This chapter provides literature review regarding relationship between infrastructure

This chapter provides literature review regarding
relationship between infrastructure and economic growth. It covers brief
definition of infrastructure, theoretical overview followed by empirical
literatures about the relationship between infrastructure and economic growth. 2.2. Definition of
infrastructure Infrastructure is a
broad concept which can often be defined differently but it is generally defined
as fundamental facilities and systems that are necessary in order to run its
economy. Grigg (1988) classifies infrastructure into 6 groups: road, transport
services, water, waste management, building and outdoor sports, and energy
production/distribution which clearly focuses on physical aspect of
infrastructure so called “hard” infrastructure or in other words, economic
infrastructure. On the other hand, World Bank (1994) divides infrastructure
into three categories: economic infrastructure, social infrastructure such as
education and health, and institution infrastructure such as financial system
and the system of government. Their definition includes institutions and
services aspect of infrastructure so called “soft” infrastructure as well as
the hard infrastructure. In this paper, the focus is on hard infrastructure. 2.3.
Theory OverviewThe importance of
infrastructure investment in growth literature is first popularized by Arrow
and Kurz (1970). Since then the role of infrastructure on economic growth has
been well documented in the number of literatures. In these literatures, it is crucial
to identify which type of infrastructure investment affects growth. The first
conventional channel was introduced by Aschauer (1989), they believe that public
infrastructure investments have significant contribution to the increase of
private productivity which results in economic growth. Furthermore, two new channels
were introduced to the literature by Agenor and Moreno-Dodson (2006) which are complementarity
and crowding out effects. Complementarity means, growth is encouraged by
private capital formation. Crowding out effect simply means, increase in infrastructure
stocks can potentially crowd out private investment in short run and situationally,
this negative effects can be found in long-run as well. In many studies, quantity of infrastructure investment
is often the center of the debate however, it is important to note that quality
of infrastructure can be equally important. That is, better
infrastructure enhances access to education and healthcare (Cockburn et  al., 2013) and the effect of better access to education
and healthcare goes back to infrastructure by increasing the efficiency of use
of infrastructure. In addition, low-quality infrastructure tends to have a high
cost of maintenance and could directly increases the production cost of certain
products. (U.S. International Trade Commission, 2009). However, quantity nor quality do not really matter if large inequality in
access to infrastructure services exist. Conversely, better access to
infrastructure services magnifies the positive impact of infrastructure
investment. Agenor and Moreno-Dodson
(2006) state that better access to infrastructure
services increases labour productivity by reducing the commuting duration and stress.  Increase in labour productivity can be
considered as one of the many positive externalities of public infrastructure. Other positive externalities include, higher
competitiveness due to reduced production costs, greater trade flows (Ismail and Mahyideen, 2015) and greater foreign direct investment inflows and outflows (Asiedu, 2002; Globerman and Shapiro, 2002) The
question of which subsector matters on which circumstance is very important as
investment on one type of infrastructure could have a positive effect while
others might have negative effect. Moreover, the impact varies depending on
country’s circumstances such as the stage of the development and geography
however they are often not pursuit in many studies. The amount of variables
that affect the impact of infrastructure on economic growth makes the empirical
research difficult and somewhat weak thus, many empirical studies seem to focus
on specific variables to get more accurate result. The next section reviews the existing empirical
research. 2.4. Empirical ResearchThere is a number of empirical
researches supporting the idea of infrastructure investment being one of the
main factor of economic growth. Their methods of analysing the impact of
infrastructure can be divided into two different ways, one is based on the
physical measures of infrastructure such as length of total roads, and other is
based on the numerical measures of infrastructure such as expenditure on
capital investment. A study by Devarajan et al. (1996) uses public
investment expenditure as a measure of infrastructure. They find that in
developing countries, misallocation of infrastructure expenditure happens
frequently which causes relationship between infrastructure and economic growth
to be negative. Furthermore, he concludes that there is a strong negative
correlation between public investment expenditure and economic growth in
developing countries. Sanchez-Robles (1998) supports the result of Devarajan et
al. (1996) with bigger sample. Moreover, Prichett (1996) explains that the
reason for negative impact is because developing countries tend to use public
investment for unproductive projects. Despite a number of empirical studies
support positive relationship between infrastructure and economic growth, these
studies show opposite result. This suggests that numerical measure such as
investment expenditure could be a poor measure. Canning (1998)
provided one of the most significant research in this field. His data set
includes physical infrastructure stocks such as telephones, telephone line
transportation, roads, paved road, electricity-generating capacity and rail
lines. The data is based on physical measures of infrastructure such as kilometers
of roads and number of telephone main lines. This is because, he argues that
numerical data does not necessary reflect the actual quantity of the
infrastructure provided nor efficiency use of investment whereas physical
measures provide the result of the investment spending. To be more specific,
difference in unit cost between the countries causes efficiency of capital
investment to be different as well. A study by Sanchez, et al. (1998) also concludes
that monetary data is not good proxy for infrastructure compared to physical
measure by using both measures. However, Canning then argues that the
difference in quality and efficiency of use across countries can still be
significant with physical measures therefore, he includes some measures of
quality and efficiency use of infrastructure. He then finds that out of five
physical infrastructure stocks, only paved road and telephones had notable
impact on growth.   In Canning’s later
study with Pedroni (2008), they examine the long-run impacts of different types
of infrastructure on economic growth. They find that infrastructure investment
generally causes economic to grow in long-run however, the result varies across
countries. They then look at the level of infrastructure stocks of each country
in relation to the optimal level. Their assumption is that if the level of
infrastructure investment is above the optimal, there will be negative effect
on the level of output and vice verse. Using physical measures of identical
hard infrastructure as Canning’s previous study (1998), they find that average
level of overall infrastructure investment is at around optimal level on a
global scale. However, if individual countries or subsectors were to look at, the
result varies. In fact, they find that on average telephones and paved roads
are provided at around optimal level but a large number of countries are
suffering from under-provision of electricity. This result suggests that
country specific studies are necessary especially for policy purposes. Calderon and Serven
(2004) examines the effect of quality and quantity of infrastructure stock on
long-run economic growth as well as income inequality. This allows them to
investigates whether investment on infrastructure stock can be used for poverty
reduction. Their panel data set contains 121 countries between 1960-2000. Their
infrastructure includes transport, telecommunication, power and some water sector.
The result shows positive relationship between infrastructure stock and
long-run economic growth and negative relationship with income inequality.
Although, lack of social infrastructure variables and missing some data
indicate that the results should be interpreted with caution. The impact of social
infrastructure is however, difficult to measure as some of the benefits are too
complex to express in the form of numerical data. Recent studies tend to
focus on specific country or regions as it could potentially generate policy
implications. A few studies have specifically focused on the role of
infrastructure in East Asia. For example, Seethepalli, Bramati, and Veredas
(2008) use panel data for 16 East Asian countries. They focus on five different
subsectors of infrastructure: energy, sanitation, water supply, transport, and
telecommunications. In order to make policy debates easier with the analysis, they
add five additional variables: geography, the quality of governance, income
levels, the degree of private participation in infrastructure and rural-urban
gap in access to infrastructure services. The study shows a positive
relationship between infrastructure investment and economic growth. However,
the five additional variables reveal the different impacts of each subsector on
different countries. The impact of roads infrastructure on per capita growth
contradicts with the quality of governance while telecom and sanitation
infrastructure boost economic growth in East Asian countries with better
governance and low inequality in access to infrastructure services. Out of five
additional variables, national income level and governance quality affect the
degree of growth to the level of infrastructure the most. High quality of governance
seems to be correlated with greater positive effect of sanitation, electricity
and telecommunication infrastructure investment on growth. A higher country
income level gives greater impact of sanitation, water and telecommunication. Greater
inequality in access to infrastructure for its part worsens the impact of
telecom and water infrastructure investment. He then compares the results
with other results from Sub-Saharan Africa sample (Estache et al, 2005) and
finds that the impact of telecom and electricity is large in both East Asia and
SSA. However, further comparison shows that the impact of roads is low and
sanitation is significant in East Asia but vice verse in Sub-Saharan Africa. Seethepalli
et al. (2008) results suggest that the differences in results between East Asia
and Sub-Saharan Africa can be explained by differences in his five additional
variables. In fact, his results show that the roads infrastructure is
insignificant in countries with better governance and higher income and
sanitation infrastructure is more significant in countries with higher income
level, better governance, higher private participation in sanitation and low
levels of inequality of infrastructure access. Given the results, it can be
concluded that East Asian countries have higher incomes, better governances,
less inequality of infrastructure access and higher private participation in
sanitation than Sub-Saharan countries. The main problem with Estache
et. al (2005), Seethepalli et al. (2008) and many other researcher’s studies is
the robustness of the results. This is because more than or close to 50% of
data observation for each variable is missing and overall dataset is not big
enough to be accurate. The attempts of aggregating the observations by both
authors meant to solve the problem of the robustness however, missing
observation in some sectors are still above 50% which suggests the needs for
better dataset.Straub and Hagiwara (2011) investigates
the level of infrastructure stocks in developing countries in Asia. They apply
two approaches, growth regressions and growth accounting to analyse the
relationship between productivity, infrastructure and growth. The study
concludes that the infrastructure stocks have been growing in the recent past and
shows positive correlation with economic growth. However, both quantity and
quality of infrastructure stocks are still not up to world averages. In
addition, cross-country regression shows a positive impact on per capita GDP
growth rate in most of the countries as the result of factor accumulation. On
the other hands, growth accounting approach suggests that positive impact of
infrastructure on total factor productivity can only be observed in few
countries. Caldero?n and Chong (2009)
investigate the impact of infrastructure development on economic growth in
Africa. Like many other studies, they use physical measures in the
telecommunications, roads and electricity. An instrumental variable estimation
was used, in order to avoid common econometric issues such as reverse
causation, time-specific effects and missing country. The study shows that
growth has positive correlation with the quality of infrastructure services and
faster accumulation of infrastructure stocks. The study also finds that Africa
is better off if they focus more on accumulation of infrastructure stocks rather
than on the quality of the existing infrastructure. Straub (2011) critically
analyses the existing macro-level literature on infrastructure investment and
economic growth. He then addresses the needs for more microeconomic data from
households and firm in order make the macroeconomic results more accurate. This
is because fair amount of infrastructure services such as water, energy and
telecommunication is consumed as final goods by households and as intermediate
goods by firms. (Prud’Homme, 2004). Moreover, he empathises the importance of
economic geography.  He believes that
inclusion of better and more realistic infrastructure sectors can generate
useful policy debates.Similarly, Snieska and
Simkunaite (2009) also evaluates the existing literature on the impact of infrastructure
investment on economic side of growth as well as the social side. They believe
that the most of existing literature lacks “unique methodology” to exam the impact
of infrastructure investment on economic growth. Their study finds that the
impact of infrastructure investment on economic growth in Baltic states varies
even though they are all in the same income group. This suggest that “regional
peculiarity” is crucial for analysing the relationship between infrastructure
and economic growth.

It is undeniable that
infrastructure investment plays a significant role in the growth of economy through
different channels as a number of studies provide the evidence. However, it is
clear that most of the studies face the statistical challenges such as missing
observations and a lack of established model.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now