Some things seem to need no explanation, but are not obvious at all on reflection and, if you wonder about them, suggest something of interest about the economic system. Consider the question of why the informal economy is so small, leading to the question of how much more productive the formal economy must be than the informal economy to make sense of how little informal economic activityy there is. See over the fold for the full argument.

 
A typical example of the informal economy is someone who gets paid to repairs roofs for others but doesn’t pay GST or taxes on these activities. Informal activities can however also include the small-scale production of alcohol, medicine, health services, childcare, financial advice, transport services, etc. In essence, informal activities are merely small-scale activities that usually have large-scale formal counterparts.
Estimates of the informal economy, which are by definition untaxed, hover around 6 to 20% for OECD countries. The OECD puts the informal economy in OECD countries at 15% to 18%. These numbers mainly seem to derive from the work of Friederich Schneider and his co-authors, who estimate the informal economy by looking at the volume and speed of money (or electricity) and working out which bit of the wear and tear of dollar bills (or use of energy) is not due to the formal economy (see the article from The Economist).

Trevor Breusch has expressed his dissatisfaction with this technique, essentially arguing it overstates the level of the informal economy, so the true number in developed economies is probably closer to 10%.
To see why it is so informative that there is so little informal economy, reflect on the hassle formal companies have to go through in order to be formal: their workers have to pay about 30% income tax. The company profits are taxed by roughly 50% whilst the interest rate payments are taxed elsewhere. On top of this is the 10% GST, as well as a whole host of small official taxes, like garbage collection and vehicle duties. As a rough rule of thumb, it immediately costs about 40% of total revenue to be formal merely in taxes (see figures).

Then consider the administrative costs of being formal. Being formal means having to comply with thousands of pages of income tax regulation, health and safety regulations, labour regulations, etc. Authoritative numbers on just how much time in total is spent on regulations one wouldn’t have to adhere to in the informal sector do not exist, but we can generate a wild guess by looking at our own backyard.

Universities are a classic case in point of just how much admin is concerned with complying with government regulation of one type of another. If we put in the conservative guesstimate that half the university administrators spend their time complying with legislation (including central HR, payroll, health officers, etc.), then about 1 in 3 employees at universities are busy keeping the university formal (the ratio of admin to academics is about 2 in some Australian universities), another 33% of total activities (the admin are not paid less than the academics!). For other organisations, it is hard to say how much of their total activity goes into complying with government regulation, but a third sounds about right (for a litany of some of the burdens of compliance, see Chamber of Commerce and Industry Queensland or the Productivity Commission).

What do these numbers mean? If we add to the above calculations the fact that being in the informal sector also brings with it some access to the welfare system, then in order to make economic sense those who work productively in the formal sector (the non-rule compliers) must be about three times more productive in the formal economy than they would be in the informal economy. And this is only the ratio that is needed at the margin, i.e. the ratio of 1 to 3 needs to hold for the 90th% percentile of those working in the total working economy (formal plus informal). For the vast majority of productive workers, the ratio is therefore probably quite a bit higher. Perhaps the true ration is thus even 1 to 4 or higher, but whatever it is, we’re a long way off 1 to 1.
Why is this so interesting? This basic ‘factoid’ tells us that big is indeed beautiful. There must be all kinds of increasing returns to scale advantages to being formal that are not open to those who are informal. What are these advantages of being formal? They include formal channels of advertising, the ability to combine with many others in recognisable dwellings, the ability to trade in bulk, the ability to credibly offer long-term contracts, the ability to generate reputation, etc. The astounding thing is that these advantages are not only worth some 70% of total sold production to manufacturing firms for which one would easily accept this, but the advantages must also be worth this amount to medium sized service companies and even to most small-scale entrepreneurs.
How can we interpret these advantages? Many of these advantages are essentially solutions to market failures, like a non-zero cost of information, increasing-returns-to-scale technology, and the notion of asymmetric information. We can then conclude that the organisation of our modern economy is only possible because of market failures rather than in spite of them! We thus have come full-circle in the sense that the usual economic argument (we should get rid of all market failures) is turned on its head; in that we wouldn’t be able to sustain the present tax and regulation system if we indeed managed to eliminate market failures everywhere. We need market failures for our way of life throughout the whole economy, so that their solution within the visible regulated sector makes the informal economy an inferior choice! For theory, this musing also means that any perfect-market constant-returns to scale technology at the micro-level is essentially incompatible with the small size of the informal economy.

Author: paulfrijters

Professor of Wellbeing and Economics at the London School of Economics, Centre for Economic Performance View all posts by paulfrijters