on production and labour markets. The size of the hidden economy (unobserved variable)
is influenced by a set of (measurable) indicators/causes such as the burden of taxes and
regulation as well as tax morality, and change in labour force participation. Based on
the statistical theory of unobserved variables, the unknown coefficients are simultaneously
estimated in a set of structural equations typically by maximum likelyhood. The indicator
variables are used to capture the effect of the unobserved variables indirectly. Different
combinations of causes and indicators can then be used to provide different estimates
(Chaudari et al. (2006)). This method provides a time-series index for latent variables,
an ordinal index which is then converted into a cardinal series of values of size by scaling
up ordinal values to cardinal values previously obtained through other indirect methods
(i.e. currency-demand approach). One of the main criticism of this method is the lack
of a theoretical link of causes and indicators with informality being the only unobserved
variables linking them (Perry et al. (2007)).7
In response to some of the criticisms of the MIMIC approach, Solomon (2008) uses
a RBC model as a theoretical framework for choosing causes and indicators used in the
estimation of the size of the informal economy.
2.2 The size of the informal economy
What is the size of the informal economy in the world? In general, data on the size,
composition and contribution of informality is limited and international comparison of the
scale of the phenomenon is difficult due to different measurement strategies.8 For this
reason, the main findings are often supported by indirect measurement methods.
In general, as we saw above, estimates of the size of the informal economy can be in
terms of income or in terms of employment.9 Chen (2007) reports that informal employ-
ment is about one-half to three-quarters of non-agricultural employment in developing
7See Appendix A in Schneider (2005) for a detailed analysis of the main methods used to estimate the
shadow economy and to Dell’Anno (2003) for a detailed analysis of the MIMIC approach and its main
criticisms.
8Many countries exclude agriculture and others include only an urban informal sector.
9We also would like to stress that direct methods depends on the definition of informality used. Henley
et al. (2006) point out that the choice of the definition is often dictated by data availability. In their paper
they compare different definitions of informality based on employment contract legislation, social security
protection and employers/employees’ characteristics using a complete household survey for Brazil.