Assessment of the effect of broadband expansion on the economy reviewed

• Published date: 01 November 2021
• Author: Mário Jorge Mendonça,Paulo R. A. Loureiro,Antônio Nascimento,Roberto Ellery
• Date: 01 November 2021
• DOI: http://doi.org/10.1111/rode.12780

2414

|

wileyonlinelibrary.com/journal/rode Rev Dev Econ. 2021;25:2414–2432.

© 2021 John Wiley & Sons Ltd

1

|

INTRODUCTION

The telecommunications industry has experienced a profound change in the introduction of new

technologies. Inserted in this context appears, in particular, the dissemination of broadband services,

whose technology allows the transfer of data at high speed.1 Although studies seeking to evaluate the

effect of broadband on gross domestic product (GDP) are based on entirely distinct methodologies

Received: 24 June 2020

|

Revised: 18 March 2021

|

Accepted: 21 March 2021

DOI: 10.1111/rode.12780

REGULAR ARTICLE

Assessment of the effect of broadband expansion on

the economy reviewed

Mário JorgeMendonça1

|

Paulo R. A.Loureiro2

|

AntônioNascimento Jr.2

|

RobertoEllery Jr.2

1Instituto de Pesquisa Econômica Aplicada

(Ipea), Dirur (Ipea), Brasília, Brazil

2Universidade de Brasília- UnB, Brasília,

Brazil

Correspondence

Paulo R. A. Loureiro, University of Brasilia

– UnB, Campus Darcy Ribeiro, Asa Norte,

CEP: 70910- 900, Brasília, DF- Brazil.

Email: pauloloureiro@unb.br

Abstract

Despite consensus about the positive impact of broadband

expansion on the economy, from an aggregate perspective,

this impact does not seem to be homogeneous for the whole

economy. By using a two- way multivariate panel data

model subjected to Bayesian estimation, this study aims to

determine the effect of broadband expansion on Brazilian

economic sectors. To account for regional differences, the

sampled data were grouped into clusters (clustering) to spot

municipalities with similar characteristics. The results indi-

cate that the impact of broadband expansion is not positive

for all sectors. Actually, it was positive in more dynamic

and technology- intensive sectors (e.g., industrial and ser-

vice sectors) but negative for the agricultural sector.

KEYWORDS

broadband, clustering, Gibbs sampling, multivariate model, panel

data

JEL CLASSIFICATION

L69; O47; O57

|

2415

JORGE MENDONÇA Et Al.

and databases, it is widely agreed that broadband expansion has had a positive effect on the economy

by increasing access to information and communication technologies. These gains are due, among

other things, to increased productivity and innovation, which have a direct impact on economic growth

(Czernich etal.,2011; Koutroumpis,2009; Qiang & Rossotto,2009).

In addition to these effects of broadband on the economy, other important issues permeate this topic.

For example, based on the fact that the impact on economic growth is more felt in more technology-

intensive economies, one may question whether it is necessary to have a level of “critical mass” from

which the effect of network expansion can be felt. Evidence of this appears in Koutroumpis (2009),

which shows that network access is greater for countries with high penetration. Another interesting

point is whether there is a critical point from which network access shows decreasing returns. In fact,

Czernich etal.(2011) use the dummy variable technique to check this hypothesis. Dummies are indi-

cators of 10% and 20% of broadband network access. It was found that only the 10% dummy access

was statistically significant. In Brazil, studies have shown findings that are in line with the literature,

demonstrating that the impact of broadband expansion on output is significant and has a positive sign

(Carvalho etal.,2017; Fonseca,2015; Macedo & Carvalho,2010a,2010b).

However, even though the impact of broadband expansion on aggregate output is positive, this

impact is not homogeneous when one studies the different dimensions of the economy. One such

dimension concerns the differentiated impact of broadband on sectors of the economy. Stiroh (2002)

states that the use of information technology (IT) has substantially increased the productivity of the

American economy. The increase in efficiency is also felt most strongly in sectors with high trans-

action costs, such as the financial sector, or labor- intensive sectors, such as tourism. The effect on

economic growth seems to be related to the regions with the highest technological intensity. In less-

developed regions, the expansion of access to the network may lead to the incorporation of more effi-

cient production processes, resulting in the substitution of labor by capital. The effect of expanding the

network on small- and medium- sized companies is not immediate due to the need to restructure firms

to adapt to the new technology. The effect of expanding the broadband network seems to be most felt

when combined with stimulating innovation.

Katz (2012) claims that the impact is more widely observed in IT- intensive sectors or sectors with

high transaction costs (e.g., financial sector). Thus, due to the effect of economic growth in less-

developed regions, the expansion of network access leads to more efficient production processes, with

consequent replacement of labor with capital land and even unemployment. Dutz et al. (2017) con-

duct a comprehensive study in Brazil about the effects of the expansion of the Internet, a by- product

of broadband services, on the different branches of economic activity and find evidence of quite

heterogeneous effects on the sectors, with possible negative effects on employment and productivity,

depending on the segment of economic activity analyzed.

Another dimension that should also be analyzed is the impact of broadband associated with re-

gional differences. Kandilov and Renkow (2010) used a difference- in- differences approach to analyze

the effect of a broadband in U.S. rural areas, concluding that it did not have a significant effect on their

economic development (measured by employment, payroll, and business establishments). Whitacre

etal.(2014) using a similar approach and data for the U.S. counties from 2001 to 2010 conclude that

median household income, employment, and the number of firms increased faster in counties with

more broadband infrastructure adoption, whereas they experienced lower unemployment. In addition,

Whitacre etal.(2014) suggest that higher download speed is associated with less poverty and more

creative class employment.

For Brazil, Jung and López- Bazo (2020) using state (regional) data at county level analyze the

effect of broadband on regional productivity, intending to find out if the economic impact is uniform

across all territories in the country. The results show that the impact of broadband on productivity is