Blogs & Comment

Sizing up the underground economy

How statisticians size up the underground economy.

A native Kahnawake police officer stands with SQ and RCMP officers with confiscated drugs, weapons and cash in Montreal. (Photo: Ryan Remiorz/CP)

Statisticians have a beef with criminals. Simply put, most felons do a lousy job of bookkeeping. Biker gangs rarely publish financial statements disclosing their cash flows from cocaine and heroin trafficking operations. Brothels are even worse about filing reliable tax returns. Strongly incentivized to avoid regulations governing working conditions, pimps are seldom forthcoming about their activities. Bootleggers and tobacco smugglers—you can’t trust anything they say.

Come to think of it, seemingly innocuous individuals aren’t blameless either. That stay-at-home parent wheeling her infant around the neighborhood is engaged in a valuable service—just ask anyone who’s paid for a week of daycare. Ditto for your neighbor when he’s hanging his laundry on the line or vacuuming his living room carpet. Such activities are seldom measured, but they are also of considerable economic importance.

If you think such a venerable concept of GDP is quite uncontaminated by crime, you’d be wrong. To purists, even the $15,000 paid to a hitman to knock off somebody should be included in GDP—he’s providing a service, after all. Fortunately, the number-crunchers at Statistics Canada aren’t afraid to get their hands dirty. StatsCan says it’s important to measure underground economic activity whenever possible so that published measures of gross domestic product (GDP) are as exhaustive they can be. And doing so makes our GDP figure more comparable to those published by other nations, many of which are also struggling to measure what goes on in the shadows. The most recent product of this ongoing work is a StatsCan report released this week entitled Estimating the Underground Economy in Canada, 1992-2008.

How does one measure the seemingly unmeasurable? It’s not easy. Fortunately our intrepid statisticians needn’t start from scratch. The Organization for Economic Cooperation and Development (OECD) is an international body that seeks to foster trade and economic growth. Nearly a decade ago it published a handbook entitled Measuring the Non-Observed Economy that addresses this question. “Non-observed” activities include overtly prohibited activities like creating and distributing pirated movies, but also “hidden” activities such as a construction company underreporting revenues or a waitress misrepresenting the value of tips she collects. And there are also “informal” activities such as cooking, washing the dishes and taking care of the kids. Statistics Canada has integrated the handbook’s recommendations into its own procedures for measuring the national accounts.

Consider drugs. Chances are, the meth lab on your street failed to notify Industry Canada of its burgeoning production last year. To capture its contribution to our economy, statisticians might consult the quantities of crystal meth seized by police and multiply it by an estimate of how much that represents of the total trade. The result: a ballpark guess of the industry’s total production. Another route is to estimate the number of meth users present in a country, and multiply that by an estimate of average consumption per head. Naturally this approach would fail to capture export activities, but it’s a start.

What about prostitutes? Customers may be hesitant to disclose their spending in this particular category to Statistics Canada during its Survey of Household Spending. And sex workers are sometimes accused of taking considerable liberties on their tax returns. To get a sense of what’s happening in the bawdy houses of the nation, the OECD suggests statisticians begin with an estimate of the total number of prostitutes. “An estimate of the number of clients to each worker then provides the volume of output and multiplication by the average prices of each service gives an estimate of gross output,” the OECD explains in a guide published in 2002. “Intermediate consumption—rent, exotic clothing, electricity etc.—is usually assumed to be some low, fixed percentage of gross output,” the document adds helpfully.

And how about “underground construction?” (In this context the term refers to unreported building activity, rather than building subways and bomb shelters.) One technique Statistics Canada tried was to consider the growth rate of the ratio of intermediate inputs (lumber, for example) to total outputs in the construction industry. The idea was “to observe the movement of this ratio over the years and determine if differences in movements could be attributed to underground economy activities.” If inputs were rising more quickly than outputs, it might suggest construction companies were up to their usual tricks: inflating costs or hiding revenue. But then again, it might only reveal that the companies have increased their productivity.

And how about illegal tobacco? One can estimate the amount of cigarette smoking in Canada by asking people about their habits. Or you can look at how much cigarette taxes are collected by government. If taxes collected are falling faster than reported smoking behaviour, one might infer rising illicit tobacco production and smuggling.

Given the sweeping assumptions and methodological unsoundness evident in the above examples, sizing up the underground economy has much in common with voodoo. (One OECD document on the subject uses the word “guesstimates” in describing the process. When statisticians invoke such terms, it’s time to pass the salt.) Statistics Canada admits its estimates on the underground economy “are not as accurate or reliable as other estimates in the national accounts.” That’s an understatement. But unless we can convince our economy’s troglodytes to come out into the light, it’s the best we’ve got.

Here are a few noteworthy takeaways from StatsCan’s report:

• Worth an estimated $36 billion in 2008, the underground economy is relatively small in Canada. According to the report, it accounted for just 2.2% of GDP that year. In some countries, such as Kazakhstan and Armenia, it’s been estimated at more than a quarter of total activity.

• Crims are losing market share. StatsCan found that although the underground economy has grown since the last estimates in 1992, it did so at a slower pace than the overall economy. The main reason was that “industries traditionally considered to be involved in underground activity are a declining portion of the overall economy.” Those industries include construction, retail trade and accommodation and food services.

• Underground goods and services accounted for as much as $24 billion in 2008, representing 2.7% of total household expenditures.