Beeronomics: The Novelty Curve, Economies of Scale and the State of Craft Brewing

Whew, how's that for a title?

Over at Beervana, Jeff Alworth has blogged a couple of times recently about some new data purporting to show the health and growth of the craft brewing industry.  He discusses how the data reveal an interesting fact about the industry: there are a few big players (Boston Brewing, New Belgium, Sierra Nevada, Full Sail, Deschutes) and a TON of very, very small breweries.  He and I have also had conversations about the overall trend.  Craft beer cheerleaders look at the remarkable growth in market share over the last decade and the still small share in general to suggest that the future is bright and that there is no reason we should not expect craft beer to continue to grow both in overall volume and in the number of breweries. I am skeptical.

The model I have in my head of the craft beer industry in the US is one that is mainly characterized by a tension between two forces: novelty and scale.

The first comes from the natural desire of consumers to consume variety. Most of us don't eat the same meal for dinner every night and we tend to get tired of any one meal we eat too often.  But the variety desire at the market level can be also arise because of static and different tastes. Finally, tastes can evolve - consumers get used to hoppy beers, for example.  So dynamic tastes, where consumers of beer find their tastes evolving and are always looking for new and interesting experiences, can also drive the market demand for variety.  All this leads to the advantage new breweries have in brewing new styles reach customers that were previously badly served because no beer was available that matched their preferences well, or to consumers looking for new things, or to consumers looking for more adventurous drinking experiences.   Jeff has coined this the 'novelty curve,' and suggests that new breweries have an advantage over established ones because of new, fresh products. [Ninkasi comes to mind]  Established breweries, like Bridgeport for example, have a more difficult time trying to keep a standard line-up selling while at the same time satisfying the desire for new tastes.

The second is a natural aspect of many industries - the larger the volume of the firm, the lower the average cost of producing the product. In brewing the size of the brew house, the ability to have your own bottling line, efficiency in distribution, number of taps, brand recognition, and many other aspects make brewing an industry that is subject to significant economies of scale.  What this means is that larger breweries can undercut smaller ones, all else equal.  [My colleague, Vic Tremblay, has estimated economy of scale effects for the large industrial breweries up to about 23 million barrels, if memory serves]

Because of these two forces and how they interact with the craft brewing industry, I expect that the industry will always be characterized by a lot of volatility.  I think that there will be some stable mainstays (the Deschutes and Full Sails of the world) that have the scale advantages and try hard to keep up a line of unique, one-off, specialty beers to feed the novelty curve.  And a lot of churn (in economics-speak) among the smaller breweries.  I think the brew-pub model is the way to go here as a stable revenue stream from food and an in-house captive audience for the beer can help tame the vagaries of a fickle consumer base.  

All that is to say is that I don't think the number of breweries continuing to increase is a sustainable path and I expect another round of shake outs in the market, unless small brewers can establish a successful restaurant operation.  And in general, I think we can always expect breweries to come and go regularly.  But this is not necessarily a bad thing - the better breweries, those that brew good beer, are well run businesses, market well, etc. will stick around and new start-ups will keep offering new, interesting beer.

This doesn't mean that the overall market share of craft breweries won't continue to increase, but as this increases it may reinforce the economies of scale advantages of bigger craft breweries.  That said, mainstays like Boston Brewing have seen sales stagnate, while lots of little breweries have opened, so we may be in a period where the variety motive in the market is ascendant.  The good news in all of this is that the market should always be interesting from here on out for this observer - I think craft beer is here to stay.

Note: the picture is of Hangar 24 craft brewery in Redlands, CA, where my brother is a brewer.  Check it out if you are in the LA area.

Beeronomics: Back to the Future?

The craft beer cognoscenti often talk about how the new craft beer movement in America is a return to the past where breweries were decidedly local (and perhaps reflected local tastes though most beers were pilsner style beers imported by german and eastern european immigrants).  I agree that the local brewery is a return to the past, but the localization of breweries in the late 19th and early 20th centuries was largely a consequence of the difficulty of transporting a bulky and heavy product.  The modernization of transport, especially post WWII, in my mind at least, was a major factor in the consolidation of brewing which was able to exploit economies of scale in all aspects of the industry.  But of course you loose something in the process: variety (such as it was).  

But this supposes that there was a lot of variety in local markets prior to the great consolidation.  I suspect not, for the very same reason that led to the creation of local breweries: transportation was costly.  On most local shelves in the US in the early 20th century, I suspect, was a choice of only a few locally produced beers. 

So really then, we are currently in an unprecedented era where we now have a dizzying array of beer choices available to us in our local stores.  I used to buy Rogue in Ithaca, NY, and in Portland I can find obscure Belgian beers, microbrews from all around the US, and tons of special brews.  What does it all mean?  Well, for one, we can identify regional tends in styles and flavors more readily and as producers try and distinguish themselves in a crowded marketplace the incentives are probably to try and produce more and more unique beers.  

Which is the point of this rumination: the rise in specialty brews (things like Deschutes' Dissident, Full Sail's Slipknot, etc.) is part and parcel of this newly crowded marketplace.  I think breweries are finding it ever more important to create some buzz and attention through the creation and marketing of these brews.  It is a wonderful trend for the beer enthusiasts, but is also a sign of an industry that is still evolving.  I think a lot about where it will all settle (e.g. how many breweries can the market sustain), but I suspect that this is an industry that will always have a lot of churn - many new breweries starting up and many old ones closing.  This is not necessarily a bad thing, it means lots of constantly changing variety.  Or perhaps the big brewers will successfully capture the market by introducing their own micro-style brews, it all remains to be seen and is pretty fascinating.  

Okay, so in the final analysis this post is pretty aimless, but this is what happens to someone shut in for over a week with sick kids.  Hopefully it makes a little sense.

Beeronomics: Specialization and Product Variety

Last Friday I had the distinct pleasure of having lunch with John Harris, brewmaster at Full Sail and the man responsible for, among many other amazing beers, slipknot Imperial IPA (shown here - go get some, it is fantastic). I learned many interesting things from John about the beer business including learning about an outfit called 'Microbeer Source.' A number of very small bottling breweries use Microbeer Source to bottle their beer, including Full Sail for the 'Brewmaster Reserve' line that is brewed at the Pilsner Room in Portland. Microbeer source is, you see, a mobile bottling line which brings the line to local breweries and bottles their beer for them. Upon researching this, I discovered that this is very common in the wine industry, which makes sense, the wine industry is mostly made up of many very small producers whose product is needs to be bottled in order to send to market. Bottling lines are pretty expensive, take a lot of space, and without this kind of specialization, it would be hard for small producers to survive. Economists often talk about such industries in the context of "natural monopolies," industries where there are such high fixed costs to begin production, that the market can only support one producer (which means having large economies of scale, but in a particular way).

In this context, if all small wineries had to have their own bottling lines, many would not be able to cover the cost of the line from the sales of their wine and would, therefore, not exist. There would not be monopoly, but the number of wineries would be drastically reduced. This is the same, albeit on a smaller scale, in the microbrewing industry. It is unlikely that breweries like Roots, a small Portland outfit (whose "Woody IPA" is fantastic and available in bottles down here in the sticks thanks to Microbeer Source) would be able to sell in bottles without such a bottling service available to them. So, through specialization of tasks, we can not only improve efficiency (as in Adam Smith's archetypical pin maker story), but in these cases, we can improve product variety as well. For what this means for consumers of beer and wine is not just lower prices for the increased efficiency that specialization brings, but many more choices available to them when they wander into their local supermarket or beer and wine store. Ah the wonders of economic organization...

P.S. John raised an interesting question, to which I have a number of plausible answers, but I will ask it as an open question first: why are just about all of the major bottling breweries in Oregon today the very same ones that were here 15 years ago? In other words, why have there been no new entrants into the industry in the last 15 years (with the exception of Terminal Gravity)? Ideas?

Beeronomics: Oh Happy Day!!

A distant uncle of mine once said:

Be not the slave of your own past. Plunge into the sublime seas, dive deep and swim far, so you shall come back with self-respect, with new power, with an advanced experience that shall explain and overlook the old.

-Ralph Waldo Emerson

This just about sums up how I feel about craft beer. The day I first tasted a Hair of the Dog 'Fred', everything I ever assumed I knew about beer was changed and I have never looked back (thanks Alan).

Which is why today is such a good day. Rare indeed is the day as good as the one in which you are able to get your hands on two bottles each of Full Sail's Top Sail and Deschutes' The Abyss. I now possess two of the finest examples of the brewmaster's art. These are limited releases and don't last long. Whispers in my ear from two knowledgeable sources (thanks Jeff and John) began my quest which ended quickly at Corvallis' University Market (in case you would like a weekend as good as mine -Portland is runnig out of Abyss quickly today). They are not cheap but worth every last penny. Go. Buy. Enjoy.

Which brings me to economics once again. I have been thinking a lot about preferences, experience goods, variety, branding and the Oregon craft beer industry lately. Some of this I have blogged about before. Seems to me that variety is also sometimes risky because by producing a beer at a far end of the spectrum, you may capture some first time drinkers of your brand but quickly loose them forever if they dislike what you have offered. Beer is a type of product that economists call 'experience goods,' goods whose quality you cannot determine until you consume it (as opposed to clothes, say, which you can touch, feel, try on, etc., and have a pretty good idea of the quality before you buy). Since many people more familiar with macro pilsners don't really know what to expect from an aggressively hopped Oregon IPA, it may be better to introduce them first to a milder pale ale. But you can't really control this if you are a beer company offering a lot of distinct beers - so are beer companies hurting themselves by offering too many distinct beers thus increasing the chances of a bad first encounter? What is the 'optimal' level of product variety in this case? Open questions.

Another take is that by producing beers like Top Sail and The Abyss, breweries establish their credentials with beer snobs and the beer press and add luster to their brand - somehow a Full Sail pale is that much better knowing it comes from the company that brought you Top Sail.

Once again, I understand that brewers' motivations are many, but they do have to sell to survive so economics considerations remain very important.

Beeronomics: Product Variety and Spatial Economics

When thinking about product variety in a tractable model, economists often rely on spatial models, the classic being Hotelling's linear boardwalk model. If you think of consumers tastes as being spread out on a line, like a boardwalk where one end might, in the case of beer, be very light pilsner beer and the other be very dark stout ale, then you can conceptualize it in terms of distance from one spot to the next. In other words, suppose my preferred beer is a medium bodied IPA, then if you are only offering light pilsners I would not be willing to pay as much for that beer as I would an IPA. Similarly if I were at a particular spot on a boardwalk, and the only hot dog vendor was a quarter of a mile away, I would not pay as much for that hot dog as I would if the hot dog vendor were only steps away. Having to travel the half mile there and back is costly to me and thus I figure it in when thinking how much a hot dog is worth to me. Now obviously it is a gross simplification to describe beer varieties along only one dimension (light-dark), but that is precisely the point of modeling, stripping away a lot of real world complications (as long as none are fundamental) to come up with a description of reality that is tractable and that can inform us about the real world (think Newtonian physics).

The point is, that in these models, often the equilibrium strategy if you can only offer one variety is the locate in the very middle of the spectrum: this minimizes the distance between you and the mass of customers and allows you to sell more at a higher price. This is why, in my opinion, Pale Ale/IPA is the most common flagship beer among Oregon Breweries and the one beer most often found on the supermarket shelves.

There is also, however, a strong incentive to offer more variety, the closer you are to customers' tastes, the more you can charge them. So it is not surprising for the major Oregon breweries to offer a standard line up of 4-6 beers. This, of course, creates a tension between continuing to increase variety versus selling a lot of the most popular beers. However, it appears from casual observation that the equilibrium is about 5 varieties.

So it does not surprise me when, for example, Terminal Gravity enters the market with an IPA (or at least that's all I can ever find on supermarket shelves). It does surprise me, however, when a whole segment of boardwalk is ignored. It could be that there are not many people sitting on that end so it does not make sense to add a hot dog cart there given the cost and the limited demand. But in Oregon, the part of the boardwalk that is not served is the lager market, and I have a hard time, given the success of Sam Adams, that there is not a strong demand for lagers, even here in Oregon. I, for one, am a huge lager fan and bemoan the fact that I generally have to look to the east coast for tasty lagers.



Well finally, we have a new entrant in the Oregon beer market that is making lagers: Heater|Allen of McMinnville. They are not producing much beer yet so I have yet to sample the wares, but I anticipate a strong demand if they do ramp up production as long as the product it good and it appears to be. Let's hope they ramp up soon.

Beeronomics: Polling Results and You

I deliberately made the question, "what is the best Oregon brewery?," vague and open to interpretation. I didn't, for example, ask for "your favorite" brewery, or the "highest quality." I did this because I wanted, in a completely unscientific way, to examine some economic theories about consumer preference. What led me to think about this was the thought experiment: if you were starting up a new brewery, what would you do for your first offering. You have some choices, go for a style that is well known and liked in Oregon (IPA) or try something distinct and unique (like Hair of the Dog). Obviously, as I mentioned before, brewing is not all about business for many Oregon brewers, but you can't live the dream if you are not making a living.

Economists believe that consumers like variety in what they consume (you don't have the same meal for dinner every day of your life). They also believe that consumer's tastes vary depending on the person and that income matters a lot in the types of goods you consume - higher income individuals have preferences for higher quality or luxury goods. So what is a producer to do given these stylized facts? Well the fact that individual consumers love variety may lead you to try and offer them many beers of many styles (Rogue and Deschutes come immediately to mind), but this is expensive. It is hard to brew in smaller batches, have many different products for which you have to find tap handles and shelf space, and have to market many different beers. Other breweries seem to concentrate on a 'flagship' brew and hope that leads consumers to their other offerings (Widmer and Bridgeport come to mind here). The fact that there are also many different types of consumers out there has two competing effects for which there is often no equilibrium. Do you try and stake out as big a section of consumers by offering them a product somewhere in the middle of the spectrum? Doing so almost assuredly mans you will face tough competition (Terminal Gravity's IPA). Or do you go for a niche market that may be smaller but you have all to your own (HotD Fred)?

Once a brewery is established if faces another set of questions: do we try and grow sales of our established beers or create even less distinct beers to appeal to a wider audience to take advantage of economies of scale, or do we grow our variety to appeal to more customers and to keep current customers happy? I think the trend toward macro-style lagers (Full Sail Session) is a strong move toward the former, while Rogue, for example is a clear example of the other strategy. Finally, I think there is a reputation effect that is hard to break. Widmer is a clear example. When Rob and Kurt started Widmer, they were pioneers brewing styles no one had ever heard of before. In those early days of green palettes what became a hit was what are now considered less-distinct beers. So it is hard to break that early reputation a brewery establishes.

So at last, to the poll results: In a strange and suspicious flurry of voting Rogue surged ahead in the space of a few hours. But I am extremely pleased as I am taking a group of OSU students to meet with Jack Joyce, the owner, and I can share with him the exciting news. He will be so pleased. Deschutes, Full Sail and HotD also garnered much support. For big breweries, Widmer and MacTarnahans (which had for most of the poll had zero votes) had few votes, quite obviously because their beers are not generally considered on par with the others. But I think if I were to try and make something out of them I would say that the strategy of trying to produce a variety of quality beers is what begets a reputation as a great brewery. I voted for Deschutes for this reason. With the exception of Quail Springs, which has now been replaced by the excellent Inversion IPA, every single one of their beers is, in my opinion, a absolutely exceptional, flawless beer. I am stunned at how well they have achieved this feat and thus my vote. Rogue is perhaps a more interesting brewery, always trying out new beers and new takes on old styles, but their offerings, while generally exceptional, have a few less than stellar examples. Terminal Gravity, by contrast makes what may be my favorite IPA, but is not known as a great brewery by my voters probably because that is about the full extent of customers experience with them. Since they have decided to compete on such a popular style, I fear for their continuing success (they have already lost their taps in two local joints here in Corvallis). The most surprising outcome, in my opinion, is Bridgeport. I have such fond memories of the old Bridgeport brewery in the Pearl, when I was an undergrad and the Pearl was still industrial, and I was appalled when I returned after the big remodel. Perhaps this is such an iconic part of Bridgeport's identity that I am not the only one that cannot forgive the injustice. But their IPA is perhaps the quintessential NW beer. Blue Heron was perhaps the pioneering bottled NW micro-brew. What gives, no love for the Ponzis?

Thanks for everyone who participated in the poll and now, go buy some great NW beer and rethink your choice.