Some thoughts on hidden economies

If you read most business articles these days, they usually contain a line like, “things are changing faster than ever.” That seems like a bit of a blanket statement. If we dig a bit deeper, we can identify a few really big trends that make it seem this way:

• The move from atoms to bits in almost everything we do, reducing the friction in systems.
• The always-on, digitally linked second world we’ve colonized with cellphones giving us around-the-clock knowledge workers.
• The shorter lifespan of companies, meaning most of us won’t work for one employer all our lives.
• The rapid dissemination of technology and learning, speeding up the pace of innovation and requiring lifelong retraining rather than one-time education.

One of the ways these trends have manifested themselves in our everyday life is in the rise of new, and often disruptive, hidden economies. If you were a newspaper publisher, you probably didn’t anticipate that Craigslist was going to kill you. Yet localnewspapers lost $5 billion in revenue between 2000 and 2007as a result of Craigslist’s entry into the market. The company may only make modest revenues from a subset of listings—but if you value it by the amount of ad revenue it eliminated, it’s worth Billions.

Hidden economies have always existed, but they’ve never been so close to the moat. If you’re innovating within a large organization, your biggest competitor may be lurking in a hidden economy, waiting to go legit—as AirBnB, Lyft, and others have—once technology eliminates barriers to entry. But how do we classify hidden economies in the first place? And if they’re not new, why are we hearing so much about them lately?

On the one hand, it’s clear that something new is afoot from a business standpoint. E-commerce, lean startup models, alternative funding, new ways of recognizing revenue, and more have transformed the way we launch, grow, compete, and sell. On the other hand, it’s not clear that this is genuinely different, rather than the same business wolf in electronic clothing. Part of the problem with naming and identifying hidden economies is that they are, by definition, hard to track.

Consider the following two examples:

• When we dry clothes in a drier, we capture the economic cost of that activity in drier sales, fabric softener, electricity, floorspace, and warranties. But when we dry clothes on a line, that value isn’t captured—despite the fact that it’s better for the planet. I heard about this from Tim O’Reilly at Foo Camp in 2012; it’s known as the Clothesline Paradox.
• Similarly, one of the most lucrative events to happen to an economy is an oil spill. The ensuing lawsuits, cleanup costs, rescue efforts, and replacement of the lost oil are a boon to Gross Domestic Product even as they savage the shoreline.

Externalities and hidden economies are inextricably linked.

How do we define a hidden economy?

Is a hidden economy simply one that’s off the grid, untracked and unregulated? If we speak of a hidden, peer-to-peer economy that includes Etsy and Kickstarter—because they’re “grey markets” that aren’t easily analyzed—do we need to lump pornography and offshore gambling into that list for similar reasons? Or is a hidden economy something more? Here are some of the definitions I’ve heard over the past couple of years:

• Is it simply part-time businesses that wouldn’t exist if it weren’t for the rise of free marketplaces?
• Do they only emerge from two-sided marketplaces in which the company only benefits when both participants benefit?
• Is there an element of subsidy or externality so that the vast majority of value is free, and money is recouped indirectly or not properly accounted for?
• Does it rely on other currencies, such as reputation or attention, rather than money?
• Is it something that’s only possible when the coefficient of friction for a transaction falls to near zero?
• Is it a newly free input, or a newly free output, to an existing system?
• Is it a switch from a market of scarcity (where we make money from what we sell the thing for, as IBM does from mainframes) to a market of abundance (where we make money because the thing is freely available, as Red Hat does from Linux)?

All of these are elements of this new, peer economy. But like the parable of the blind men describing an elephant, no one definition will suffice.

An Occam’s Razor for “hidden”

What we need is a clean definition with which to slice the hidden economy from the simply electronically-enabled one. We need an Occam’s Razor for hidden. One of my favorite stories of superb jurisprudence is the British judge who ruled on biscuits.

The British love their biscuits and cakes, but for reasons too Byzantine to fathom, the two are taxed at different rates. Jaffa Cakes, a chocolate-sponge-and-orange staple of the Empire, claimed that it was a cake, and consequently, subject to the lower tax applied to cakes. So it fell to this judge to decide on a definitive rule that would separate biscuits from cakes.

After much reflection he concluded:

A biscuit gets soft when it gets stale. A cake gets hard when it gets stale.

The Jaffa Cake was saved. Some well-respected economists and analysts with whom I’ve spoken insist—sometimes angrily—that there are no such things as hidden economies, and that talk of alternate currencies is bunko. “The only thing that counts, and gets counted, is money,” they proclaim. At the same time, companies maintain “goodwill” as a tangible, valuable asset on their balance sheets; and firms like Etsy use “residual lifetime value” as a better way of valuing their organizations than discounted cash flows. So something’s got to give.

Back to basics: inputs and outputs

Here’s a really simple model of how businesses function. Organizations take resources, capital, and labor; apply their processes to it in order to generate value, and then distribute it to consumers—often with the help of middlemen who will break bulk orders (selling a thousand oranges individually) and provide assortment (selling apples and bananas alongside oranges.)

First: new inputs

One thing that has changed is that we have a significant number of new inputs into the economic model (which we might mistake as a genuinely new economy.) Consider:

• Once, a moonlighting entrepreneur grew produce, working in the evenings in addition to their day job, and sold it at a local market. The capital was minimal; the resources were free (sun, water, and a back yard); and the labor was subsidized (by a day job.) The middlemen were simple (a market) and the transaction was opaque to regulators and tax authorities.
• Today’s entrepreneur can make digital music on a home studio, or design artwork in Photoshop, or commission collaborative artwork through Mechanical Turk. They can use a printing service to create prints on demand, without ever seeing the finished product. And they can then sell it on Craigslist, eBay, or Etsy. They can accept payment pseudonymously with cryptocurrency. They might even finance the product with Kickstarter, shooting a promotional video with free software and uploading it to Vimeo or YouTube.

The transactions are similar; but in the modern example, the inputs, capital, and other elements of the economic model are new. Let’s put “old” and “new” grey markets on this diagram. Simply put, there are more resources, more labor sources, more capital, more channels, and more consumers than before.

Second: new ways to track things

A second aspect of this “new” economy is that we’re able to track much more of the commerce in the world. For one thing, much of it is digital, and digital things can be analyzed. Every click, like, share and retweet leaves a digital breadcrumb trail, which means it’s a candidate for accounting. But more than this: it’s easier to track. Governments used to ignore grey markets on a de minimis basis.

It is an open secret that the Spanish jobless rate—double the European average—is a fiction. Hundreds of thousands of people claim unemployment benefit when they actually have some kind of work; millions are not registered as working, which means that neither they nor their employers are paying social security contributions. One proof, say employers, is that when unemployment fell to 8.5 per cent at the height of the boom in 2006–07, they could find no workers to hire. (Europe: Hidden economy, by Victor Mallet and Guy Dinmore)

One of the big questions is whether the “hidden” economy is a “new” economy, or whether it’s just the same old wolves—pornography, illegal immigrant work, predatory lending, extortion, drugs—in new clothing that can now be tracked more easily. According to Canadian economic analysts, one of the big reasons for the huge variances in reporting hidden economies is that there’s considerable disagreement about whether illegal activities should be included in assessments of the hidden economy.

One reason why estimates of the underground economy’s size vary is that no consensus exists on what exactly should be included in this category. Some researchers focus exclusively on lawful transactions that are not reported to government authorities, while others are interested in both hidden (but otherwise legal) activities as well as illegal transactions that generate income.

Today, computerization and ubiquitous computing are creating a world of sensors, where much more can be tracked for less effort.When Athens received only 300 tax payments for pool owners, it took an enterprising civil servant only a short while with Google Earth to determine that there were more than 16,000 pools in the city. Before, satellite imagery would have been prohibitively expensive; today, it’s an app.

So two big things are changing, largely as a result of the switch from atoms to bits that’s happening all around us. One is that there are new inputs into economic models, giving rise to new businesses; the other is that there are new ways to capture business activity, making more of them targets for regulators and tax authorities.

How do we pick our battles?

We need to pick a specific set of criteria to describe this new/hidden/peer economy, and decide what makes it real, interesting, and disruptive. But those criteria can’t be arbitrary, either, or the topic won’t be compelling. Here’s a 1997 Fraser Institute analysis of underground economic activities: [table id=2 /] We need a taxonomy like this, with examples and clear contrasts. It seems like this new, hidden economy we’re trying to describe:

• Is based on a new model, and that the new model is likely enabled by technology and networks. Traditional hidden economies like moonlighting might be hidden from regulators and tax collectors, but they’re not disruptive.
• That’s unreported in traditional ways. This may be because it’s an alternate currency (done for respect); because it’s hard to track (turking across borders); or because it’s displaced (Apple benefitting from Youtube content or Facebook activity.)
• Isn’t inherently illegal. While criminals will always embrace new technology, they’re also strongly incented not to report their activities. So we’re more interested in legal activities that somehow aren’t reported.

The Venn diagram of those three might look something like this: This isn’t a great definition. In fact, it’s pretty awful. We need to look at a number of other threads, and try to find the facets of a hidden/peer/new economy within each of them.