A substantial literature documents the transformative effects of information and communication technologies on productivity and growth in advanced economies. Yet, dramatic cross-country differences in ICT abundance between rich and poor countries suggest that the benefits from the digital revolution may not have been shared globally. In light of this, there has been substantial discussion in policy circles regarding the need to remove barriers to technology adoption. For example, the World Bank’s forthcoming World Development Report, “Digital Dividends”, is devoted to this topic.
New research
In our recent paper (Eden and Gaggl 2015), we take a step back and ask whether ICT capital stocks are indeed too low in low-income countries. For this purpose, we estimate ICT capital stocks and compare them to benchmark levels based on capital-labour ratios, the relative prices of ICT and non-ICT capital goods, and industrial composition.
- Our benchmark levels of ICT reflect the view that cross country differences in productivity, purchasing power parity and specialisation should drive efficient variation across countries in ICT abundance.
Benchmark values of ICT capital stocks are lower in poor countries for three reasons:
- The overall capital-labour ratios are lower;
- The prices of ICT capital goods are higher relative to the prices of non-ICT capital goods; and
- Low-income countries tend to specialise in agriculture, which has a relatively low ICT capital intensity.
We find that income per-capita is roughly uncorrelated with the deviation of ICT from its benchmark level. This suggests that there are no systematic barriers to ICT adoption in low-income countries.
Measuring ICT abundance around the world
We begin our analysis by compiling a new dataset that allows us to measure ICT and non-ICT capital stocks for a sample of 70 countries at various levels of economic development. To construct our dataset we build on information provided by the Information Technology and Services Alliance as well as the International Telecommunication Union and combine these sources with data from the Penn World Tables and price data from the International Comparison Program.
We then document that, in low income countries, the value of ICT capital indeed represents a smaller share of the aggregate capital stock. The differences are even larger in real terms, as we find that ICT capital goods are relatively more expensive in low income countries (Figure 1). This suggests that the scarcity of ICT in developing countries is likely not explained simply by lower capital-labour ratios.
Figure 1. ICT abundance and income per person
Why do poor countries have less ICT?
We find a surprisingly simple answer to this question: after accounting for overall capital-labour ratios, differences in industrial composition explain the bulk of the cross-country variation in ICT per capita. For example, if agricultural products comprise a larger share of output in developing countries and agriculture is inherently less ICT-intensive than other industries, one would expect there to be relatively less ICT in developing countries. Systematically more expensive ICT goods (relative to other capital goods) in poor countries exacerbate this result and explain the remaining difference.
To illustrate our point, we use a simple theoretical framework in which ICT and non-ICT capital intensities differ by industry. This framework allows us to predict relative ICT abundance based on estimates of sector-specific value added for a wide range of countries and an estimate of the sector-specific ICT intensity based on data from the US. The exercise suggests that cross-country heterogeneity in industrial composition, and thus specialisation, predicts virtually all of the cross-country variation in relative ICT abundance. As an additional piece of evidence for this result we further document that there is no systematic relationship between income per-capita and measured ICT spending – a direct proxy for ICT abundance – within industry. This is in stark contrast to the well-documented fact that capital labour ratios within the same industry vary widely across countries.
This is an important insight, as it suggests that plausible mechanisms emphasised by previous research are perhaps less relevant in the case of ICT. One such mechanism is delayed technology adoption, which suggests that due to frictions in learning and adopting new technologies, low-income countries may be slower to accumulate ICT capital (e.g. Comin and Hobijn 2010, Gust and Marquez 2004). Another mechanism relates to the relative prices of capital and labour. If labour is to some degree substitutable with ICT capital (e.g. Autor and Dorn 2013, Karabarbounis and Neiman 2014), then low-income countries will likely opt for lower ICT capital stocks, as labour is relatively cheaper in these countries. Both of these explanations would suggest that, within a given industry, ICT should be relatively less abundant in poor countries – a hypothesis we reject.
In sum, our results suggest that the variation across countries in ICT abundance is predominantly between-industry variation rather than within-industry variation. This suggests that frictions associated with the accumulation of ICT are reflected in changes in industrial composition, rather than in changes in the production structure within industries.
Does India have too little ICT? An illustrative example
To illustrate our benchmarking procedure, consider the following example. According to the International Telecommunication Union (a body of the UN), internet subscriptions are 30 times more abundant in the US than they are in India. Is this 30-fold difference too large or too small? In other words, is ICT under-utilised or over-utilised in India?
Assuming that internet subscriptions are proportional to the physical abundance of ICT, our model suggests a benchmark level of ICT capital based on the capital labour ratio, the relative price of ICT and industrial composition. In fact, this benchmark ratio of ICT in the US relative to ICT in India is 63. Thus, for this specific example, an observed 30-fold difference in broadband subscriptions relative to the US is actually suggestive of ICT utilisation rates that are high compared to our benchmark. Put differently, given capital labour ratios, price differentials and industrial composition, India appears to be utilising ICT beyond what would be expected.
It is worth emphasising that the above benchmarking procedure does not utilise our estimates of ICT and non-ICT capital stocks. The procedure provides a benchmark for ICT prevalence, that requires as inputs only aggregate capital-labour ratios, value added by industry, and estimates of the relative prices of ICT and non-ICT capital goods. Since capital-labour ratios are available widely from the PWT and industrial composition is available widely from the World Bank’s World Development Indicators, the only input that is not readily available is the relative price of ICT. Our ICP-based estimates for relative prices are available for a wide variety of countries and can be utilised in this benchmarking procedure.
Of course, our analysis here ‘explains’ cross-country differences in ICT abundance only in an accounting sense. However, this accounting exercise illustrates that in order to understand the source of the differences, one must look deeper into the fundamental questions of development economics, and understand why capital-labour ratios are lower in developing countries – why non-tradables are relatively cheaper; and why production is concentrated disproportionately in agriculture and other industries that do not require much ICT capital. Thus, our cautious conclusion is that poor countries don’t necessarily ‘need more IT’.
References
Autor, D H, and D Dorn (2013), “The growth of low-skill service jobs and the polarisation of the us labour market”, American Economic Review 103: 1553–97.
Comin D, B Hobijn (2010), “An Exploration of Technology Diffusion”, American Economic Review100: 2031–59.
Eden M, P Gaggl (2015), “Do poor countries really need more IT? The role of relative prices and industrial composition”, Policy Research Working Paper Series 7352, The World Bank.
Gust C, Marquez J (2004), “International comparisons of productivity growth: the role of information technology and regulatory practices”, Labour Economics 11: 33–58.
Karabarbounis, L, B Neiman (2014), “The global decline of the labour share”, The Quarterly Journal of Economics 129: 61–103.
This article is published in collaboration with VoxEU. Publication does not imply endorsement of views by the World Economic Forum.
To keep up with the Agenda subscribe to our weekly newsletter.
Author: Maya Eden is an economist in the Macroeconomics and Growth Team of the Development Economics Research Group at the World Bank. Paul Gaggl is an Assistant Professor of Economics in the Belk College of Business at the University of North Carolina at Charlotte.
Image: A USB lead is plugged into a laptop. REUTERS.
