Estimacion de los efectos de la formalidad sobre microempresas informales mexicanas: Un enfoque multivariado conjunto.

AutorGiannatale, Sonia Di
CargoReport

Estimating the Effects of Formality on Mexican Informal Microfirms: A Joint Multivariate Approach

Introduction

According to Fajnzylber et al. (2009), microfirm owners and their workers accounted for approximately 50 per cent of the Latin American labor force during the last decade. In all Latin American countries, and especially in Mexico, (1) a significant proportion of these microfirms operates in the informal sector of the economy; that is, they neither pay taxes nor fulfill their legal obligations toward their workers. Yet, they make use of many public goods, such as the public health system, the sidewalks, streets and parks where they conduct business, and, in many cases, do not pay for the electricity or water they use as production inputs. However, the quality of those public goods might be compromised because of the resulting loss of tax revenues. The discussion over the determinants of informality is still ongoing. A country's regulatory burden and its capacity to enforce the law have both been identified as determinants of informality, but there is empirical evidence that suggests that as the latter grows stronger the former becomes less important (Dabla-Norris et al., 2008).

At the individual microfirm level, the owner's choice of operating in the formal or the informal sector of the economy should be part of a profit maximization decision, as Levenson and Maloney (1998) discuss. Under this perspective, Fajnzylber et al. (2009) analyze whether formalization has an impact on the performance of Mexican microfirms and which treatment (credit supply, entrepreneurial training, and access to government services) is more effective on business performance. They find that formalization has potential positive effects in profits and the survival rates of microfirms.

In this article we concur with the idea that formality is an input of the productive process, where its marginal value increases as the size of the firm increases. For larger firms, meeting institutional requirements becomes more important because a larger operation is more likely to draw the attention of governmental agencies. For some firms, however, either because of specific characteristics of their type of activity and capital restrictions, or their owner's entrepreneurial ability, the optimal decision is to remain in the informal sector and to maintain a small scale of production. For instance, Dabla-Norris et al. (2008) empirically establish a positive association between small firms and informality. On the other hand, Cunningham and Maloney (2001) show that, since the informal sector in Mexico is highly heterogeneous, a reasonable explanation for microfirm owners' decision to remain in the informal sector can be found in differences in their entrepreneurial abilities and heterogeneous preferences, and not in distortions in the labor or credit markets, as it is usually assumed.

In this paper, we focus on the decision-making process of informal Mexican microfirms trying to answer two questions: a) are, ceteris paribus, formal firms more profitable than informal ones?, and b) is, as suggested by the model of Rauch (1991), the informal sector the optimal choice for small firms and the formal sector the preferred choice for larger firms? To answer these questions, we perform quantitative comparisons on three measures of firm size: net profit, investment in physical capital, and number of workers. Our study explicitly assumes the endogeneity of the firms' decisions. That is, we assume that the decision of whether or not to participate in formal institutions is part of the entrepreneur's profit maximization decision problem (Levenson and Maloney, 1998). Accordingly, assuming profit-maximizing owners, a microfirm operates in the formal sector only if the net profits (monetary or not) obtained from operating in this sector exceed those obtained in the informal sector.

We perform a joint simultaneous analysis of a vector of interrelated microfirm responses resulting from the sector choice by using a Roy model with multiple response variables (Heckman and Vytlacil, 2007). We estimate all the parameters in the model using full information maximum likelihood by means of a Montecarlo EM algorithm. This methodology allows us to control for the endogeneity of the responses, account for the correlation among the different responses, and, therefore, obtain more efficient estimates of the treatment effect. Additionally, we include a heteroskedastic error covariance matrix in the estimation. This multivariate approach distinguishes our analysis from other related empirical papers that analize microfirm responses separately using either standard selection models or matching methods. Differently from most of the related literature on treatment effects, in which the effect of treatment (formality in this case) on the treated (formal) sub-population is the focus of the estimation, we estimate the effect that treatment would have on the untreated (informal) firms. We study informal microfirms for three reasons. First, previous studies have focused either on the formal sub-population or on the whole population. The proportion of the informal sector in the economy has increased steadily in the last decade in Mexico (Duval and Orraca, 2011). This fact suggests that in Mexico there are incentives for operating in the informal sector, contradicting the results of Fajnzylber et al. (2009), who, by using both matching and standard control function techniques, found positive effects of formality on the profits and survival rates of formal Mexican microfirms. McKenzie and Sakho (2010), on the other hand, by using instrumental variables and matching techniques on Bolivian data, found that effects of formality on profits depend on firm size. In particular, they found that, for small firms, lower profits are associated with registering for taxes; while middle-size firms might benefit from formalization. Second, Latin American governments have shown an increasing interest in reducing the size of the informal sector by providing incentives to informal microfirms to move into formality; therefore, it is interesting to explore how well informal firms, as they are, would perform under the rules of the formal sector. Our third reason is methodological. Although we use data from more recent waves of the survey used by Fajnzylber et al. (2009), the data still have the same problem that those authors encountered, which is the difficulty to satisfy the common support assumption for the treated sub-population; an issue that is less problematic for the untreated sub-population. In simple words, for each informal entrepreneur it is possible to find a similar (in observable) formal counterpart; the opposite, however, is not always true as there are types of formal entrepreneurs (e.g. very highly trained or educated ones) that cannot be found in the informal sector.

Our empirical results indicate that the main determinants of sector choice are the microfirm owner's level of formal education and access to credit, and the microfirm's type of economic activity. Even though the nature of our estimation is static, our results show that the right incentives for a microfirm's growth are only found in the formal sector of the economy, and that this growth occurs if the entrepreneur's education level and ability to obtain outside funds are adequate, as well as if the microfirm operates in the sectors of commerce and services. Moreover, when comparing the behavior of currently informal microfirms and their expected behavior were they to operate in the formal sector, we find that these microfirms would obtain an annual net profit in the formal sector that is lower than what they currently obtain in the informal sector, even after adjusting their capital and labor levels. Hence, those entrepreneurs would optimally choose to stay in the informal sector.

This article is organized as follows: in section I we describe the data we use and present some descriptive statistics. In section II, we present the econometric model and the methodology we use to quantify the impact of formality on net profits and the levels of physical capital and labor. In section III, we discuss our results. Finally, in the last section, we offer our concluding remarks.

  1. Data

    We employ data from the Mexican Microfirm National Survey (Encuesta Nacional de Micronegocios, Enamin) for the years 1998 and 2002. This survey is part of an additional questionnaire that comes with the National Survey of Urban Employment (Encuesta Nacional de Empleo Urbano, ENEU), where people who say they are owners of a firm with ten or fewer employees must answer the Enamin questionnaire, which gathers specific information about the characteristics of the microfirm.

    The numbers of registered questionnaires in Enamin-1998 and Enamin-2002 were 10 738 and 11 106, respectively. We use data from both surveys in our estimations and, because our tests of the existence of structural change between 1998 and 2002 were statistically rejected, in this article we present and discuss only the results of the joint sample, which comprises a total of 21 844 microfirms.

    Many informal microfirm owners, however, use the informal market as a source of subsistence while seeking a wage-earning job. Mexican informal firms are heterogeneous in the sense that a fraction of their owners are true entrepreneurs, while some owners are trying to earn a living while searching for a job with a paid salary and benefits, (Bruhn, 2012). This means that many microfirms have a very short life span, a situation that is not representative of microfirms owned by individuals who have actively decided to be their own bosses. In order to avoid this bias, in the final estimation we use only microfirms whose owners are 20 years old or more, who invest more than US $100 in capital, and who have been operating for at least 4 years. We decided to distinguish the true entrepreneurs from...

Para continuar leyendo

Solicita tu prueba

VLEX utiliza cookies de inicio de sesión para aportarte una mejor experiencia de navegación. Si haces click en 'Aceptar' o continúas navegando por esta web consideramos que aceptas nuestra política de cookies. ACEPTAR