A large empirical literature estimates the determinants of cross-country

growth differences. Recent studies focus on Bayesian model averaging (BMA)

to address the substantial model uncertainty inherent to growth regressions
(Rockey, Temple, 2016). These studies focus their analysis on identifying
growth determinants either for the growth period 1960–1992 or the growth
period 1960–1996. Using a new data set, we provide the first systematic
assessment of the stability of inferences on growth determinants across
growth periods. Our findings indicate that inferences are unstable across
growth periods, but determinants related to demography and education tend
to become important in recent growth periods.

Many variables have been suggested to determine economic

growth. Durlauf et al. (2005) tabulate 145 growth determinants that have been
discussed in the literature with the vast majority of them being statistically
significant at least in one study. However, many of these variables may appear
statistically significant due to specification
searching. Hendry (1980) and Leamer (1983) have pointed to the flexibility of
econometric research and the wide range of estimates that can be obtained for
a given coefficient of interest. Estimates that appear to be insignificant in the
research process may remain unreported, while estimates that are statistically
significant may be selected for presentation in an article. This specification
searching and selective reporting is the response of researchers to the
incentive system of academic publishing as discussed
by Ioannidis (2005) and Glaeser (2011) and it may be particularly prevalent if
model uncertainty is large with little consensus about what control variables
should be included in the regression model.

Recent empirical evidence indicates the prevalence of specification searching

in economics. For the top economics journals, Brodeur et al. (2016) find a lack
of p-values just above the significance thresholds of 0.05 - 0.10 and conclude
that researchers may have preferred (marginally) significant p-values
resulting in an abundance of p-values just below the significance threshold of
0.05. This indicates that authors may have searched across a variety of
specifications to turn their marginally insignificant estimates into statistically
significant findings. Similarly, Bruns et al. (2019) and Vivalt (2019) find
evidence for discontinuities in the distribution of published t-values, with
marginally significant t-values being overrepresented compared to marginally
insignificant t-values. Ioannidis et al. (2017) show for 159 research fields in
economics with more than 60,000 estimates that power is low in many
economics studies while the point estimates of these studies are frequently
statistically significant. This finding indicates that authors with sample sizes
that are actually too small to reliably detect the effect of interest may engage in
specification searching to exaggerate the point estimates in order to achieve
statistical significance.
These findings demonstrate the need for methods that reduce some of the
flexibility of econometric research by basing inference on the coefficient of
interest on a large set of models rather than on a single model or a small
selective set of models. The growth literature put very early emphasis on
model uncertainty to improve credibility and reliability of growth regressions.
Prominently, Levine and Renelt (1992) and Sala-i-Martin (1997) assess the
robustness of growth determinants with respect to different sets of control
variables based on Extreme Bounds Analysis (Leamer, 1983, Leamer,
Leonard, 1983). Fernandez et al. (2001b) and Sala-i-
Martin et al. (2004) pioneer the use of BMA to identify robust growth
determinants. Many studies followed in this vein including studies that
analyze the jointness of growth determinants (Doppelhofer, Weeks, 2009, Ley,
Steel, 2007), the relative importance of alternative growth theories
(Durlauf et al., 2008), the specific growth determinants for African countries
(Masanjala and Papageorgiou, 2008), and studies with a more methodological
focus (e.g. Ley, Steel, 2009, Eicher, Papageorgiou, Raftery, 2011).

Literature review
Denison (1962) affirmed that economic growth is the increase of real GDP or GDP per capita, an increase
of national product that is measured in constant prices. Economic growth is influences by direct factors
like for example human resources (increasing the active population, investing in human capital), natural
resources (land, underground resources), the increase in capital employed or technological
advancements. Economic growth is also influenced by indirect factors such as institutions (financial
institutions, private administrations etc.), the size of the aggregate demand, saving rates and investment
rates, the efficiency of the financial system, budgetary and fiscal policies, migration of labour and capital
and the efficiency of the government. There are four major determinants of economic growth: human
resources, natural resources, capital formation and technology, but the importance that researchers had
given each determinant was always different. Renowned economists provided, over time, the most basic
ingredients which appear in modern theories of economic growth.

The determinants of economic growth are inter-related factors influencing the growth rate of an
economy. There are six major factors that determine growth with for of them been grouped under
supply determinants and the other two are efficiency and demand. The four supply factors are natural
resources, capital goods, human resources and technology and they have a direct effect on the value of
good and services supplied. Economic growth measured by GDP means the increase of the growth rate
of GDP, but what determines the increase of each component is very different. Public expenditure,
capital formation, private or public investment, employment rates, exchange rates etc. have different
impacts on economic growth and we should take into account that these determinants have different
implications if the states are developed or not. There are also socio-political factors and events that have
a major influence on the economic advancement of a country. There are also differences between
economic and non-economic determinants. “Proximate” or economic determinants refers to factors like
capital accumulation, technological progress, labour and “ultimate” or non-economic sources refers to
factors like government efficiency, institutions, political and administrative systems, cultural and social
factors, geography and demography (Acemoglu, 2009).

https://ashraffeps.yolasite.com/resources/EuroMed/Fall2018/The%20main%20determinants
%20affecting%20economic%20growth.pdf

There is a large part of economic theory analyzing the causal relationship between exports and economic
growth. Certainly, since exports consist one of the main determinants of economic growth, an
increase of exports contributes to an increase of economic growth. However, there are also some
other indirect factors, which affect the causal relationship between exports and economic growth.
Ricardo in his study in 1817, notes that trade facilitates products output with a comparative
advantage in a country resulting to a higher level of national wealth. Recent empirical studies are less
convincing relating to the causal relationship between exports and economic growth, because the main
interest focuses on which methods are used for economic growth through trade expansion.

The basic a priori argument is that exports expansion contributes to economic growth increasing
the percentage of gross fixed capital formation and productivity factor. If there are incentives for
investments growth and technology advance the marginal productivities factors are expected to be
higher in exporting sector than the remain economic ones. Since the ratio of exports to gross
domestic product denotes an open economy index, a higher ratio indicates a relatively higher open
economy. On the other hand a lower ratio of exports to gross domestic product reflects to a
limited trade policy and a more close economy. Solow (1956) in his study suggests that the larger the
investment and saving rate are the more cumulative capital per worker is produced.

Tyler (1981) examining a sample of 55 developing countries resulted that exports and investments are the
main determinants of economic growth. New growth theories stress the importance of investments,
human and physical capital in the long-run economic growth. The policies, which affect the level of
growth and the investment efficiency determine the long-run economic growth. Theoretically, the gross
capital formation affects the economic growth either increasing the physical capital stock in
domestic economy directly, Plossner (1992) or promoting the technology indirectly, Levine and Renelt
(1992). Recently, many empirical studies emphasized in diversified role of private and public
investments in growth process. The public investments on infrastructure, in extent in which are proved
to be complementary to the private investments, can increase the marginal product of the private
capital, augmenting the growth rate of a domestic economy.

Κhan and Kumar (1997) supported that the effects of private and public investments on
economic growth differ significantly, with private investment to be more productive than public one.
Knight, Loyaza and Villanueva (1993) and Nelson and Singh (1994) confirmed that public investments
on infrastructure have an important positive effect on economic growth over the period 1980-
1990. Εasterly and Rebelo (1993) evaluated that public investments on transportation and
communications are positively correlated to economic growth, while there were negative effects of
public investments of state-owned businesses on economic growth. The effect of foreign direct
investment on economic growth is dependent on the level of technological advance of a host
economy, the economic stability, the state investment policy and the degree of openness. FDI
inflows can affect capital formation because they are a source of financing and capital formation
is one of the prime determinants of economic growth. Inward FDI may increase a host’s country
productivity and change its comparative advantage. If productivity growth were export biased then
FDI would affect both growth and exports. A host’s country institutional characteristics such as
its legal system, enforcement of property rights, could influence simultaneously the extent of FDI
and inflows and capital formation in that country.
Βlomstoerm, Lipsey, Zejan (1994) found a unidirectional causal relationship between FDI inflows
as a percentage of GDP and the growth of per capita GDP for all developed countries over the
period 1960-1985. Ο Zhang (1999) examines the causal relationship between foreign direct
investment and economic growth with Granger causality analysis for 10 Asian countries. The results of
this study suggested that there is a unidirectional causality between foreign direct investment and
economic growth with direction from FDI to GDP in Hong Kong, Japan, Singapore, Taiwan, a
unidirectional causality between exports and economic growth with direction from economic growth to
exports for Μalaysia and Thailand, also there is a bilateral causal relationship between FDI and GDP
for Kina and Indonesia, while there is no causality for Korea and Philippines.

Βorensztein, De Gregorio and Lee (1998) highlight the role of FDI as an important vehicle of
economic growth only in the case that there is a sufficient absorptive capability in the host economy.
This capability is dependent on the achievement of a minimum threshold of human capital. Μoudatsou
(2003) suggested that FDI inflows have a positive effect on economic growth in European Union
countries both directly and indirectly through trade reinforcement over the period 1980-1996.

https://silo.tips/download/the-main-determinants-of-economic-growth-an-empirical-investigation-with-
granger

Over time, there have been various theories about what determines economic growth. The first modern
school of economic thought is generally believed to be classical economics. Many of the ideas
fundamental to classical economics were proposed in 1776 by Scottish economist Adam Smith. The
book, entitled ‘The Wealth of Nations’, was first published during the British Agricultural Revolution, and
was considered a landmark contribution condoning the benefits of free market economies that
promoted the concept of laissez-faire4 and free competition. In his book, Smith asserted that the
maximisation of a nation’s wealth occurs when its citizens pursue their own self-interest. These ideas
were later extended by David Ricardo (1817) who put forward the labour theory of value. This theory
was based on the idea that in a competitive environment, the prices of goods sold have a tendency to be
proportionate to the labour costs associated with producing them. John Stuart Mill (1848) elaborated
this theory by putting it into the context of contemporary social environments.

Classical economics focuses on the relationship between the law of diminishing returns and population
growth (Jackson & McIver 2001). The law of diminishing returns contends that “as successive equal
increments of one resource (e.g. labour) are added to a fixed resource (e.g. land), beyond some point
the resulting increases in total output (marginal outputs) will diminish in size” (Jackson & McIver 2001,
p. 530). This concept is then considered alongside what is known as the ‘optimum population’ for a
society. This is where given an economy’s scare resources, the population of a nation grows to a point
that will yield the most income per person. Another attribute of classical economics was that the output
of an economy was assumed to be distributed among different social groups in accordance to the costs
borne by those individual groups in producing the output (Library of Economics and Liberty 2002).

However, as prices in the market are not always reflective of the ‘value’ (i.e. the cost of producing the
product), this suggested the idea that ‘value’ equates to how the person obtaining the product perceives
it, leading to the concept of supply and demand. This was known as Marginal Revolution in economics,
which formed to what became later known as neoclassical economics. Neoclassical economics is based
on the following principles: that people have rational preference among outcomes; individuals maximise
utility and firms maximise profits; and people act independently on the basis of full and relevant
information.

The Harrod-Domar model was developed during the 1930s, using savings levels and capital productivity
to explain economic growth within an economy. Independently developed by Sir Roy F. Harrod in 1939
and Evsey Domar in 1946, the model implies that long-run equilibrium economic growth is not natural
within an economy and is virtually impossible to achieve. Using the savings ratio, the capital-output
ratio, and the rate of increase within the labour force, this model suggests that if any of these
aforementioned variables were to shift from equilibrium, this would have a detrimental effect on the
economy in the form of growing unemployment or prolonged inflation. To achieve equilibrium
economic growth, the Harrod-Domar model focuses on the comparison between the natural rate of
growth (which relies only on the increase in labour in the absence of technological change) and growth
that is dependent on the saving and investment patterns of households and businesses. In the case that
savings ratios are increased, this would provide more funds for borrowing, allowing greater investment
in capital and therefore technological progress. A key assumption in the Harrod-Domar model is the
concept of ‘fixed proportions’ in which labour is unable to be replaced by capital in production.

It was after the Harrod-Domar model that the exogenous growth model was developed. One of the key
contributors to the exogenous growth model was Robert Solow, who in 1956 developed a model which
utilised all the basic assumptions of the Harrod-Domar model, 35 apart from the ‘fixed proportions’
concept. This allowed increased capital to be distinct from technological progress. The model builds on
the Harrod-Domar model by adding a productivity component. More specifically, Solow extended the
Harrod-Domar model by treating labour as a production factor. This was achieved by introducing the
concept of diminishing returns to labour and capital separately, and constant returns to scale when both
components were combined, as well as incorporating an exogenous technology variable that was
distinct from capital and labour. This model fitted available US economic growth data at the time, and in
1987, Solow was awarded with the Nobel Prize in Economics for his work on the model.