Low Latency Internet and Economic Growth: A Simultaneous Approach
19 Pages Posted: 7 Mar 2016 Last revised: 11 Aug 2016
Date Written: August 2016
Empirical studies on the impact of internet growth relationship suffer from limited data availability to assess long-run growth effects and usually cover only small sets of countries. The aim of this study is to introduce a novel measure for internet quality and show that it can be used in a conventional growth model. The latency of an internet connection can be measured directly and one does not rely on intergovernmental agencies for collecting the data.
Apart from the lack of availability some existing measures also have conceptual problems. For example, the penetration rate, i.e. the share of population using the technology, is easily available from the World Bank and therefore widely used. Comparing the penetration rate across countries is dangerous as the survey methodology differs across countries and it neglects any information about the quality and frequency of internet access. Consequently, any increase in the penetration rate in a developing country will result in a smaller “divide” between the industrialized and the developing country, despite an increasing divide in qualitative terms. One possibility is the use internet bandwidth per user (Rohman and Bohlin 2012), which provides information on one aspect of quality, but it relies on survey methods for data collection.
In my contribution I try to assess the suitability of latency as a proxy for internet quality to quantify the influence of internet quality on economic growth. To assess the suitability of the indicator I build upon the theoretical framework proposed by Röller and Waverman (2001). The model consists of a standard growth equation which is augmented by ICT capital, where the stock of ICT capital is endogenously determined by separate demand and supply equations. These equations are then simultaneously estimated. Different from Koutroumpis (2009) my research uses latency as an indicator of internet quality, instead of directly including the ICT capital in the growth equation. Additionally the data allows to consider a larger sample than existing papers. The latency data for the research at hand is obtained from the Stanford PingER (Ping end-to-end reporting project), where the latency of end-to-end connections between research institutes in different countries is measured. The remaining variables describing the state of the economy are obtained from the World Bank database while prices of broadband and the annual ICT investments are taken from the ITU database.
Preliminary research has provided evidence that latency provides useful insight into the internet quality of individual countries. Moreover, it can be shown that latency is indeed determined by supply and demand dynamics. There is ample evidence in the literature supporting the hypothesis that broadband internet access increases economics growth. It remains to be shown how the inclusion of latency into standard growth models can be used to reflect this phenomenon and how the endogenous modeling of internet quality can be used to solve problems of simultaneity with respect to income and internet supply.
Keywords: Economic Growth, Internet, Broadband, Operationalization, Simultaneous Equation Model, Latency
JEL Classification: C33, F43
Suggested Citation: Suggested Citation