Productivity and resource misallocation in Latin America¹⁾

See Blyde and Fernandez-Arias (2004); Cole et al. (2005); Restuccia (2008); Daude and Fernández-Arias (2010); Ferreira, Pessôa, and Veloso (2013).

See Parente and Prescott (1994, 1999, 2002); Howitt (2000); Klenow and Rodriguez-Clare (2005).

See Bergoeing et al. (2005), Cole et al. (2005), Banerjee and Duflo (2005), Restuccia and Rogerson (2008) and Hsieh and Klenow (2009) propose an alternative explanation.

As the elasticity of substitution between plant value-added s increases, intermediate inputs become closer to perfect substitutes. At the limit, only the highest-productivity good is produced.

As reported by the Manufacturing Industry Database hosted by the NBER.

In the Section 4, we assess the robustness of the main results to alternative hypothesis about these two parameters.

The average TFPQ is taken via a geometric average:

The average TFPR is:

We also used information from the WBES for Argentina, Chile, Bolivia, Ecuador, Mexico, Peru and Uruguay. See Table A3.

Most of the official data reside in the national institutes of statistics. Therefore we relied on a network of research teams that helped us running the codes. The computations in each country were made by the following reseachers: Argentina programs were run by A. Neumeyer and G. Sandleris; in Bolivia by C.G. Machicado and J.C. Birbuet; in Brazil, by C. Ferraz; in Chile and with the WBES, by M. Busso, L. Madrigal, and C. Pagés; in Colombia, by A. Camacho and E. Conover; Ecuador, by C. Arellano; El Salvador, by J.P. Atal, M. Busso and C. Cisneros; Mexico, by P. Martínez; Uruguay, by C. Casacuberta and N. Gandelman; and in Venezuela, by L. Kolovitch. In all cases the computations were done using a common program and, as much as possible, the same variable and sample definitions was used, too. All codes are available from the authors upon request.

Interestingly, there is more productive heterogeneity in these two economies than in China as reported in Hsieh and Klenow (2009), regardless of the measure of dispersion employed. This may be, at least partly driven by the fact that Chinese data covers plants with revenues above US$600,000 and therefore excludes the smallest, possibly least productive firms.

It should also be noted that some of the dispersion is not due to real differences in productivities but to differences in quality within the sector which translate to prices.

In the Appendix Table A1 we provide other dispersion methods for the distributions of the three variables.

Duarte and Restuccia (2010) point to the lower degree of competition in the service sectors in relation to manufacturing as one potential reason why across countries there is more convergence to the world frontier in manufacturing than in the service sector.

See, Banerjee and Duflo (2005); Buera, Kaboski and Shin (2011); Moll (2012); Midrigan and Xu (2010); D’Erasmo and Boedo (2012).

The argument is that productivity determines size, with more productive firms growing to be larger, rather than the other way around: i.e., larger firms become more productive as a result of their size. Yet a positive relationship between total factor productivity and size can also be driven by economies of scale. This is because most methods of computing TFP assume constant returns to scale; therefore, increasing returns to scale would wrongly show up as higher TPF for bigger firms.

In particular, the survey asks “Is O_isc ‘No Obstacle’, ‘a Minor Obstacle’, ‘a Major Obstacle’, or ‘a Very Severe Obstacle’ to the current operations of this establishment?” where the obstacles O_isc are, for example, access to finance, labor regulations, functioning of the courts, etc.

The scales are based on the following variables: (I) Restricted Access to Capital: (1) Degree of obstacles for current operation: Access to finance; (2) Access to finance listed as a top 3 obstacle; (3) Has a line of credit or loan from a financial institution; (4) Has its financial statements certified by an external auditor. (II) Restrictive Labor Regulations: (1) Degree of obstacle for current operations: labor regulations; (2) Degree of obstacle for current operations: inadequately educated labor force; (3) Degree of obstacle for current operations: practices of informal competitors; (4) Labor regulations listed as a top 3 obstacle; (5) Practices of informal competitors listed as a top 3 obstacle; (6) Percentage of workforce unionized; (7) Labor regulations affect decisions of hiring or firing permanent workers; (8) % of workers declared for payroll taxes. (III) Bad Functioning of Courts: (1) Degree of obstacle to current operations: functioning of the courts; (2) Functioning of courts listed as a top 3 obstacle; (3) Agreement with: The court system is Fair, impartial and uncorrupted; (4) Agreement with: The court system is Quick; (5) Agreement with: The court system is Affordable; (6) Agreement with: The court system is Able to enforce its decisions. (IV) Detrimental Regulations and Institution Instability: (1) Degree of obstacle for the current operation: licensing and permits; (2) Degree of obstacle for current operations: customs and trade regulations; (3) Any regulation listed as a top 3 obstacle; (4) Degree of obstacle for current operation: political instability; (5) Degree of obstacle for current operation: corruption; (6) Degree of obstacle for the current operation: macroeconomic instability; (7) Any instability listed as a top 3 obstacle. (V) Unfair Taxation: (1) Degree of obstacle of taxes for the current operations of the establishment; (2) Degree of obstacle for the current operation: tax administration; (3) Taxation listed as a top 3 obstacle; (4) % of sales declared for corporate or sales taxes; (5) Establishment was visited and or inspected by tax officials; (6) Located in capital city (i.e., easier to monitor).

The code is available at Matias Busso’s webpage.

The views expressed here are those of the authors and do not necessarily reflect the opinions of the Inter-American Development Bank.