Dynamics of informal microenterprises and poverty in Peru: a panel approach

This is the authors' abstract. We don't add key points for this paper.

Studies into the dynamics of micro-enterprises and poverty The recognition of the importance of taking into consideration jointly the characteristics both of production units and of workers is relatively recent in empirical economic studies and stems from the fact that workers with similar individual characteristics, but working in production units with different characteristics, have different wages (Abowd et al., 1999; Mortensen, 2003). In the case of developing countries the theme has been indirectly addressed, through strategies of diversification towards non-agricultural productive activities in the case of rural households (Lanjouw, 2008). The explanations for this phenomenon have sought to consider conjointly the factors linked to labour supply and demand and in particular the characteristics of enterprises. According to Hamermesh (2008: 664), ‘One of the most interesting developments in labour economics in the past decade has been the new ability to study both the demand and the supply side of the labour market’. For Hamermesh, a crucial question in the income’s equation is to know the relative importance of the characteristics of enterprises compared to those of the workers. This is important for public policies as these two factors call for two different types of intervention. It rapidly transpired that there was a strong correlation between wages and the size of firms. Oi and Idson (1999) have found three hypotheses to explain this: (1) the most productive employees pair off with the most skilful entrepreneurs in order to minimize the total amount of salaries and supervision costs; (2) the biggest enterprises pay the highest salaries in order to retain them; (3) the biggest enterprises adopt a discretionary salary policy in order to participate in the annuities. The size of the enterprises, a factor which is closely linked to their productivity, is considered in several empirical studies as the principal determinant for inequalities in wages. Brown and Medoff (1989) remarked, in the case of the United States, a positive correlation between the size of the firm and workers’ salaries. These disparities between wage levels persist even when controlling the effects ofobservable characteristics as much for workers as for enterprises. Groshen (1991) also investigated the sources of the intra-branch wage dispersion of North-American workers by taking into account the differences linked to the occupation and the establishment (size, presence of unions, branch, per cent of women, etc.) and found that the characteristics of establishments explain nearly half of the salary disparities. Brunello and Colussi (1998) studied the case of Italy which is more similar to that of developing countries because of the relative importance of small-and medium-sized enterprises. The authors estimated mincerian income equations by considering the allocation of workers in enterprises of different size as being explicitly endogenous. The authors found that the differential linked to the size of the enterprise is not different from zero: but this result is sensitive to subjacent hypotheses of worker distribution in the enterprises. Disparities in salaries might be explained by individual characteristics observed among the workers and by selection effects. The informal sector has been generally considered to be a subsistence sector, a refuge during periods of crisis. Another vision describes the informal sector as a hive of activity, where micro-entrepreneurs carry the torch of the capitalist spirit (de Soto, 1986). Although the incomes of informal entrepreneurs are on average relatively low, they are also quite dispersed and, for one segment of them, quite near to, if not higher than, formal sector salaried workers. In the ‘choice’ motive explanation for becoming an informal entrepreneur two antagonistic visions coexist. For some it is a choice linked to expected advantages (and not only pecuniary ones), while for others it is a situation which is not freely chosen and is linked to the fact that workers have insufficient qualifications to access salaried work. According to Cunningham and Maloney (2001), these opposing visions can be reconciled once the existence of a high level of heterogeneity within the informal is acknowledged. Very few studies have examined the links between workers’ incomes and the characteristics of enterprises in developing countries. Funkhouser (1998) studied a small sample of enterprises in Guatemala (data for 256 big enterprises gathered by the Labour Ministry). The results of the regressions show that half of salary dispersion is linked to the observed heterogeneity of the enterprises and that about 15 per cent is linked to the fixed effects of the enterprises. By using a representative sample of micro-enterprises, Cunningham and Maloney (2001) showed that in the case of Mexico, the high level of heterogeneity of microenterprises does not correspond to a segmentation of the labour market but rather to the disparities specific to a sector where small enterprises proliferate. The observed disparities in terms of enterprise size, permanence and degree of formality are argued to be due to the unequal distribution of human and physical capital and not to any dualism whatsoever in the labour market. The factor and cluster analysis applied to the survey of Mexican microenterprises data effectively reveals that micro-enterprises serve as a refuge for those unable to find salaried work but the authors also point out that this is nevertheless a small phenomenon compared to the vast majority who have chosenvoluntarily this type of work out of a desire to be independent and because of the perspective of a higher income. The profile of those who leave it corresponds more to that of workers who have failed and voluntarily quit the micro-enterprise sector. The authors conclude that there exists only a weak link between capital intensity, size and human capital returns (Cunningham and Maloney, 2001). For Tokman (1989), workers in the informal sector are part of the workers who do not possess the required qualifications to accede to the salaried jobs which are insufficient in number owing to low productive investment. Conversely, for Soto (1986), the micro-entrepreneurs in the informal sector could develop and manage bigger enterprises. Regulations and the weight of administrative costs constitute a serious barrier to their development. Even if there is no doubt about the existence of a dynamic segment of microenterprises capable of evolving into bigger ones in developing countries, very few studies have examined the micro-economic determinants which make these transitions possible and the few rare existing studies show contradictory results which are very sensitive to the methodological approach used. Thus, de Mel et al. (2008), basing themselves on panel data from households surveys in Sri Lanka, show that about two-thirds to three-quarters of self-employed workers have characteristics similar to those of salaried workers rather than those of the managers of big enterprises. The authors conclude that credit access is not the only constraint to the growth of micro-enterprises. The analysis of the panel data from the Chilean household survey (CASEN) serves as an empirical base for Nopo and Valenzuela (2007) for the analysis of the consequences of the reconversion of salaried workers into entrepreneurs. Using matching techniques to construct a counterfactual, the authors show that the gains are positive and significant while for the cases of inverse transition there is loss of income.