.2240

Cleveland

20006 Detroit RoadSuite 100Rocky River, OH 44116440.331.3531

Stamford

262 Harbor DriveStamford, CT 06902646.616.3000

San Francisco

One Maritime Plaza, 9th FloorSan Francisco, CA 94111415.646.7200800.858.9316

Washington D.C.

2900 K Street, NWSuite 520Washington, DC 20007202.868.5300

International Location

Cowen International Limited

London

1 Snowden Street - 11th FloorLondon EC2A 2DQUnited Kingdom44.20.7071.7500

COWENRESEARCH COWEN INC.

COWEN.COM62


cannabis: $75b opportunity; category …...source: state reports, nsduh and cowen and company...

Documents