Abstract
Estimating the size of the hidden economy continues to attract the interest of both researchers and politicians alike. However, and despite the recent increase in the technical sophistication of the two estimation methods most commonly employed in the empirical literature, namely the monetary approach and the multiple causes-indicators technique, it is necessary to continue making advances in our knowledge of the relationships between the hidden economy and its economic and institutional environment. Against this background, this paper sets out to provide empirical evidence on this question. Using the estimation made by Gadea and Serrano (Citation2001) of hidden income in Spain during the period 1964–1998, a univariate and multivariate causality analysis is carried out between hidden income and the representative variables of the legal economy, taxation, the labour market and currency. The results allow one to draw a number of interesting conclusions on the complexity of these relationships.
Notes
The recent paper of Schneider and Enste (Citation2000) offers an overview which analyses the different estimation methods proposed in the literature.
The most recent works which adopt the currency demand approach are characterized by maintaining a more critical attitude towards the assumptions and giving greater emphasis to econometric aspects. From these, one can cite the works of Bhattacharyya (Citation1990), Schneider and Neck (Citation1993), Spiro (Citation1994), Hill and Kabir (Citation1995), Schneider (Citation1997, Citation1999) and Gadea and Serrano (Citation2001).
For example, the works of Frey et al. (Citation1982), Frey and Weck (Citation1983a, Citationb) and Frey and Pommerehne (Citation1984).
See Aigner et al. (Citation1988).
On this last point, it should be noted that, as compared with other disciplines, relatively little attention has been paid to the statistical properties of the series (Atkins, Citation1999). Only a limited number of works, for example, those of Giles (Citation1999b) and Gadea and Serrano (Citation2001) have concerned themselves with this aspect.
Details of the estimation can be found in Gadea and Serrano (2002).
Other possible causes, such as regulation or fiscal morality, have been rejected in the Spanish case (Gadea and Serrano, 2002). Together with the tax burden, an indicator of the probability of detection and punishment (PDP) has also been included.
The unit root tests confirm that in all cases the variables are I(1).
The graphical analysis of the IRF, omitted for reasons of space, reinforces the results of the FEVD in both models.
Giles (1997), using the Granger causality test, highlights the causality relationship that exists between the legal and the underground economy in New Zealand. This author concludes that periods of strong economic growth give rise to the appearance of greater incentives to carry out irregular activities which escape fiscal control. However, by withdrawing productive resources, irregular activity can also have negative effects on economic growth, with these appearing weakly in the Spanish case. As Schneider and Enster (Citation2000) indicate, a number of studies have obtained opposite results on the relationship between regular and hidden income, with it being necessary to devote more attention to this aspect.
These conclusions are in line with those of Ahn and De la Rica (Citation1997), who analyse the relationships between the underground economy and unemployment in Spain. These authors indicate that the hidden economy arose in Spain not only as a response to taxation, but also as an alternative to unemployment. The higher probability of working in the informal sector – with this coming from the position in the family unit, gender, education level – appears with respect to the type of workers who have fewer incentives and greater difficulties in finding work, with this being a faithful representation of the structural features of the Spanish labour market.
Tanzi (Citation1999) points to the need to pay greater attention to the relationship between the hidden economy and labour supply. This author indicates that irregular activity feeds off workers who, for a variety of reasons, for example, being retirees or housewives, pensions recipients, etc., do not form part of the active population and who, as a consequence, are not accounted for as being unemployed. Furthermore, people who are officially listed as unemployed are, in fact, working in underground activities.
The sign of the male activity rate is negative, and that of the unemployment rate positive, with this latter variable dominating.