[caption id="" align="alignleft" width="400" caption="Whole Foods, 2017"][/caption]

Sometimes the most interesting things are right in front of you. Earlier today, I was walking past a local bar and I noticed a sign for the “drink exchange” over the door. I’ve walked by this bar many times before and I’d never seen that sign.

What was really interesting was the logo, which is a ticker tape coming out of a martini glass. It turns out that the Drink Exchange is actually a system that bars can buy that allows stock-market-like pricing of drinks (and other things, if requested).

What does this mean? For example, let’s say the price of a gin and tonic is $5 (I obviously do not live in New York anymore). The price can fluctuate over the course of the evening, and the system displays the price history and current price on a screen in the bar. Bartenders can also simulate a “stock market crash” in which all drinks become extremely cheap, to get patrons to buy.

There are a few interesting things about this idea.

First, the ways in which it does not work like a stock exchange. As far as I can tell, the demand for particular drinks has very little to do with the price; it’s just a random change or one that’s promotionally motivated. It would arguably be much more interesting if the prices fluctuated with demand, although of course this makes very little sense when resources are not constrained (a bar shouldn’t have a limited supply of vodka, for example).

Amusingly, one of the commenters posted that “You literally have a ‘stock market’ on certain nights, where if people drink more of a particular beer, it becomes cheaper!” No. That is not how a stock market works. I worry about this one.

Second, what effect this has on sales and margins. By getting people to focus on price, does the bar do well on volume or does it lose out on margin? The site notes that the drink exchange increases sales by 3000%… that tells you very little about the effect on profitability. If you click through to their chart, it’s not clear that the scale is meaningful; on an average night patrons purchase half a Bud Light? Very small N (statistical sample) here.

Third, the applicability of this idea to other sales contexts. What if you were in a supermarket and you could bid on tomatoes? Or books at a bookstore? Wharton uses auctions to allocate seats in popular classes, but the system is being used appropriately (if perhaps not always fairly) to allocate scarce resources; tomatoes shouldn’t be scarce at a grocery store. Would this be an effective promotional strategy? Would it be an effective allocative strategy for items that sell out? What if Apple just auctioned off all of its new iPhones on the day they came out, rather than having people wait in line?