Two Software Developers are Seeing to it That You Never Get Ripped Off Buying Weed Again

 

It’s Friday, and I do have shit to do. But if I’m to believe priceofweed.com, an anonymously crowdsourced, regional index of the going street rates for marijuana, I could steal away from this desk, wander around outside and begrudgingly fork over $500 for an ounce of high quality bud. If I were back in Chicago, that figure would be more like $480; out in Denver, about $290.01; and in Portland, something like $253.35.

These are averages, of course. And maybe this is a no-brainer bit of Manifest Destiny (probably not), but it magnifies a central tenet of Economic Theory 101: Criminalize a good and you not only forfeit all control over it, but you fracture its underground economies. By nature, black markets aren’t disposed to rigorous study and analysis. They often lack empirical data and are full of skittish research participants that are flung across all possible environments.

It’s only fitting that the economic-information geographers behind “Data Shadows of an Underground Economy”, a new draft report that “drew upon over 16,000 spatially-referenced, retail purchases of marijuana” amassed at PoW to “cartographically and statistically analyze” price differentials throughout the States, are so dispersed. Matthew Zook is an associate professor at the University of Kentucky, where an ounce of high quality bud will set you back $386.63. Monica Stephens is a PhD student at the University of Arizona, where that same ounce goes for $370.54. Mark Graham is a research fellow at the Oxford Internet Institute, Oxford University, where I have no credible way of knowing what a decent O runs for these days. After poking around PoW late last year, Zook tells me, “We thought, ‘Wow, this data is really interesting. It’s really this sort of nice, fine grain. We could do something interesting with it!’”

They reached out to PoW, a two-man operation launched in 2010. Both are software developers, and by now are used to academics coming around, asking for help. One helmsman, who I’ll call Cory, says they get requests every week from master’s students and researchers from across the world looking to make sense of their copious time- and date-stamped data. “We’re happy to send it out to them,” he tells me.

So after ridding the data from this volunteered geographic information (VGI) of outliers, the researchers – all members of the Floating Sheep collective – drew up a continuous surface through kriging, a geostatistical interpolation formula, to flag the average variance within price spreads with a spherical semivariogram model. Their analysis illustrates that this particular underground economy, in terms of “supply and transportation effects,” adheres to established economic theory. It also affirms the idea that any one state’s legalization regime has a significant (see: negative) impact on the price you pay for the good shit. These findings aren’t at all surprising, really, but they illustrate the effectiveness of merging crowdsourced geographic data with studies of economic habits “that would otherwise be impossible to conduct.”

Of course, the Floating Sheep collective aren’t solely bent on handling street prices of illicit drugs. But the equivalent of a PoW website could be made for almost any activity, Zook claims, and likewise pull from user-generated reports. And what’s remarkable here is we’re talking about a nearly full-blown illegal economy: “You just don’t see this sort of market data,” he explains.

It’s true. Poke around the internet and you’ll find a lot of homebrew “indexes” of street values within underground economies, but most are slipshod and sorta slimy and, with regard to illegal drugs, largely catering to the pill-head set. PoW is far more together. Maybe that’s testament to the simplicity of a design that’s allowed its creators to sit back and busy themselves with other projects as the data comes rolling in. “We haven’t had a chance to work on it too hard in the last little while,” Cory admits.

But their blog, dormant for nearly a year, was updated around the time he and I spoke. “Has it really been that long?” the post begins. “Anyway, here’s what’s been going down.” PoW has reached 30,000 submissions, and their data, via Floating Sheep, is featured in a nice Infoporn spot in this month’s issue of Wired. They’ve also got some unreleased features they’ve been tinkering on. These will plot data “for all countries” (particularly Europe), break down prices by city and create “an animated heatmap of price change over time."

Will PoW continue reaping submissions? There’s an inherent paradox behind all this. Today’s snowballing crowd-source movement rests on legions of subjects who may have no idea how the innocuous personal information they give up may be flipped against them. “There’s so much data on what people do, when they do it, where they do it,” Zook says. A dearth of information used to be a problem. “You couldn’t find any good information about local consumer behavior. Now, we’re sort of at the opposite end. We have too much data. We have these massive data sets, and no one quite knows how to work with them.”

o