Poverty changes in Manaus: Legacy of a Brazilian free trade zone?

• Published date: 01 February 2019
• Author: Marta Castilho,Aude Sztulman,Marta Menéndez
• DOI: http://doi.org/10.1111/rode.12560
• Date: 01 February 2019

REGULAR ARTICLE

Poverty changes in Manaus: Legacy of a Brazilian

free trade zone?

Marta Castilho

1

|

Marta Menéndez

2,3

|

Aude Sztulman

2,3

1

Instituto de Economia, Universidade

Federal do Rio de Janeiro, Brazil

2

Université Paris-Dauphine, Université

OSL, IRD, LEDa, DIAL, PARIS,

FRANCE

3

IRD, LEDa, DIAL UMR 225, PARIS,

FRANCE

Correspondence

Marta Menéndez, Université Paris-

Dauphine, Laboratoire d'Economie

Appliquée (LEDa), DIAL Unité Mixte de

Recherche (UMR) 225, 75016, Paris,

France.

Email: Marta.Menendez@dauphine.fr

Abstract

This study contributes to the literature on the social impacts

of Special Economic Zones by analyzing poverty changes

in the Brazilian state of Amazonas, where the Free Trade

Zone of Manaus (FTZM) is located. Using census data, sta-

tistical micro‐decompositions and counterfactual simula-

tions, we show that labor income was the major driver of

poverty declines during the 2000 to 2010 decade in the

municipality of Manaus (and all the more so for households

with members working in the FTZM, who were less poor

to start with). Comparison with ex‐ante “similar”munici-

palities, in regard to socioeconomic and demographic crite-

ria, corroborates the relative success of Manaus in terms of

poverty reduction as well as the essential role played by

labor income. Wage regressions also show a significant

FTZM premium effect for workers, though diminishing

over time. In the rest of the state of Amazonas, nonlabor

income remains the main factor for poverty reduction and

the FTZM appears to have limited spillover effects, even at

short distances. Our contrasting results illustrate both the

benefits and limitations that trade and industrial policies

face in underprivileged areas and suggest that a better tar-

geting of social policies is needed to improve distributional

outcomes and spillovers for the whole state.

1

|

INTRODUCTION

The number of special economic zones (SEZs), that is, demarcated geographic areas within a coun-

try where the rules of business are different from those that prevail in the national territory, has

surged in the world, particularly in developing countries.

1

Despite this widespread use of SEZs as

DOI: 10.1111/rode.12560

102

|

© 2018 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/rode Rev Dev Econ. 2019;23:102–130.

a policy instrument to foster economic development, their impact on labor standards and human

development is still a subject of considerable debate. For example, some studies find how SEZs

might offer better job opportunities, particularly for more vulnerable groups (women, young peo-

ple, minorities and poor people in general), which may include higher wages (Amengual & Mil-

berg, 2008; McCallum, 2011) and better labor conditions (Aggarwal, 2007; Amengual & Milberg,

2008; Cling, Razafindrakoto, & Roubaud, 2005, 2009; De Hoyos, Bussolo, & Nunez, 2012; Glick

& Roubaud, 2006; Jenkins, 2005). These factors may lead to income poverty and/or inequality

reduction. At the same time, several case studies (Glick & Roubaud, 2006; International Labour

Organization, 2003, 2008; Kwaku Akpokavie, 2001; McCallum, 2011; Oberai, Sivananthiran, &

VenkataRatnam, 2001) find some troubling signals associated with the development of SEZs, such

as lax enforcement of labor laws (unrestrictive firing regulations and lack of social protection),

lower labor standards (with respect to social dialogue, freedom of association, collective bargain-

ing, etc.), high turnover owing to “sweatshop”working conditions in factories (working hours,

pace of work, health and safety conditions), and persistently low wage levels over time.

2

However, almost none of the abovementioned studies focuses on the ultimate effects of SEZs

on household poverty and indeed few are based on individual data (average sector data or aggre-

gate firm‐level data are more frequently used). Among the exceptions, Glick and Roubaud (2006)

and Cling et al. (2005, 2009), for Madagascar, do focus on workers’earnings using individual‐

level labor force surveys in the Malagasy capital (Antananarivo), and De Hoyos et al. (2012), for

Honduras, is one of the rare studies to look at the effects of SEZs on total household income pov-

erty. For example, Cling et al. (2005) find that, in the 1995 to 2001 period, firms in the Malagasy

Zone Franche offered wages as high as the formal private sector ones, while employees’working

conditions were in general more generous (with the exception of working hours).

3

The econometric

study of De Hoyos et al. (2012) also shows for Honduras that workers in the maquila sector

earned higher wages between 1991 and 2006 and the authors evaluate, through a partial equilib-

rium simulation exercise, that poverty would have been higher without the maquila sector.

Depending on the set of characteristics and the environment of each SEZ, the consequences for

poverty at a local or regional level can be very different.

4

Consequently, the role of SEZs in pov-

erty alleviation and, more generally, in income distribution changes remains an empirical issue.

Our aim is to contribute to the literature by analyzing poverty changes in the Free Trade Zone of

Manaus (FTZM), in its municipality, Manaus, and, more widely, in the Brazilian state of Ama-

zonas in the past decade, to investigate spillovers.

The FTZM in Brazil is an interesting case study to analyze the impact of SEZs on distributional

outcomes. First, the creation of the FTZM, in the center of the Amazon Forest, was designed from

the outset to contribute to the economic and social development of the Amazon Region, one of the

poorest regions in Brazil,

5

and to step up its political integration into the rest of the country. The

municipality of Manaus is an isolated enclave where the only means of transportation are the Ama-

zon River and airplane. Despite these problems with transport and market access (local, regional,

domestic and foreign), the creation of the FTZM in Manaus prompted the development of an

important industrial hub, integrated into international markets, that has posted ove r the last decade

particularly good economic outcomes. By way of illustration, between 2000 and 2010, gross

domestic product in Manaus posted high levels of growth (4.4% on an annual average basis),

resulting in major job creations.

6

Defenders of the FTZM argue that this industrial boom generated

economic gains as well as positive environmental and social effects. However, if the FTZM's good

economic performance in the last decade can be related to particularly relevant improvements in

poverty indicators in the municipality of Manaus, and more widely in the poor state of Amazonas,

still needs to be documented.

CASTILHO ET AL.

|

103