Sunday, January 28, 2018

Weed is Impacting Wine Sales


         Everybody Must Get Stoned ...

I spent last week at the Unified Wine Symposium in Sacramento. I've been coming to the conference since 1995; the first year ASEV and CAWG merged their separate conferences into one. Life was very different back then.


In 1995 you could find a really good bottle of Napa Cabernet for $15, Fed Funds were 8.5%, and marijuana possession was considered a Schedule I substance under the Controlled Substance Act of 1970 and prosecuted as such - right along with LSD, mescaline, and magic mushrooms.

The very next year - late in 1996, California became the first state to allow the legal use of medical marijuana, leading to the collapse of the Unified Conference and all wine sales as we knew them.

OK that last part's not true. Since that first year of Unified, wine sales in the US have experienced 20+ years of growth, even with medical marijuana coming on the scene. And Unified's Trade Show which back then didn't even fill up the first floor of the Sacramento Convention Center, now has two floors of trade show participants and a waiting list to get on the floor. Needless to say, the Unified Conference has thrived.

      They'll Stone You When You're Trying to Be So Good ...

The conference has gone so well that this year fourteen-thousand people came from all over the globe to Sac, selling out hotel rooms from Roseville to Davis. They came to gather and discuss trends, discover new equipment, meet with colleagues, connect with vendors, and after-hours, they wined and dined in the food establishments of River City. Most people were working hard and trying to be good.

But in an interesting sign of the times, this year I noticed 15% of the attendees skipped the food and wine scene altogether on Tuesday evening, and instead went up H street to the Capitol Park to form a large circle and everybody got stoned while listening to Bob Dylan.

OK that's an exaggeration. Those 2,100 people who went to the park brought plenty of munchies, notably Italian ice cream bars, cold pieces of juicy cantaloupe, M&Ms, and Dr. Pepper; the official beverage of Stoners International. But they didn't bring wine. Why would they want to drink wine when they are already going to be high?

      They'll Stone You When You're Trying to Make A Buck ...


Given all the good vibes being spread around this past week with a still improving economy, record stock prices, and increasing demand for wine - other than everyone taking time out from the sessions to watch the replay of the SVB Wine Report, it was quite interesting to see the water cooler talk was ignoring the aforementioned good vibes, and focusing instead on a somewhat dower Forbes article that came out last Monday with the headline: Alcohol Sales Dropped 15% in States With Medical Marijuana Laws. 

You have to admit that sounds ominous! Are we going to see weed kill off wine demand? Is that why we're seeing wine sales decelerateNo wonder I saw 15% of the attendees headed for Capitol Park last week!

Of course California has just opened the floodgates to recreational marijuana use this month. With this research focused only on medical marijuana legalization, what is that going to mean for California wine sales with recreational pot available to the masses? 

I asked around the convention and you can imagine the responses from the convention attendees I ran into. They ranged from, "That's bullshit," to "I'm going to buy stock in Italian ice cream bars," and of course, "I told you so."

      They'll Stone You When You're Walkin' on The Floor ...


This topic has been gathering steam since Washington and Colorado first legalized recreational marijuana in 2012 and to me it's not intuitive that pot is a pure substitute for wine. So wandering around the trade show the first day, it occurred to me I needed to get into the proverbial weeds and to the foundation of the discussion by reading the research

It's pretty routine that I can discover some level of bias in research when I look beyond the headlines, like cancer research founded by the Tobacco Industry, or poor research methodology that includes bad modeling or surveys that lack statistical significance. 

So I dug in and read the original report [PDF], then exchanged a couple emails with one of the report authors. I appreciate his engagement with me and while I was hoping to find something damning, I have to say from all I can tell, the research was set up beautifully with no detectable construction or funding bias.

They researched 2/3rds of the counties in the US so there was clear statistical significance, and the methodology was as good as anything I've seen

The only clear issue with the work was they used Nielsen data that's biased to the three-tier and omits some stores selling premium wine like Costco and Total Beverage among others, as well as direct to consumer sales. 

There is some channel shifting taking place at the same time of the research, but I'm not sure that's enough to toss the findings. Confused I started thinking maybe I needed to go to the Capitol Park with some Thin Mints and do some research myself? 

I went to bed that night and processed, then woke up with a question: How can the research conclude that there is a decline in wine sales when US wine sales were going up in 2016 - 2015 during the term of the research? Said a different way, how can you conclude wine sales are decreasing when sales are rising?


      They'll Stone You When You're Trying to Keep Your Seat ...

I had to sit down and read through the research again to answer that one since the compound growth rate of wine sold in the US was 3.5% per year during that time. It turns out the article title in Forbes wasn't quite reflective of the report finding. 

The way they came up with the link to impaired sales was by comparing counties where pot became legal, with counties where it wasn't legal. To the extent sales were positive in both counties, but the growth rate was less in the county after pot was legalized, they could conclude there was a link and quantify it. And before you go there, they also did a commendable job trying to filter out many other coincident factors.

Saving you a little reading time, here is the research conclusion:
We find that marijuana and alcohol are strong substitutes. Counties located in medical marijuana states reduced monthly alcohol sales by 13 percent, which is a consistent finding across several empirical specifications. ...When disaggregating by beer and wine we find that legalization of medical marijuana had a negative effect on corresponding sales by as much as 12 and 14 percent, respectively.

       They'll stone you and say it's the end ...



What do I conclude about the research? I may be on an island here because there has been serious research to prove weed and wine are strong substitutes, meaning those who consume both are largely ambivalent about the choice between the two. Maybe the math works out but well ... I'm just not there yet.

Using the high estimate of the total number of cannabis users in the US of just over 50 million people, 22% of 230 million total wine consumers could be consumers of both, only if all pot smokers were wine drinkers. So 14% impact still seems heavy to me, especially considering many were consuming pot before it was legal and arguably wouldn't change their behavior post-legalization. Those who consumed wine would still continue to consume some volume of wine.

While consumers smoke pot to get high, they don't drink wine, or premium wine in particular, with the same end goal.

So based on my own predisposed biased view of consumer behavior and backed by some prior research, to the extent wine is negatively impacted by weed, I think it's going to occur largely in lower price segments. I can make some sense out of a consumer coming home to blow off steam and rather than drinking a six-pack of Coors Light, decides instead to get high. 


The other research I just mentioned above drew a stronger correlation with beer versus wine and lower price versus higher, but this new work has me open to perhaps finding negative impact in wine as well in lower price points.

With that as a working theme, I asked the researcher if they considered price in their research and was told that price wasn't considered. So there you go. I guess we need more research.

To my thinking, premium and luxury wine is used in socially different ways and for different reasons versus weed. While possible at the fringe, I don't think we're going to see restaurants succeed that pair sinsemilla with salmon. Here's the acid test: Can you imagine dropping by a friends house with a 1961 Chateau Latour and asking, "Should we pop the cork or do a bowl?" Maybe it's me because I like wine, but that's not a real choice.

I think I'll switch off Dylan, tune in, open that Latour with a friend, put on Bruno Mars 24K Magic, and find some blue cheese, fig jam, and some crackers. Sadly, there goes my Italian ice cream bar investment.





What are your thoughts? Is Weed Hurting Wine? Are you buying Italian Ice Cream Bar stock?


  • Please join the site in the upper right hand of the page.
  • Offer your own observations and opinions for the benefit of the wine community below. 
  • Promote this discussion on your favorite social media platform please.

29 comments:

  1. Great assessment Rob. One of the best I've read on this particular debate over the last 5 years. I know the line about the 1961 Chateau Latour is tongue in cheek, but I think it's an important litmus test, and perhaps a better question we should consider (and one whose answer isn't so obvious) for people who are into both is "Can you imagine dropping by a friends house with a 2015 Cakebread Chard or 2012 Duckhorn and asking, "Should we pop the cork or do a bowl?". I think the '61 Latour is obviously no realistic tradeoff to consider, but for some recent vintages that are in the ultra premium and luxury space we might see some sort of impact as marijuana becomes de-stigmatized. There is a certain buyer, who was already buying and probably won't change the data much moving forward, but then there is the "soccer mom" demographic, the " 5'5" right handed woman shopper who the wine retail shelves are set for" that Gallo taught us about in the first month of training. She is the demo that I believe will be the X factor. Today I agree with you that Coors Light and Barefoot have more to worry about than a 61' latour, but there is a middle to that spectrum too. My opinion is that the fate of the impact pot has on the $13-$39 bottles will depend on just how "normal" vaping/smoking in typical public and private social settings becomes for the "alpha female" mom shopper. Thanks for another great piece of content Rob. Look forward to the next.

    ReplyDelete
    Replies
    1. Kloberdanz
      Thanks for logging in and the comment. You're right with the Latour I was making a point that quality wine and cannabis have different uses socially and going to the edge to make the point. There is a demand curve and I think this research does have me considering there can be substitution impact at the low end. What about the middle? I don't really know - except to say there are ends of the curve so I can't even define the middle today. The impact could be immaterial or perhaps modest, but IMHO of even greater impact is the shift in consumer preference that has to do with an aging consumer and evolving value/preference curves.

      Thanks again. Appreciate the excellent perspectives!

      Delete
  2. Perhaps, but those who drink 1961 Chateau Latour are not really the point here, I think. The main concern is more about those who drink, "The Prisoner."

    ReplyDelete
    Replies
    1. TCarl - If there is a preference curve with substitution effect which is zero at the luxury end, and something at the cheap end, the discussion really is about what that looks like in the middle. As I noted above - that's where this research fell short. It found pot was a strong substitute for alcohol but didn't look at price which I believe is the differentiator. More work needed there.

      Delete
  3. Well said, Kloberdans. I also believe that we are severely underestimating the impact that edibles will have on the market. This will soon make up a substantial percentage of overall cannabis sales and is more widely accepted in social settings.

    ReplyDelete
    Replies
    1. Was recently talking to a market research firm about Gron, a craft cannabis chocolate maker in Portland, and without me saying anything about wine, this firm tossed out the idea of consumers bringing some fine THC and/or CBD-infused chocolate to a dinner party instead of a nice bottle of wine....

      Delete
    2. Thanks Marshall and Pete. Today I believe firmly that premium wine and cannabis aren't good substitutes. People aren't likely to come to a party (social sharing) and keep to themselves with edibles or anything else. The social risk is probably too great. Over time that could evolve to some extent but today, premium wine isn't consumed solely to get ripped. There is a social and food component that plays, and the buzz isn't the whole sale, while with pot, it's is all about the high. We'll see if that changes.

      Delete
  4. Your assumption is that all weed is low-end. Today's weed game is much different than what you might recall. There are now multiple tiers from low-end to premium, just like wine. Plus your underestimating cannabis users affinity to take a THC edible with that 1961 Latour. Some would argue it enhances the experiences as there senses are heightened. Thanks for the great blog!

    ReplyDelete
    Replies
    1. Anon 10:13 - I'm not assuming all weed is low end at all. It's clearly being grown for THC content and has more quality control today. I'm only suggesting that price of wine is a component that impacts the sustainability of wine.

      Delete
  5. I spent my career in the restaurant business and then in wine sales. I'm still an almost daily consumer of wine and a frequent marijuana consumer as well. My wife hates the smell of pot so I keep it in my wine room and the ritual is to go there with any like minded guests and burn one while picking out the wine we have with dinner. No 61 Latour unfortunately, but no Prisoner either.

    ReplyDelete
    Replies
    1. Kent - Your point is one of those I was making. The impact attributed to medical marijuana seems too extreme. Consumers of both products don't necessarily take a binary approach. Likely if they were fine wine consumers and picked up marijuana when it became legal - still continued to drink good wine.

      Delete
  6. I won't try to fool anyone. I get a buzz from wine. But. Having a choice between pot and wine is a no brainer for me. I like wine with food. I like pairing an everyday inexpensive or a rare beautifully balanced wine with a nice meal. Smoking a doobie or eating a chunk of pot infused chocolate just isn't going to do it for me. For those that just want a buzz maybe pot is the way to go...I think time will tell but I do not see pot as a wine substitute. By the way Rob...where did you get that pic of the Coors Light and that big beautiful bud ?:)

    ReplyDelete
    Replies
    1. Paul - thanks for weighing in. I of course see the uses for each as pretty distinct. Sure we can pair pot with food, but really you are just satisfying cravings versus pairing anything. More important, the social uses are pretty different.

      This research has me more open to some impact at the low end, but limited impact at the luxury end.

      Delete
  7. I always thought cigar dinners were silly, and I would never try to match pot with a meal. Wine and food pairings are one of the great delights in my life. Now, I am not a huge volume marijuana consumer, but I don't thinl the two are mutually exclusive.

    ReplyDelete
    Replies
    1. I don't see them as mutually exclusive at all. If they were, they wouldn't be any kind of substitute. You can clearly have both or either, but it's hard for me to believe someone who was a wine lover and could have all the pot they wanted for free, would stop consuming wine.

      Of course, the devil is in the detail and the interesting research will include more delving into the impact on recreational marijuana and price of wine/beer.

      Delete
  8. As a side bar - I had a drive-time spot this morning on KABC in LA where the host talked about a restaurant owner who noted that younger consumers were using weed to pre-charge so they didn't have to buy as much wine with dinner.

    That indeed does make some sense and points to a level of substitution, but to me that's more about the argument price pays in the substitution effect - to my OP above.

    Consumers are getting very fatigued by paying 3-4x on a bottle of wine at the restaurant.

    ReplyDelete
  9. I work for a winery located in WA. I have recently seen some proprietary data on LDA (Legal Drinking Age) consumption in states where marijuana is legal - and the impact of legal weed is significant both on that state's LDA consumption index (vs. rest of US) and average amount of $ spent on alcohol. The other area that are increasingly affected is labor: if you are a migrant farm worker in a large wine producing state, are you going to collect your $10/hr picking grapes, or your $25/hr picking buds? Additionally, if you are a liquor store owner and can afford to pay $12-$13/hr for workers - it's hard to compete with the marijuana industry who is paying double that rate in hard cash.

    ReplyDelete
    Replies
    1. Velouria6 - Thanks for joining in from WA!

      There is more work that needs to be done for sure. The anecdotal evidence that flies around is hard to rationalize. For example, another colleague just emailed me data he has on tax based data on servings in Colorado since legalization. Adjusting for population growth, there was no change from when it became legal.

      And use MJ growers have a better deal for farm workers. Working in vineyards is better than row crops, but labor is one place cannabis and wine don't meet up.

      Delete
  10. I was mostly piping up to provide anecdotal evidence of my consumption habits, not disagreeing with your point. As to substitution, if you show up with a 61 Latour, I'm skipping my martini as well as my bowl. Prisoner, I might need both.

    ReplyDelete
    Replies
    1. This is where I wish this platform had emojis so I could put a LOL one up.

      Delete
  11. As a manager at a winery in CA, I have noticed a large decline in our millennial demographic. While our main base are aged 45-70, we used to have a decent amount of younger traffic as well. Now it's mostly seen for wedding or birthday parties. With the recent legalization here in CA, it stands to reason we would lose traffic, the main question I have is:
    Will this current level of impact continue to increase and for how long? Or will it begin to taper off/decline as the novelty of marijuana has worn off?

    ReplyDelete
    Replies
    1. I should say "novelty of legalization has worn off"

      Delete
    2. Anon 2:55 - I mentioned above that a friend just passed over to me solid data from Colorado showing no changes in servings since pot was made legal there. So careful on attribution for the present. I might add, information that I have from a store in Oregon is that the boomers have returned to MJ once legal, and the dispensaries are just as likely to have a 72 year old as a 27 year old come to purchase.

      Delete
  12. Rob,

    Great post. Much better than I did on my blog. One important thing to note about the study, though: it looked at consumption after medical marijuana legalization (MML), not recreational legalization. It may very well be that going from a total ban to MML is much more of a shift than going from MML to totally ending prohibition. This could have important implications.

    ReplyDelete
    Replies
    1. Gabe - Moving to recreational laws could have another down shift. I struggle with how much that might be and suspect it will be minor. Hope I'm right!

      Delete
  13. Rob, great post. The part that just doesn't add up to me is that there was this dramatic drop in wine sales *only after* the state legalized marijuana. Doesn't that presume that the cannabis behavior changed significantly after the state opened for pot?
    Here are some anecdotal observations from Colorado, where I've been both before and after cannabis prohibition was lifted:

    * Pot consumption didn't seem to change much after legalization, especially for my generation (Gen X) and Boomers. My friends were already smoking pot - it just got a lot more convenient to get it.
    * Pot prices are down substantially today compared to during prohibition.
    * There's still a stigma attached to pot. it's not common at parties or most social situations.
    * I haven't noticed any less wine consumption at all among the folks I interact with.
    * Obviously there's only so much disposable income, and choices need to be made. My speculation is that younger folks probably are spending more of their disposable income on pot now compared to before legalization. That would logically come at the expense of alcohol.

    In general, I just can't wrap my brain around a 15% reduction (which is substantial!) in wine because of pot. those numbers don't jive with anything that I've observed here. What am I missing?

    ReplyDelete
    Replies
    1. Jeff - thanks for logging in aND posting. Since I wrote the blog, I've seen two sets of data from State government taxx data that show since recreational pot was approved in the states of WA and Colorado, there has been no change to wine consumption.

      I appreciate the rigor of the research noted in this post, but it's hard to see how medical MJ could have the kind of impact noted. In fact, if it did, wouldn't recreational pot cause even more of a decrease?

      Apparently the science isn't settled just yet.

      Delete
  14. I enjoyed your article and it really had me thinking about the before and after effects of MML implementation and the market forces of a competing product.

    I read the original report back to front, and I agree with the author, their analysis shows that the implementation of MMLs is correlated with decreased alcohol sales; but you hit the nail on the head and I believe that the tipping point for causation depends on the price of the alcohol that you are substituting for. While gross sales information for wine and beer was the primary focus in the analysis along with timing and policy variables, the analysis does not take into account all exogenous factors or correct for omitted variable bias. Put simply, price is integral for economic behavioral modelling; MML's in general will have a huge impact on supply side economics for marijuana. Your photo of the Coors light going head to head with a huge bud so perfectly sums this up graphically. I'm sure many people now stop to consider their choices on a Friday night given the economics of a 12 pack "Taste from the Rockies" versus a Bob Marley baseball bat...

    An interesting aspect I noted with the report's analysis was depicted in the whisker plots of sales after the MML became effective. While there is a downward trend or more negative parameter effect in both beer and wine, the distributions become more spread (the variance is increasing), this tells me that the standard deviation is increasing, this happens around 8-9 months after MML's were implemented and the variance further increases with time. As I said, I agree with the author of the report, but it seems there is something changing in the underlying population; maybe the novelty of legal marijuana is running its course. Either way, thank you Rob for your article on this interesting topic.

    ReplyDelete
    Replies
    1. Sorry for the delayed response Jesse. I appreciate your well-thought comments.

      Delete

Please sign into the community to post. Common-sense guidelines apply: Disagree with author but offer your own thoughts. Disagree with other posters but please attack the post versus the person.

Flaming, spamming, off-topic posts, advertising and offensive posts that would not be suitable for work will probably be deleted. Drunken posts will be forwarded to your mother.