Tuesday, May 10, 2011

'Growth of Session Beer'? Not Yet, Friends.

I've got some data for you today. Good, hard evidence. I was going to title this post "Decline of the Session Beer," because that's what it's going to show. But then I saw the Boston Globe blog headline, "The steady growth of session beer." That was based on this Advertising Age article, "The New Drinking Session: How Craft Brewers Are Drawing in More Consumers."

Not so fast, folks.

Are brewers really drawing in more drinkers with low-alcohol beers? Is session beer really growing in America?

That's what I've been trying to find out, more or less fed up with anecdotal evidence and wishful thinking. I think I finally have a fairly definitive answer: a big, fat NO. That's despite all the buzz, the high hopes of drinkers like myself, new initiatives like Chris Lohring's Notch, and campaigns like Lew Bryson's Session Beer Project.

The data come from Ratebeer.com, thanks to the patient efforts of Joe Tucker and his crack team of site administrators. Whatever else you love or hate about Ratebeer, it probably has the most comprehensive database of beers available over the past 10 years.

These numbers refer to beers added to the Ratebeer database. Generally speaking, that means newly available beers, but they can also be beers that were available for a while but not added to the site until later. The data set begins in April 2000. Older beers that were no longer commercially available at that time are theoretically excluded, since by the site's rules they would not be "rateable." Thus, to say it plainly, these are commercially available beers from the past decade. I'm not saying the data are perfect, but they're pretty damned good.

Also, because we are talking about session beers in America, these are only U.S. beers.

By year: Total number U.S. beers added, number of them at 4.5% abv or lower, and percentage of total at 4.5% or lower.
2000: 1978 222 11.2% (note: partial year, data set begins in April 2000)
2001: 1223 107 8.7%
2002: 2638 188 7.1%
2003: 4523 436 9.6%
2004: 6395 615 9.6%
2005: 4401 318 7.2%
2006: 4555 263 5.8%
2007: 5053 262 5.2%
2008: 5668 350 6.2%
2009: 7860 445 5.7%
2010: 8796 419 4.8%
2011: 3419 152 4.4% (partial year, through the first week of May)

My assertion is simple: If we were truly seeing the growth of session beer as a category, we would be seeing more session beers available in the marketplace. The opposite is true: The rate at which they are entering the marketplace is currently decreasing.

Surprised? So was I.

Session beers will need more than buzz if they are to become a real trend. In my view, they will also need to make money. For more on that issue, stay tuned for my article in an upcoming issue of Draft magazine.

*I was going to make a handy line graph for you to illustrate this, but I frankly don't have time to learn how. The numbers speak for themselves, but if anyone wants to take a few minutes and plug these numbers into a graph, I'd be much obliged.


  1. These are interesting numbers, and certainly suggest it's not as simple as the Advertising Age story, but I think I disagree with the argument that because the percentage of new session beers on Ratebeer is declining, so is session beer consumption.

    A brewery can release multiple big DIPAs and be assured of selling out a small run, no matter the price. Indeed, big beers keep, so there's little risk in releasing a new DIPA in bottles and selling them at high price points. Part of what distinguishes a session beer, though, is its affordability. So while it may be a better long-term, sustained option for a brewery, it makes a lousy short-term offering thanks to the high time and materials costs of launching a new beer.

    So, looking at sheer volume and percentages may well not tell the full story.

    Also, while I believe Ratebeer has good data, there's no question that the user base skews toward big beers; if any type of beer would be underrepresented, it's session beers. That said, I think the issues with this study are more on the side of content validity than on the side of sample purity.

  2. Forgot to add: There are holes that can be poked. Poke away.

    For example: I went with the 4.5% limit as more indicative of the session beer trend. In my view 5% is mainstream strength. But if there's a real clamor to see the same numbers for 5% and below, I'll pester Joe T. again and ask for them.

    Also: It's worth looking at Ken Weaver's analysis from last year on rising abv levels.

  3. "but I think I disagree with the argument that because the percentage of new session beers on Ratebeer is declining, so is session beer consumption."

    I don't think I made any assertions about consumption. I'll concede that it's possible that more people are drinking from the already available session beers in greater quantity. However, I think it's more likely that if that were in fact happening, the brewers would be responding by making more session beers available. That is apparently not happening.

  4. That's fair enough, and I am tempted to agree, but I think it's tough to measure decline by product mix percentage in an industry still in a relatively steep growth phase.

    A couple filters could help. For example, given the vast disparity in how well session beers keep, it would be interesting to see how draught offerings have changed, or how year-round offerings have changed. Heck, even measuring sales volume would be better. In other words, anything that could eliminate one-off brews that don't necessarily reflect overall demand for would give us a clearer picture.

  5. Greg: Very good points all around. In my mind sales volume would be the second-best metric, right after the selfish one of what's available to me when I walk into my favorite beer shop or taphouse.

    Another way could be to pick a large sample of bars and restaurants and track their beer lists over time, including abv and sales. But that would be quite a project, better suited to marketing/industry types who gets paid more than I do.

  6. Glass half-empty, glass half-full. I look at those numbers and see the number of new session beers in 2009 and 2010 as the 4th and 2nd highest in the past ten years, and 2011 on-track to do the same. I'm also questioning the significance of new intros as a measure of popularity for session; these are beers that sell in volume, not limited edition woo-woo beers. Yards, for instance, as you know, sells over 50% of their output at 4.5% or under...yet those are the only session beers they make out of their whole portfolio. This is one facet, and in the perspective of how session beer is likely produced and consumed, maybe not the most useful one.
    That said...a reality check is ALWAYS a good thing! Right now I'm looking at all the excitement as a validation for brewers who've been thinking about doing session beers. Let's have a look at the numbers next year. I'd love to track 2011 month-by-month.

  7. There are also small remote breweries, such as ours, that make sessionable one-off's, and also seasonal regulars, that don't show up on ratebeer.

  8. Funny you should post! The photo above was taken at your pub last August. The Wotcha bitter, I think.

    Brewers Union and five of its beers are on RB, but your point is solid and troubling. There must be a lot of beers that never get rated, though I think they'd be a small percentage overall. A bigger problem for the data is if those unrated beers lean toward low or high alcohol... And we could suppose that missing beers would lean toward the low-abv side, since many typical raters would not find them as interesting.

    That is a problem.

  9. A few thoughts. I've had 7 styles out on draft or cask only that have never been reviewed, so maybe reviewers are not draft people, or maybe session beer fans are not reviewers. Don't know, but I think it's the later.

    I also think the "trend" spotting may be premature. There is still much work needed in terms of education and acceptance. And brewers who jump on the bandwagon with 5% and up "session" beers are not helping. Sorry if that offends, but the easy part is using the word, brewing it is the difficult task.

  10. Chris: You could be right, about session-oriented drinkers being less likely to review. I can say that most of the tickers/note-takers/reviewers/raters I know are fairly non-discriminatory about what they try.

    How often do brewers add their own beers to the RB database, rather than ignore the whole thing or just rely on drinkers to do it for them? I wonder. I have known some brewers to add their own.

    I'm going to continue acknowledging that the data are imperfect, but... I'm also going to stick to my guns, as far as those numbers being meaningful. I believe they are.

    To risk repeating myself, I believe that if there were really a trend toward more session beers, then we would see more session beers being added to the not-totally-comprehensive-but-well-populated RB database. We are not seeing that, yet.

  11. You may be correct about the decline, and that's why those of us who would like to drink more than one beer and stay upright are making so much noise about session beers.

    I would raise the caveat of selection bias in your database. It's a lot easier for the craft beer enthusiast to get enthusiastic about the latest hop bomb or velvety dark imperial stout or double whatever. Unfortunately, it's the best proxy we have for such valuable information as volume or revenue.

    My hope is that we are at an inflection point in which a long term decline is finally reversed this year. We're certainly talking about it more!

  12. Pam,
    Your last point is, FWIW, congruent to what we saw in bourbon about ten years ago; a slowing decline that coincided with a LOT more press talk about it...and then an upturn. Looks a LOT like that, actually. The crush of new brands didn't start till there actually was an uptick in sales of the established brands, though...like we've been seeing in Irish whiskey. Just a thought.

  13. You cannot draw any meaningful conclusion about the size/growth/decline of a market solely from the number of new launches into it, particularly as recorded on a database that is not guaranteed to be 100 per cent comprehensive. So your entire argument fails, I'm afraid.

  14. "You cannot draw any meaningful conclusion about the size/growth/decline of a market solely from the number of new launches into it, particularly as recorded on a database that is not guaranteed to be 100 per cent comprehensive."

    "Market" to me means sales. Obviously these data cannot track sales at all. Instead I was aiming to track a purported trend. It is no great leap to assert that American brewers establish and/or follow trends by launching new beers. In fact, that is what they have always done.

    So no, Martyn, with great respect for you, my argument does not fail.

  15. Given that the ratebeer sample may be biased because ratebeer raters as well as beer sommeliers and cicerones are strange guys who seem to be inclined toward beer extremism, I would like to share my beer experience in Brazil to illustrate a theoretical argument that may explain why extreme beers are becoming more popular (and session beers more difficult to find).
    Brazil has experienced in the last three years or so a boom in the craft brewing industry. For a country that was dominated by pale, tasteless lagers (as everywhere else in Latin America) Brazil now has a growing (but still small) range of good beers either produced in Brazil or imported from Germany, Belgium and more recently USA.
    My point is that the big producers always argued that the presence of pale lagers (most bellow the 4,8% ABV range) was a demand from the consumer because Brazil has a sunny, warm weather most of the year that is not proper for consuming European style beers that are maltier and stronger than most Brazilian beers. That said, it’s amazing to see how the strong Belgian ales and extreme American beers have become the synonym of quality amongst Brazilian “beer experts”. Even the German beers are much criticized by those experts because they judge the Rheinheitsgebot too restrict and German brewers uncreative. How all these translate into more extreme and less session beer available? Well, many of the newly created craft breweries in Brazil are producing dubbels, tripels, quadrupels, imperial stouts that sold for 30 dollars a 0,6 liter bottle, weizenbocks refermented with Champagne yeast that sold for 40 dollars a 0,75 liter bottle, and so on. Many of these beers, designed to be copies of Trappist and Abbey Belgian ales are much more expensive than the original ones. So what is the point in creating these copies? I would like to see more session beers, affordable and available, than spending my hard won duckets in expensive copies of Belgian style beers.

  16. The problem is, you are looking at it from the perspective of a researcher that is using his own observations and a site that gives you statistical data that is incomplete.
    How many new breweries? Of those new breweries, how many beers have have they released and what is the spread of ABV. I see a new brewery launch four or five beers, and yet one might be low. How many breweries have had a lower ABV beer in the arsenal and only recently have seen a greater demand for it, or really started to use it. If a brewery like Sam Adams who releases all sorts ABV's suddenly releases one low ABV Rustic Saison and sees an up tic in sales of some others how is that quantified? Percentages show a decline but they would see an increase in volume. If breweries across the country have a low abv and it suddenly starts to pick up sales, who tracks that ground swell. You certainly don't. As a retail owner I see this trend in the selections being presented and requested. As a frequent bar goer I see this in the selection at the taps. You don't have enough data to make an educated analysis. You got so many holes in your sieve you should just call it a hoop.