How important are mobile broadband networks for the global economic development?☆
Introduction
Throughout history, new technology has always been an important driver of productivity and economic development. We are currently experiencing a technological revolution based on ICT. One of the major innovations within ICT, during the last decade, is the use of mobile broadband. According to GSMA (2018) mobile broadband connections have increased from approximately 27 thousand in 2001 to 4.8 billion in 2017 i.e. an average growth of 113 percent per year.1
The use of data being sent via mobile networks has been increasing exponentially at approximately 65 percent on a year-on-year basis during the period, 2010–2015 (Coyle and Williams, 2011, Ericsson Mobility Report 2016). The basis for this development has been the introduction and expansion of 3G and 4G mobile network systems and the development of smartphone devices. According to Cisco (2017) wired devices accounted for 51 percent of global IP traffic in 2016, compared to 41 percent of Wi-Fi and 7 percent for mobile devices. Cisco forecasts that mobile devices will account for 17 percent of global IP traffic in 2021.
Despite the enormous expansion of mobile broadband, it is still unclear to what extent, it has contributed to global economic development. Previous research has shown that ICT has had a large economic impact in many countries (Brynjolfsson and Hitt, 2003, Oliner and Sichel, 2000, Röller and Waverman, 2001). However, most of these studies have focused on established technologies such as fixed telephone lines and computers. Only a handful of studies have focused on mobile technologies (see for example Gruber and Koutroumpis, 2011). As more data becomes available it has become increasingly easy to also study the impact of newer technologies. This paper investigates the macroeconomic impacts of mobile broadband based on econometric methods applied to a cross-country panel data set. The primary questions that will be investigated are:
- •
To what extent has mobile broadband affected macroeconomic development in terms of GDP globally?
- •
If there is an impact from mobile broadband, is it an effect of mobile broadband introduction and/or a gradual process along mobile broadband penetration?
Chinn and Fairlie (2007) pointed out that there is a digital divide across countries in personal computers and Internet penetration. Thus, our questions are important from a policy perspective, because if mobile broadband has an important economic impact many countries could leapfrog in their economic development. We therefore also investigate the impact from mobile broadband in high- and low-income countries and OECD and non-OECD countries.
The paper shows that mobile broadband is positively associated with GDP based on 135 countries (90 once controlling for capital, employment and human capital). Introducing a dummy variable for mobile broadband introduction in a difference-in-difference specification, there is evidence of an introductory effect from mobile broadband. Moreover, there is also a contemporaneous effect from mobile broadband penetration. Furthermore, based on moving averages, we find stronger and larger effects from five-year differences compared to first differences. This is an indication that a lagged effect from mobile broadband penetration on GDP also exists.
Based on a two stage model controlling for simultaneity and reverse causality, we find strong evidence that mobile broadband introduction and penetration causes GDP growth rather than vice versa. The results suggest that a 10 percent increase in mobile broadband penetration causes a 0.8 percent increase in GDP. Moreover, once we control for the years since mobile broadband was introduced, we find that the economic effect gradually decreases over time. Finally, we find that our results are robust once we distinguish between low- and high-income countries. However, the effect from mobile broadband on GDP is considerably larger and more significant in non-OECD countries compared to OECD countries.
The paper is organized as follows. In Section 2 we summarize findings from earlier research and position our study in the current literature. In Section 3 we present the methodological framework, in Section 4 we describe the data, in Sections 5 and 6 we present our results based on both a fixed effect and an instrumental variable approach. Section 7 provides robustness checks and Section 8 concluding remarks.
Section snippets
Related literature
Throughout the 1980s it was unclear to many economists to what extent information and communication technology (ICT) impacted economic growth at the macro level (Solow, 1987). However, ever since economic and productivity growth took off in the US economy in the mid-1990s, there have been a plethora of studies showing links between ICT and economic development (see for example Jorgenson et al., 2008, O'Mahony and Vecchi, 2005, Oliner and Sichel, 2000, van Ark et al., 2008). Most of these
Methodology: production function framework and econometric specification
The model applied in this paper is based on the framework of the neoclassical production function. The production function framework relates output to labor, capital, intermediate inputs and TFP. In this paper we measure output as GDP, which is obtained by deducting intermediate inputs from gross output and adjusting for subsidies and sales taxes. Assuming an augmented Cobb-Douglas production function, we have the following equation:where Yi,t is real GDP, Ki,t is
Data
The data used in this paper has been collected from a number of different sources. Data on GDP, employment and human capital were retrieved from the Penn World Tables (Feenstra et al., 2015). The Penn World Table publishes different GDP series (see Feenstra et al., 2015). This paper uses a measure where the levels of GDP have been constructed based on multiple PPP benchmark years and therefore correct for changing prices between these benchmarks. Feenstra et al. (2015) argue that this measure
Results and discussion
A fixed effects (FE) model controls for or partials out the effects of the country specific components. An alternative to the fixed effects model is the random effects (RE) model which is used when variation across countries is assumed to be random and not correlated with the dependent and independent variables in the model. Based on a Hausman test we reject the hypothesis that the random effects model is most appropriate and instead conclude that the fixed effects model is most appropriate.8
Simultaneity
The methods used thus far have determined a correlation rather than a causal effect of mobile broadband introduction and penetration on GDP growth.
One way of addressing simultaneity is by using instruments that are correlated with the explanatory variable but not with the error term. Some of the instruments proposed from earlier studies are tax credit for ICT investment and specific types of housing (Abramovsky and Griffith, 2006, Dettling, 2013). However, none of these instruments are
Robustness
This section tests the robustness of our results based on different country groups and additional data. Due to space limitations we will only include one version of the specification, our robustness checks are themselves robust to other versions of the specification, results of which are available on request. Table 9 shows the results for the different specifications for the log of predicted mobile broadband penetration once we control for the years since mobile broadband introduction (based on
Conclusions
A number of different studies have shown that ICT is closely connected to macroeconomic development in terms of GDP. Most of these studies have focused on ICT as a whole or established technologies such as fixed telephone lines and computers. This paper investigates the effect of a much more novel technology, namely that of mobile broadband, on GDP.
Mobile broadband is measured as a percentage of total connections. Mobile broadband connections are defined as SIM cards registered on mobile
References (68)
- et al.
Mobile and more productive? Firm-level evidence on the productivity effects of mobile internet use
Telecommun. Policy
(2016) - et al.
Initial conditions and moment restrictions in dynamic panel data models
J. Econ.
(1998) - et al.
International R&D Spillovers
Eur. Econ. Rev.
(1995) The role of the telephone in economic development
Telecommun. Policy
(1980)- et al.
Is newer better? Penn world table revisions and their impact on growth estimates
J. Monetary Econ.
(2013) The economic impact of broadband on growth: A simultaneous approach
Telecommun. Policy
(2009)- et al.
Economic growth, telecommunications development and productivity growth of the telecommunications sector: Evidence around the world
Telecommun. Policy
(2010) - et al.
Telecommunications and economic growth: An empirical analysis of sub-Saharan Africa
Appl. Econ.
(2012) - et al.
CEE telecommunications investment and economic growth
Inf. Econ.Policy
(1998) - et al.
Mobile, fixed-line and Internet service effect on global productive efficiency
Inf. Econ. Policy
(2007)
Economic impacts of mobile versus fixed broadband
Telecommun. Policy
The modern drivers of productivity
Res. Policy
Outsourcing and offshoring of business services: How important is ICT?
J. Eur. Econ. Assoc.
Competition and innovation: An inverted U relationship
Q. J. Econ.
Formulation and estimation of dynamic models using panel data
J. Econ.
Some tests of specifications for panel data: Monte Carlo evidence and an application to employment equations
Rev. Econ. Stat.
Broadband penetration and economic growth nexus: Evidence from cross-country panel data
Appl. Econ.
The economic impacts of telecommunications networks and broadband internet: A survey
Rev. Netw. Econo.
Trade induced technical change: The impact of chinese imports on innovation, diffusion and productivity
Rev. Econ. Stud.
Estimation and inference in dynamic unbalanced panel-data models with a small number of individuals
Stata J.
Computing Productivity: Firm level Evidence
Rev. Econ. Stat.
The determinants of the global digital divide: A cross-country analysis of computer and internet penetration
Oxf. Econ. Pap.
Cisco Visual Networking Index: Forecast and Methodology, 2016–2021
Growth and human capital: Good data, good results
J. Econ. Growth
Communication Capital, Metcalfe's law, and U.S. Productivity Growth
Broadband infrastructure and economic growth
Econ. J.
Telecommunications and economic growth: A panel data approach
Appl. Econ.
Broadband in the labor market: The impact of residential high-speed internet on married women's labor force participation
The dynamo and the computer: An historical perspective on the modern productivity paradox
Am. Econ. Rev.
Ericsson mobility report – on the pulse of the network society
Cited by (41)
How is mobile broadband intensity affecting CO<inf>2</inf> emissions? – A macro analysis
2024, Telecommunications PolicyWhat would it cost to connect the unconnected? Estimating global universal broadband infrastructure investment
2023, Telecommunications PolicyICTs quality and technical efficiency: An empirical analysis
2022, Telecommunications PolicyThe economic impact of mobile broadband speed
2022, Telecommunications PolicyCitation Excerpt :To investigate the association between mobile broadband speed and labor productivity, the methodology in this paper is based on econometric methods. There are a number of studies that have investigated the economic impact of fixed and mobile broadband penetration (see Czernich et al. 2011; Edquist et al. 2018; Koutroumpis, 2009; Thompson & Garbacz, 2011). Moreover, there are also a few studies that have used econometric methods to investigate the impact of fixed broadband speed and economic development.
Growth and growth disparities in Africa: Are differences in renewable energy use, technological advancement, and institutional reforms responsible?
2022, Structural Change and Economic DynamicsConcentration of the mobile telecommunications markets and countries’ competitiveness
2022, Telecommunications PolicyCitation Excerpt :Similarly, Grimes, Ren, and Stevens (2012) confirm that broadband adoption increased the productivity of firms in New Zealand by 7%–10%. More recently, Edquist, Goodrige, Haskel, Li, and Lindquist (2018) corroborate the significant effect of mobile broadband introduction and diffusion on GDP. Using a two-stage model, they find that a 10% increase in mobile broadband adoption increases GDP by 0.8%.
- ☆
We are grateful for useful comments and suggestions by Leonard Nakamura and the participants at our presentation at the 35th IARIW General Conference, Copenhagen 21–24 August 2018. We also gratefully acknowledge financial support from the ERSC, grant number, ESRC IAA PSB111_MSRH and Ericsson Research for funding the research project “The Economic Impact of Communication Equipment”.