Size of the US Underground Economy

Sharon Astyk had a thought provoking post last week:

Here is the single biggest question to consider about the economic, energy and environmental unwinding we are facing - what will the economy look as we go? I get more questions about this than about anything else - what should people do for work, what should they do with savings, how should they begin to prepare themselves for a lower energy world. What I find, however, is that among both the prepared and the unprepared, there's a whole lot of people kidding themselves. There are those who imagine that there is no economy outside the world of the stock market and formal jobs - that a crash in those things is the end of the world, which means to them either that it can't happen or they should buy a bunker and some ammo. Others have imagined themselves "free' of all economic structures larger than the neighborhood, cheerfully providing most of their needs or bartering and never again touching cash. Both ideas fall into the realm of fantasy.

Let us remind ourselves that the informal economy is, in fact, the larger part of the world's total economy. When you add in the domestic and household economy of the world's households, the subsistence economy, the barter economy, the volunteer economy, the "under the table" economy, the criminal economy and a few other smaller players, you get something that adds up to 3/4 of the world's total economic activity. The formal economy - the territory of professional and paid work, of tax statements and GDP - is only 1/4 of the world's total economic activity.

We know from peasant economist Teodor Shanin and others working in the field that when the formal economy fails people all over the world, they shift into the informal economy. This explains why, in the former Soviet Union, although conventional economic models showed that people "should" be starving to death, they weren't. This is how people with functionally no income can still eat - although often not well.

I frequently disagree with Sharon, but I do find some points of agreement on this question of the informal economy.  My sense of the likeliest evolution of society in coming decades is that global economic capitalism will persist but that it become more efficient by continuing to become more automated.  Thus it will be able to serve the interests of economic and cultural elites while requiring fewer resources (particularly oil) because it will increasingly not require the services of, or serve the interests of, the masses.  I've written on a number of occasions of how I think one of the earliest symptoms of the gradual approach of the "singularity" is the continued lowering of the US male employment/population ratio:

Levels are particularly low amongst the young:

and the less educated

I would expect that the increasing lack of need for many classes of men in US society would be associated with an increase in the size of the underground economy: drug production and dealing, crime, casual cash labor, off-the-books construction projects, etc.  I would also expect that the size of the informal economy would continue to grow further as the US experiences more globalization, more immigration, more automation, and less access to oil and other resources.

I was curious to know what efforts had been made to estimate the size of the informal economy.  After some poking around, I discovered Edgar Feige, a widely cited Professor Emeritus of Economics at University of Wisconsin-Madison who seems to have devoted much of his career to this question.  His latest estimates are in this paper, and here's the bottom line:

This shows two estimates of the size of the underground economy estimated as the ratio of all unreported income to reported income.  Note that the graph is not zero-scaled.  This indicates that the underground economy more than doubled in size during WWII (in response to rationing), fell during the rapidly growing years of the fifties and sixties, increased sharply in the seventies with the oil shocks, improved a bit in the eighties and nineties, but then has started to reach new heights in the difficult years of the aughties.  This is more-or-less what I would have expected.  I predict that these ratios will continue to go higher in future years.

It's worth noting that there are considerable (and rather fascinating) measurement issues that have to be overcome to produce these estimates and the results should be viewed as somewhat uncertain.  Some sense of the issues can be gained from these quotes from the paper:

One of the most reliable economic statistics is the amount of U.S. currency in circulation held outside of depository intuitions by the public. By the end of 2010, U.S. currency in circulation with the public had risen to $920 billion dollars, amounting to $2950 for every man, woman and child in the country. Over the past decades we have witnessed a host of cash-saving financial innovations, leading to widespread predictions of the advent of a “cashless society”. But contrary to these expectations, the demand for U.S. dollars continues to rise and we remain awash in cash. Over the last twenty years, real per capita currency holdings increased by 79 percent and currency as a fraction of the M1 money supply rose from 30 percent to 49 percent.

To put these figures in perspective, they imply that the average American's bulging wallet holds 91 pieces of U.S. paper currency, consisting of: 31 one dollar bills; 7 fives; 5 tens; 21 twenties; 4 fifties and 23 one hundred dollar bills. Few of us will recognize ourselves as “average” citizens. Clearly, these amounts of currency are not normally necessary for those of us simply wishing to make payments when neither credit/debit cards nor checks are accepted or convenient to use.

Federal Reserve surveys (Avery et al. 1986, 1987) of household currency usage found that U.S. residents admitted to holding less than 10 percent of the nation's currency supply. Businesses (Anderson, 1977; Sumner, 1990) admitted to holding 5 percent. It seems that the whereabouts of roughly 85 percent of the nation's currency supply is unknown. This anomalous finding suggests that the “currency enigma” (Feige 1989, 1994) and the problem of “missing currency” (Sprenkel, 1993) is still very much with us.

The currency enigma has both a stock and a flow dimension. First we must determine who holds the outstanding stock of U.S. cash. Specifically, how much of this currency is abroad, (the dollarization hypothesis) and how much is held domestically (the underground economy hypothesis) by citizens reluctant to admit to their true cash holdings? The flow issue concerns the amount of cash payments sustained by that missing currency. If half of the missing currency is hoarded and the other half is used as a medium of exchange, turning over at an average velocity of between 30 and 50 transactions per year, (Feige, 1989a) the missing circulating currency stock would give rise to a flow of “missing payments” of an order of magnitude comparable to the entire GNP of the United States.

In essence it's first necessary to produce estimates of what fraction of the unknown currency is actually overseas, and then it's necessary to estimate how fast the domestic underground cash is turning over in order to get to estimates of the amount of underground income.

The entire paper is well worth a read.