Introduction

The differences in living standards and income between countries in the world today is simply astounding. The World Bank estimates that in 2013, 10.7 percent of the worlds population lived on less than US$1.90 a day compared to 12.4 percent in 2012. That’s a 35 percent decrease since 1990. For the last half-century, we have seen a rise in growth and living standards. For the richest countries, this rise has been going on for more than a century. Britain and France have experienced large growth rates, however, some countries have failed to keep up, such as Argentina, which was one of the wealthiest countries in the 20^th century. Miraculous growth became increasingly dramatic, with countries like Japan and South Korea making the transition from bankrupt to industrial power in a single generation. On the other hand, other countries has been insusceptible to this growth and have remained desperately poor. These differences among countries pose questions as to why some countries are rich and others poor and what is the source of this growth?

Factors that point to and perhaps lead to these gaps in growth have been widely investigated. When discussing what determines the rate of economic growth, our first assumption is that each country is different, and the effects of these factors will vary. Development strategies have been closely investigated, to increase growth in developing economies. For example, experience has shown that increases in labour productivity has resulted in high rates of growth. Thus, this dissertation focuses on the factors or determinants of economic growth, specifically in developing countries.

Different models of economic growth stress diverse causes of economic development. We begin by discussing a few theories that describe and explain sources of growth.These include, the Neoclassical Growth model and Solow model. These stress the importance of supply side-factors such as labour productivity, savings rate and capital inputs as determinants of growth. We also discuss the Endogenous Growth theory, which concentrates on the strong influence of factors such as human capital and technological innovation. There has been much research that tests these theories and the determinants of growth that they strongly support. We will be discussing these literatures and whether these apply in developing countries.

This paper examines a few of these variables pointed out in past literatures and economic growth theories. These variables include factors such as inflation, government expenditure and exports. The study tests 35 developing countries over a 10-year period from 2007 to 2016. By using a panel regression, this research tests the impacts of these variables on GDP growth. As it is a known assumption that the impact on an economy varies this paper further investigates if these effects differ depending on the level of income of these economies. The sample is grouped by their income level (high, intermediate, low) to study this assumption.

Literature Review

Economic growth is the most powerful instrument for reducing poverty and improving the quality of life in developing countries. When discussing economic growth there are a number of different growth models that argue and stress alternative causes and determinants of economic growth. Here we focus on journals and literatures that contribute to this debate and test these theories.

The Solow Model

We can start by looking at The Solow Model. Robert Solow (1987) developed the neo-classical theory of economic growth. Solow builds on the classical model of the three basic determinants of economic growth that are labour, capital, and technology. Solow recognised aggregate savings as a way to increase economic development. He argued that funds from savings can be borrowed and used for investment purposes to increase capital.

In the simplest version of Solow’s growth model, the economy is closed, which means domestic savings is equal to investment and there is no technological change. Without technological change, there is no growth of output per worker. Along the steady growth path, the growth rates of labour and capital are equal therefore capital per worker is constant. As technological progress is constant, output per worker is also constant. Output can grow but only at the rate of population growth. In Solow’s growth analysis, he describes a ‘steady state point’ where savings generates just the right amount of investment to stay on the balanced growth path. If capital per worker is less than the steady state level. Investments exceeds the amount needed for balanced growth and the capital per worker rises. Hence, the economy tends towards its steady state.

Common known facts about developing countries is the lack of ability to finance their investments with national savings. Ismet Gocer et al. (2016) study the effect of the saving-investment gap on economic growth. They state that developing countries need to close the financing gap in investment with their domestic savings in order to sustain stable economic growth. This study uses clustering and panel data analysis to observe the relationship between economic growth and savings using a sample of 65 countries between the period 1981 to 2014.

In this study, countries were categorized into the two groups by using K-means clustering procedure; countries with negative saving investment gap in 2013; and countries with positive saving investment gap in 2013. Cluster 1 describes that the impact of savings on economic growth in countries which have negative saving investment gap is negative and statistically insignificant which was an expected result. This indicates the assumption that developing countries are unsuccessful at channelling their savings into investment. They state in their study that another reason for these results is that institutional structure in developing countries are insufficient in order to increase the level of savings and in turn, increase economic growth. Looking at Cluster 2, they found that the effect of savings on economic growth in developing countries with positive saving investment gap is positive. The fixed effects coefficient estimate of savings suggests that a 1% increase in savings elevates the economic growth by 0.08%. As cluster 2 consists of countries which possess savings over investment, positive and statistically significant relationship can be credible. An additional cluster was analysed that consisted of all 65 countries. In cluster 3, the coefficient is positive and statistically insignificant. As a result, this study succeeded in obtaining accurate results in estimating the effect of savings on economic growth and using these results, they were able to summarise that developing countries are required to take precautions to enhance saving rate to get sustainable debt structure and economic growth.

One conclusion made from Solow’s growth model is that the growth rate does not depend on the savings rate. Output rises due to an increase in savings, but the growth rate always returns to the balance growth rate. Therefore, increased savings benefits the economy in developing countries by raising GDP per capita, but only temporarily. It does not increase the long-term growth rate. The only factor that generally matters is the growth of labour input or in other words, population growth.

Changes in population growth affect both the consumption needs of an economy and the productive capacity. An increase in population growth means an increase in the number of working population. As labour is one the key determinant of growth, it is clear that an increase in the labour force can increase total output on a large scale. Crenshaw et al. (1997), studied effects of age-specific population growth rates on economic growth in 75 developing countries over the period 1995 t0 1990. They find that a rise in the child population hampers economic progress, while an increase in the adult population stimulates economic development. The study found that Third World countries that have dramatically lowered their fertility rates, such as South Korea, are now enjoying rapid growth in the working population without the burden of a rising child population.

The idea that there must be other determinants that either drive or hinder economic growth must be taken into account. There have been a number of equally important contributions to economic growth literature that focus either on the impact on economic growth and on the important factors and sources of economic development.

Neoclassical Growth Model

We can continue extending the neoclassical growth model, and further investigate additional determinants of growth. Barro (1997) investigated the determinants of economic growth for over 100 countries covering the period 1960-1985. Using a panel regression and three stage least squares method, the evidence indicated that the growth rate of real per capita GDP is stimulated by better focus on the rule of law, decreased government consumption and lower inflation. Low inflation helps promote stability and encourage investment. If an economy experiences periods of high and volatile inflation rates, then rates of economic growth tend to be lower. The study also found that growth is also boosted by higher levels of life expectancy and of male secondary and higher schooling, lower fertility rates. This essentially raises the labour force which in turn increases labour output.  In this study, he found that infrastructure investments, R&D outlays, the quality if education and the distribution of income and wealth are also likely to be important aspects of growth. However, further investigation is required to obtain reliable results in discussing these elements.

Gaurav Agrawal (2015) discusses the effect of Foreign Direct Investment (FDI) as a determinant of economic growth in the five BRICS economies namely, Brazil, Russia, India, China, and South Africa over the period 1989 to 2012. Neoclassical economies described the benefits of FDI as a stable source of foreign financing within the balance of payments. In recent years, much research has concluded that an increase in FDI leads to higher growth rates in developed countries. It has been claimed that foreign direct investment can create employment, increase technological development, and improve the economic condition in developing economies. FDI positively effects an economy and is considered one of the principal determinants of economic growth and development. Agrawal was able to prove this assumption in his study.

In order to analyse the FDI-led-growth hypothesis, this study performed three steps: (1) to test for stationarity or the order of integration; (2) test for cointegration; and (3) test for direction of causality. These three tests were conducted at panel level. Conducting a Pedroni’s Test, the results indicated the existence of long run cointegration relationship of between economic growth and FDI on the panel of BRICS economies. The test results showed that 5 out of 7 of Pedroni’s statistics significantly reject the null of no cointegration which implies existence of a long-run co-movement of FDI and economic growth. This also shows that there is possibility of causality between FDI and economic growth in BRICS economies. The study was able to observe a positive correlation between growth and foreign direct investment.  A conclusion was made that suggested that efforts should be made to encourage other potential factors of economic development that would in turn enhance and stimulate foreign direct investment.

Nevertheless, this study only focused on the five BRICS economies, and does not go further in depth on the impacts on other developing countries, therefore it is very limited when applying the effects of FDI on growth in lower developing countries. In contrast to Agrawal (2015), some approaches also argue about the limitations and risks of increased foreign direct investment in an economy. The Keynesian approach believes that free-market itself cannot ensure efficiency in the economy. For instance, information is not always perfect. In fact, incorrect or not enough information may attract the wrong kind of investment. In some cases, the interests of the investors may not coincide with the interests of the receiving economy. Increased Competition from foreign investors may also have a negative effect. It may push out local corporations as they are unable to compete. Therefore, rather than creating employment, the result would be lost jobs. To rectify this, some economists suggest the idea that government protection is needed to protect local activities and corporations.

In neoclassical development theory, the benefits of outward-orientation is described as a powerful source of economic growth. Outward-orientation allows the economy to use external capital for growth and development without encountering problems in servicing the corresponding debt. It is also considered to result in rapid growth of exports. Much research has shown that an increase in exports, combined with easy availability of capital, accelerates technological progress in developing economies. Exports expand aggregate demand and encourage full employment of resources. The revenues earned from exports enhance consumption and stimulate technological advancements.

Dollar (1992) investigated the sources of economic growth in 95 developing countries covering the period 1976-1985. This study develops a technique for estimating a cross-country index of real exchange rate distortion, using international comparison of prices prepared by Robert Summers and Alan Heston. The price data compiled by Summers and Heston showed that Asian developing economies are more outward oriented than their counterparts in Africa and Latin America. After controlling for differences in levels of development, Asian economies seem to have low price levels, while Latin American countries have moderately high price levels and African economies have extremely high price levels. High price levels indicate strong protection and incentives geared to production for the domestic market, whereas low price levels reflect relatively modest protection and incentives oriented to external markets. Asian economies also show evidence of low variability of these real exchange rate distortion measures, however Latin American economies have been overwhelmed with a high degree of volatility. Using these results, he constructed an index of outward orientation. This outward orientation measure is correlated with per capita GDP. For the period 1976-85, the most open quartile of countries had a per capita growth rate of 2.9%; the next quartile, 0.9%; the third quartile, -0.2%; and the most closed quartile, – 1.3%. These results strongly show that trade liberalization, devaluation of the real exchange rate, and maintenance of a stable real exchange rate could improve growth performance in many developing economies. The study therefore concluded that outward-orientated development or in other words, export-orientated policies, has an influential effect on accelerating technological advancements in an economy. Dollar (1992) states that this can be achieved through a low degree of protection and stable real exchange rate regime.

Similarly, a study by Alberto Gabriele (2006) discussed the relation between GDP growth and exports of goods and services in developing countries. In contrast to Dollar (1992), Gabriele (2006) found that export-oriented services activities in developing countries are usually under the control of foreign economic agents. This means that in most cases, they tend to be poorly integrated to the domestic economy concluding that outward-orientation may not sufficiently boost economic growth.

In Kappel and Ghani’s (2003) study on the influence of openness as a determinant of growth in developing countries concluded that liberal trade policies are recommendable, but must be complemented by sound macroeconomic management and micro-policy to reinforce competition and institutional improvements.

There are two disadvantages of outward orientation. The first is that it may be difficult for developing economies due to the competition from the more established corporations in developed countries. Another disadvantage is that developed economies provide a high level of protection for their industries producing simple labour-intensive products in which developing economies already have or can soon acquire.

Knight, Loayza, and Villanueva (1993) tested the neoclassical theory of economic growth and employs a technique for using a panel of cross-sectional and time series data for 98 countries over the 1960-1985 period to determine the quantitative importance for economic growth of both country-specific and time-varying factors such as human capital, public investment, and outward-oriented trade policies. They found that the Solow-Swan model’s predictions are consistent with the evidence. This includes the positive effects of savings ratios and public investment on developing countries. The study concluded that physical capital, human capital, public investment, openness to trade and population growth are all key determinants of economic growth.

Endogenous Growth Theory

Moving further, a more modern growth theory, namely the Endogenous growth theory can be debated. This is an area of macroeconomic research that has been very influential since the 1980’s. Endogenous growth theory believes that investment in human capital, innovation, and knowledge contribute to economic growth.

Human capital is associated with the study of human resources management. As labour is one of the three most important determinants of economic growth, economists have found that the quality of labour that a person supplies varies. The differences in the quality of workers have been used to explain differences in income among countries. To study the effects of human capital, economists have studied characteristics of people that enable them to produce more output. In developing economies, intellectual ability is far more important than physical ability in determining a person’s wage. For this reason, investment on education has become one of the most important form of investment in human capital. When discussing the effects of education on economic growth, the influence of human capital becomes strong when the focus turns to the role of school quality.

Eric A. Hanushek (2013) studied the role of human capital as a driver of economic growth. Using assessments provided by the development of international assessments of math and science, the study presented a simple model of long run growth over the period 1960-2000 for the set of 50 countries with required data on growth, school attainment and achievement. These assessments provide a common metric for measuring skill differences across countries, and they provide a method for testing the approaches to modelling growth. This basic model shows a significant relationship between school attainment and growth and explains one-quarter of the international variation growth rates. The study concluded that difference in economic growth across countries are closely related to cognitive skills as measured by achievement on international assessments of mathematics and science. This shifts attention to issues of school quality. Research has shown that just providing more resources to schools is generally futile. Evidence from the growth analysis shows that providing schools that are unable to teach basic skills provides no benefits, therefore, slowing the pace of the provision of schools to a rate that also allows the development of quality schools, appears to be a good solution.

There are a few endogenous growth theories that stress the influence of stock market and taxation as determinants on economic growth. Stock markets improve firm efficiency by eliminating the premature withdrawal of capital from firms. This stimulates the growth rate of human capital and per capita output. Stock markets can also increase growth by raising the fraction of resources devoted to firms. Levine (1991) constructed an endogenous growth model where stock market emerges to allocate risk and explores how the stock market adjusts investment incentives in ways that affects steady state growth rates. In this model, stock markets improves economic growth by enabling the ability to trade ownership of firms without unsettling the productive processes within firms. It also allows investors to hold diversified portfolios. Tax policy drives growth by altering investment incentives.

Using this model, Colombage et al. (2015) investigated ten Developing Asia nations over the period 1990 to 2008 to test this theory. The study employs panel unit root tests, Pedroni cointegration tests and panel Granger causality tests to estimate both short and long run causal relationships between stock market, tax revenue and economic growth. They concluded that this study found that stock market and tax revenue do have an impact on economic growth even though the impact might be very small or sometimes even negative. This study only investigated the effects on growth in ten nations and therefore may not be a good representative for explaining the effects of stock market and taxation on growth.

A decrease in tax rates has the effects of reducing government revenues, at least in the short term, and creating either a budget deficit or increased government debt. This is because government spending would most likely have to be financed through borrowing. This would therefore result in a decrease in government spending and would negatively affect growth in developing countries. Many economists argue that tax cuts would only be successful in the more richer and developed countries but for developing economies, tax cuts may lead to cuts in services and in turn reduce growth in the economy.

Contrary to the neoclassical argument, endogenous growth theory argues that government plays a significant role in generating growth both in the short and long run. A fiscal deficit occurs when a government’s total expenditure exceeds the revenue generated. A higher fiscal deficit implies increased government borrowing and high debt which results in government cut backs in spending. This reduces growth in human and physical capital. Thus, reducing deficits would be a solution that would stimulate economic growth. Adam and Bevan (2005) examines the relationship between fiscal deficits and growth for a panel of 45 developing countries. A fiscal deficit occurs when government expenditure is greater than the revenue the government generates. The results showed that reduced deficits appear to stimulate growth, however, the level of growth depends on how changes in the deficit are financed. Deficits positively effect growth if financed by government profits and likely to negatively affect economic growth if financed by domestic debt. If the deficit is financed by external loans at market rates, then it would result in opposite flow and stock effects.

When discussing the role of government in stimulating growth in an economy, we can also debate the impact of growth in government size on economic growth. The budget or number employees simply measure a government’s size. Gusesh (1997) investigates this theory and estimates a model by ordinary least-squares regression method using annual time-series data for the period 1960–85 for fifty-nine middle income developing countries. The results suggest that growth in government size has adverse effects on economic growth in developing countries.

The assumption of increasing returns to capital of endogenous growth model also implies that foreign aid may improve growth in the long-run. As the model stresses the role of human capital as a determinant of economic growth, it justifies the assessment of foreign aid building up human capital in receiving countries. Aid, in the form of both technical assistance and investment in education and health care systems potentially fosters human capital leading to increased production. Karras (2006) investigates the relationship between foreign aid and growth in per capita GDP using annual data from the 1960 to 1997 period for a sample of 71 aid-receiving developing economies. Results showed that the effect of foreign aid on economic growth is positive and statistically significant. The study found that raising foreign aid by $20 per person of the receiving country results in a permanent increase in the growth rate of real GDP per capita by approximately 0.16%. Using an alternative foreign-aid measure, a permanent increase in aid by 1% of the receiving economy’s GDP permanently raises the per capita growth rate by 0.14% to 0.26%. Likewise, Nwaogu and Ryan (2015) found that foreign aid have a statistically significant contribution to economic growth in their study of 53 African and 34 Latin America and Caribbean Countries.

Contrastingly, studies have shown that foreign aid may lessen growth in developing economies rather than stimulate growth. Rahanna et al. (2017) found that the relationship between foreign aid and economic growth differs between high-income developing countries (HIDC’S) which are at latter stages of development and low-income developing economies (LIDC’S). They find that foreign aid has positive effects on growth in HIDC’S and negative effects on growth in LIDC’S. They suggest that countries need to gain some ‘traction’ before foreign aid is accepted.

Data

From these literatures we can see that the common assumption is that the majority of the key determinants labelled by growth models are significantly associated with economic growth. In this study I aim to test these theories by looking at a few of these variables studies in previous literatures and their impact on GDP growth in 35 developing economies. These variables include, Inflation rates, exports, government consumption expenditure, gross capital formation, foreign direct investment, and population growth. This study uses data from the period 2007 to 2016. Even though an immense amount of earlier data is available, to keep the results of this data reliable to realities in the present, I use a recent period of a smaller range.

In this study, the sample has also been separated into three groups: 15 economies with the highest income, 10 economies with middle or average income among the sample, and 10 countries with the lowest income. Separate regressions are run with these three groups to test whether the level of significance these variables have on economic growth is dependent on the income or how developed these economies are.

In order to keep results reliable, all data has been collected from World Bank. The World Bank group is a unique global partnership that fights poverty worldwide through maintainable solutions. The World Bank Group supports countries to share and apply innovative knowledge and solutions to create sustainable economic growth. Their Access to Information policy created initiatives such as World Bank Open Data which provides free and open access to global development data. It provides a listing of available World Bank datasets, including databases and other resources.  The DataBank provides an analysis and visualisation tool that contains time series data on a wide range of topics.

Key Variables

GDP Growth is used as a dependent variable. GDP or Gross Domestic Product is the total value of all finished goods and services produced in an economy in a year. It includes any product taxes, and deducts any subsidies not included in the value of the products. The higher the GDP, the higher the economic performance of an economy. GDP is the sum of value added, measured at constant prices, by households, government, and industries in the economy. Hence, the GDP Growth rate is the rate at which an economy’s GDP changes or grows from one year to another. Using GDP growth rate will allow the investigation of the effects on economic growth.

The inflation variable is measured by the consumer price index (CPI) which reflects the annual percentage change in the cost to the average consumer of acquiring a basket of goods and services.  CPI is a weighted price index where changes in weights reflects changes in the spending of households.  It is known to be less volatile which removes the risk of volatility from the dataset. Inflation is described as an increase in the cost of living or price level which in turn leads to a decrease in the worth or purchasing power of money making it difficult to make economic improvements. This study uses inflation to measure the effects of changes in inflation on economic growth and main hypothesis is that inflation will have a significant effect on GDP growth.

Exports of goods and services describe the value of all goods and services that are provided to the rest of the world. These include merchandise, transport, insurance, and services such as, communication and construction. It commonly described as a function of international trade whereby goods and services are traded between one economy to another. The sale of such goods contributes to an economy’s gross output, thus, in this study, exports is measured as a percentage of GDP. The hypothesis regarding exports is that it is expected to have a strong influence on GDP growth.

Another variable used in the regression is Foreign Direct Investment (FDI) as a percentage of GDP. It shows net inflows in the reporting economy from foreign investors. It is an investment made by a cooperation in one country in business interests in another, in the form of either founding business operations or acquiring assets in another economy. The key features of FDI is that it is an investment made that establishes either control or influence over the decision making of a foreign business. It is a common and widely known assumption that FDI largely contributes to economic growth, thus, I will be testing this theory to see if this is the case in developing economies.

Government Consumption Expenditure (GCE) includes all government spending and purchases of goods and services as a percentage of GDP. It includes the value of goods and services produced by the government and any purchases by the government of goods and services produced by market producers that are supplied to households. Increased government spending is likely to result in a rise in aggregate demand which in turn can lead to higher growth in the short-term.  However, this depends on how the increase in spending is financed. If it is financed through higher taxes, then there will be no increase in aggregate demand as the tax rise will counter-balance the higher spending. Many economist, on the other hand, argue that increased government spending would result in a more efficient allocation of resources therefore boosting economic growth. Using GCE in my regression will test this assumption to show whether this positive relationship between growth and GCE in developing economies.

Another important variable used in my regression is called Gross Capital Formation. It includes any additions to the fixed assets of the economy plus net changes in the level of inventories. Fixed assets refer to capital stock such as equipment and machinery, tools and transportation assets. Inventories refers to stock of goods held by companies to meet temporary changes in demand. Commonly, the higher the capital formation of an economy, the faster an economy can grow. Gross capital formation is also calculated as a percentage of GDP to test the impact on an GDP growth.

The final variable in the regression is Population Growth measured annually. Population growth rates are calculated on the assumption that rate of growth is constant between two points in time. It is normally stated that an increase in population growth leads to a decline in the growth rate. Many economists argue that higher population growth leads to a lower standard of living. It will be used as an indicator in my regression to test prove this impact on developing economies.

In this study, a common panel data regression model is used which will look like:

yit=a+bxit+ϵit

Where y is the dependent variable, x is the independent variable, a and b are coefficients, and i and t are indices for individuals and time. The use of panel data allows empirical tests of a wide range of hypotheses. With observations that span both time and individuals in a cross-section, more information is available, which in turn provides more efficient estimates.

Before inputting the regressions, a correlation matrix was computed. This is used to investigate the dependence between multiple variables at the same time. This allows us to determine which variables may require further investigation and whether they should be replaced or removed from the regression. If the correlation coefficient between two variables is greater than 0.8 then it shows that the two variables are highly correlated and further investigation into whether to remove these variables is suggested. In figure 1, we can see that this is not the case for the independent variables in this study and it is therefore permissible to input the data into the regression.

Figure 1. Correlation matrix of Independent Variables
	Inflation	Exports	EBGS	FDI	GCE	Gross Capital Formation	Population Growth
Inflation	1
Exports	-0.1855	1
EBGS	-0.1945	0.4614	1
FDI	-0.1786	0.2514	0.2672	1
GCE	-0.1514	-0.2038	-0.1747	-0.0048	1
Gross Capital Formation	-0.1543	0.0634	-0.1448	0.4400	0.0136	1
Population Growth	0.2669	-0.0880	-0.3391	-0.3112	-0.1039	-0.1562	1

Empirical Analysis and Results

The results of the effect on economic growth in all 35 countries is shown in Figure 3.

Figure 3. Panel regression results of full sample
Independent Variable	Coefficient	P-value	Standard Error
Constant	1.0917	0.2008	0.8517
Inflation	-0.05968	0.0304	0.0275
Exports	-0.0093	0.0136	0.0038
Foreign Direct Investment(FDI)	6.54E-12	0.1243	4.25E-12
Government Consumption Expenditure (GCE)	-0.1618	0.0000	0.0346
Gross Capital Formation	0.1681	0.0000	0.0218
Population Growth	1.3914	0.0000	0.1601
R-squared	0.336
Adjusted R-squared	0.324
Standard Error	2.830
F-statistic	28.918
Prob (F-statistic)	0.000
Durbin-Watson Stat	1.580
Observations	350

The results from the panel regression show that, over the course of the period, inflation significantly effected the growth rate in all developing economies which was an expected result. This is most likely due to the fact that growth in GDP causes inflation over time. In periods of growth, aggregate demand increases faster than supply which causes upward pressure on prices and wages resulting in a higher inflation rate. As inflation increases, consumers tend to spend more money as they are aware it will have less value in the future. This further increases GDP growth in the short term but brings about further price increases. If inflation reaches a level that is deemed too high, it can have the reverse effect and lead to a decline in the state of the economy. This is explained by the correlation coefficient which shows a weak negative relationship between inflation and GDP growth. If economic growth is caused by an increase in aggregate demand, then this is likely to cause high inflation. If economic growth is caused by an increase in production and aggregate supply, then this is not the case.

Similar to Dollar’s (1992) study,  exports of goods and services also seem to have a significant effect on the GDP growth rate. This suggests that as exports is a large component of aggregate demand, increasing the level of exports in an economy will help rise AD and boost economic growth. The p-value is quite low, which indicates strong evidence that exports has a positive effect on GDP. On the other hand, this is not supported when looking at the correlation coefficient. Generally, the coefficient is expected to show a positive correlation but, in this case, the results are showing a very weak negative correlation. This may be because exports may not contribute to GDP growth in these developing economies as largely as would be expected.

Furthermore, foreign direct investment seems to have an insignificant effect on economic growth which goes against our general hypothesis. This suggests that the level of FDI in these economies may not have been great enough, during this period, to have a positive effect on growth. FDI flows to low-income economies however only few of these economies have been successful in attracting significant FDI flows. Looking at Agrawal’s (2015) study on FDI-led growth, we can assume these successful countries include the five BRICS economies. On the one hand, it is clear there is a positive relationship between FDI and growth, shown by the positive coefficient, however, the insignificant p-value implies that in developing economies, it may be found difficult to secure enough foreign direct investment to boost growth and productivity.

Following the assumptions and hypothesis mentioned earlier, government consumption expenditure, gross capital formation and population growth show a significant impact on economic growth. In fact, the results show that they are highly correlated as the p-value results are 0.0000 for all three variables. This implies that any change in GCE can affect growth on a large scale. The low negative correlation coefficient shows that there is a negative relationship between GCE and output. For example, higher government spending on education and training could increase labour productivity and enable higher long term economic growth. However. lower government expenditure results in reduced productive capacity.

Generally, developing economies often devote a higher percentage of their GDP to investment, which explains the p-value results for gross capital formation. A higher rate of gross capital formation means an increase in assets enabling economic growth. Also, a rise in the population growth increases the productive capacity of the economy and improves tax revenues. Hence, explaining the significance of the results for this variable. Described in the study by Crenshaw et al. (1997), higher population growth can give rise to production, boosting output in these economies.

We can conclude that the findings for the full sample of all 35 developing countries, supports the findings in previous literatures that key determinants of economic growth include, inflation, exports of goods and services, government consumption expenditure, gross capital formation and population growth. Intriguingly, the results were unable to support the theory of foreign direct investment as an important determinant of growth.

The R-squared is the percentage of the variable variation that is explained by the model. It provides an estimate of the strength of the relationship between the model and the response variables. The R-squared here indicates that the model explains 34% of the variability of the response data around its means. In general, the higher the R-squared, the better the model fits the data.

Moving further, this study, grouped the sample into three groups by high, middle, and low-income economies (Figure 4.), to analyse the results in depth and detail.

Figure 4. Allocation of sample by income level.
Group 1		Group 2	Group 3
China	Hong Kong SAR, China	Pakistan	Tunisia
Japan	Israel	Colombia	Congo, Dem. Rep.
India	Singapore	Bangladesh	Uganda
Brazil	Malaysia	Vietnam	Trinidad and Tobago
Korea, Rep.	South Africa	Peru	Afghanistan
Russian Federation		Sudan	Jamaica
Mexico		Sri Lanka	Mauritius
Indonesia		Kenya	Mozambique
Argentina		Tanzania	Rwanda
Nigeria		Ghana	Malawi
15 Countries		10 Countries	10 Countries
Total: 35 Countries

 

Highly Developing Countries

The regressions results for group 1 of the 15 countries with the highest income in the sample is displayed in Figure 5.

Figure 5. Panel Regression results for Group 1.
Independent Variable	Coefficient	P-value	Standard Error
Constant	-2.8405	0.1576	1.9995
Inflation	0.0535	0.4948	0.0782
Exports	-0.0212	0.0281	0.0095
Foreign Direct Investment(FDI)	0.1243	0.0287	0.0562
Government Consumption Expenditure (GCE)	-0.0865	0.2058	0.0681
Gross Capital Formation	0.2511	0.0000	0.0339
Population Growth	1.4408	0.0000	0.3701
R-squared	0.366
Adjusted R-squared	0.340
Standard Error	2.964
F-statistic	13.77
Prob (F-statistic)	0.000
Durbin-Watson Stat	1.821
Observations	150

In this model, the R-squared here indicates that the model explains 36.6% of the variability of the model. Compared to the results in all 35 economies, in group 1 it seems that inflation and GCE have an insignificant impact on economic growth. Inflation may have been too large in this period for these high developing economies. A rise in prices leads to a decrease in consumption and result in a decrease or a halt in growth. The coefficient for inflation is showing a positive correlation. These are very unexpected results. Commonly, as seen before, inflation is expected to have a negative correlation.

Additionally, these results also imply that government expenditure is low in these economies. Reduced government spending has a markedly impact on both aggregate demand and supply, depending on which areas of public spending were cut. A cut in government expenditure is expected to have a negative impact on aggregate demand. A fall in aggregate demand leads to lower economic growth. From looking at Adam and Bevan (2005) findings, we can assume that these economies may be experiencing a fiscal deficit and increases government spending is needed to spur economic growth. The strong negative coefficient supports this evidence and, it is clear, that government expenditure has a significant influence on economic growth.

In contrast to the results identified for the full sample, FDI seems to significantly effect GDP growth, following the findings in Agrawal’s (2015) study. These high developing economies, which includes countries such as China and Japan, are the few economies that have successfully been able to attract a large amount of foreign direct investment flowing into the economy. It is apparent that in doing so, these high developing economies have boosted their economic output during this period. In this case, we can conclude that foreign direct investment is more likely to positively effect high developing economies and is in fact, a key determinant of growth.

Intermediately Developing Economies

Figure 6. Panel Regression results for Group 2.
Independent Variable	Coefficient	P-value	Standard Error
Constant	0.9441	0.5000	1.3944
Inflation	-0.1264	0.0001	0.0317
Exports	-0.0069	0.6496	0.0152
Foreign Direct Investment(FDI)	0.1706	0.1383	0.1141
Government Consumption Expenditure (GCE)	-0.0972	0.0871	0.0562
Gross Capital Formation	0.1810	0.0000	0.0423
Population Growth	1.0483	0.0011	0.3115
R-squared	0.344
Adjusted R-squared	0.302
Standard Error	1.973
F-statistic	8.145
Prob (F-statistic)	0.000
Durbin-Watson Stat	1.419
Observations	100

The regression results for group 2, displays some rather diverse results. Firstly, the R-squared here specifies that the model explains 34.4% of the variability of the model which is lower than the model for group 1. Compared to the higher developing economies, exports is insignificant, indicating a lack of contribution to GDP growth. It can be identified that the productive capacity of these countries is low, therefore resulting in reduced exports. Through this assumption, it is evident that to reach a higher level of economic development, that is apparent in group 1, these economies should focus on export-orientated policies as suggested by Dollar (1992).

In order to improve production and increase exports, a rise in government spending is required on capital, such as equipment and construction. However, government expenditure seems to be low, which is evident by the insignificant p-value. This is similar to that of the higher developing countries. There is also a stronger negative correlation between government spending and GDP growth. Although the p-value is much lower than it is for group 1, it is still clear that for developing countries, there is reduced government spending.

Contrary to higher developing countries, FDI is once again insignificant, supporting the idea that these developing countries were unsuccessful in obtaining enough foreign investment to stimulate output. Population growth is also much less highly correlated as the p-value has changed to 0.0011. Population growth may not be as high as it is in these intermediate countries than it is for the high developing countries.

Furthermore, compared to group 1, inflation significantly affects GDP growth in group 2. In fact, it seems to be positively highly correlated in these economies shown by the low p-value of 0.0001, when compared to the results of the full sample. Similar to finding found in Barros’s (1997) study, this suggests that in these economies, there may be low inflation which is argued to contribute to a higher rate of economic growth in the long term. Low inflation promotes stability which in turn attracts investment, stimulating long term growth.

Low Developing Economies

Figure 7. Panel Regression results for Group 3.
Independent Variable	Coefficient	P-value	Standard Error
Constant	0.7172	0.7347	2.1101
Inflation	-0.1016	0.0572	0.0528
Exports	0.0194	0.4676	0.0266
Foreign Direct Investment(FDI)	-0.0236	0.7075	0.0628
Government Consumption Expenditure (GCE)	-0.0940	0.3720	0.1047
Gross Capital Formation	0.0761	0.1851	0.0570
Population Growth	1.9808	0.0000	0.3598
R-squared	0.396
Adjusted R-squared	0.357
Standard Error	3.045
F-statistic	10.17
Prob (F-statistic)	0.000
Durbin-Watson Stat	1.436
Observations	100

Compared to the other regressions, for group 3, the R-squared is at its highest, indicating that the model explains 30.6% of the variability of the model. Once again, government consumption expenditure, foreign direct investment, and exports is insignificant. However, there is a weak positive correlation coefficient. This was the original expected result for the full sample. This implies that the relationship between exports has a more positive influence in lower developing economies.

Intriguingly, gross capital formation is also insignificant in these countries. As gross capital formation measures the net increase in fixed capital, the results indicate that there is a lack of investment and spending on capital such as land improvements, plant and machinery, and industrial buildings. Some poorer economies such as these are unable to afford investment and therefore concentrate on consumption, which explains the insignificance of the relationship between gross capital formation and GDP growth.

Similar to that of group 1, inflation also seems to be insignificant, but in this case, the correlation shows an expected negative correlation. Even though the results are insignificant, it is still very close to the 0.05 cut off suggesting that any small change in inflation could reverse the effects of inflation on economic growth.

Conclusion

Following previous literatures, this study has found that inflation significantly effects economic growth for all 35 countries. Inflation is a result of short term growth, and can lead to further growth, as long as the rate of inflation doesn’t reach a level that is deemed too high. When the effects were compared between economies with different income levels, these results altered. In both high and low developing groups of countries, inflation was insignificant. In these economies, it seems inflation is too high, contributing much less to the goal of stimulating economic growth. The regression results for intermediate developing economies, on the other hand, shows significant results, which are essentially expected results and suggests that these economies have a lower inflation rate, and as low inflation promotes stability, we can conclude that low inflation is a key determinant of economic growth.

For all 35 countries, conforming to the main hypotheses, the study found that there is a significant relationship between exports of goods and services and economic growth. Growth in exports creates employment and has a large role in determining the current account deficit. Intriguingly, for group 2 and group 3, this is not the case, and results show an insignificant relationship between them. A possible explanation for this is that the level of exports in these economies is not substantial enough to contribute largely to the economies output. These countries would be advised to focus more on export expansion. In group 1, we can see that high developing economies, has succeeded in this aspect, displaying significant results. This allows us to conclude that exports of goods and services is a vital determinant of economic growth.

Puzzlingly, foreign direct investment shows to have an insignificant relationship between GDP growth, for the full sample but when the sample is grouped, the results for high developing economies shows a significant relationship between FDI and growth. For the intermediate and low developing economies, there is also an insignificant relationship, similar to that of the results for the full sample. Group 1, includes countries that are known to have successfully gained substantial amount of FDI inflows, explaining the significant p-value, however, the results for the other groups suggest that these countries are not in the same position as these high developing countries. FDI flows in these countries during this period, may not be great enough to boost output. Therefore, attracting high FDI inflows as a method of increasing growth is important for these developing countries.

Moving further, this study found that the relationship between government consumption expenditure and GDP growth is highly significant, for all 35 countries. Increased government spending results in supply-side improvements in the economy, such as spending on education and training. However, for the intermediate and low developing countries, government spending was insignificant. In order to have a significant effect, these countries would need to raise government expenditure on factors that increase an economy’s output.

Finally, following the main hypotheses, gross capital formation and population growth show highly significant results for all regressions. These are expected results and shows that these variables are important determinants of economic growth.

There have been limitations in this study as it only uses data from a 10-year period and from only 35 countries. Nonetheless, it allows further research on additional economies, so that development strategies can be advised. When critically evaluating the methodology in this study, it would have been preferable to haves a bigger sample, and have tested more factors to further investigate the favourable sources of economic growth.

The primary aim of this study was to test whether these variables influence economic growth and if focus on these determinants can help stimulate growth in developing economies. The results were proven favourable against these variables. Although a few of the results were contradictory to the main hypotheses, we can not ignore the importance of these variables as key determinants of economic growth. I believe this study has succeeded in pinpointing this, and we can conclude that inflation, exports of goods and services, foreign direct investment, government consumption expenditure, gross capital formation and population growth are all key determinants that developing countries should focus their development strategies on, to spur economic growth and development.

Bibliography

Agrawal, G., 2015. Foreign Direct Investment and Economic Growth in BRICS Economies: A Panel Data Analysis. Journal of Economics, Business and Management, Vol. 3(No.4), pp. 421-424.

Barro, R. J., 1997. Determinants of Economic Growth: A Cross-Country Empirical Study. First MIT Press Paperback Edition 1998 ed. s.l.:The MIT Press.

Christopher S. Adam, D. L. B., 2005. Fiscal Deficits and growth in developing countries. Journal of Public Economics , Volume Vol.89, pp. 571-597.

Dollar, D., 1992. Outward-Oriented Developing Economies Really Do Grow More Rapidly: Evidence from 95 LDCs. Economic Development and Cultural Change, Vol.40(No.3), pp. 523-544.

Crenshaw M. E., Ameen Z. A., and Christenson M., 1997. Population Dynamics and Economic Development: Age-Specific Population Growth Rates and Economic Growth in Developing Countries. American Sociological Review , 62(No.6), pp. 974-984.

Gabriele, A., 2006. Exports of Services, Exports of Goods, and Economic Growth in Developing Countries. Journal of Economic Integration, Vol. 21( No. 2 ), pp. 294-317 .

Kappel, R. and Ghani, E., 2003. Openness, Institutions, and Policies: Determinants of Globalisation and Economic Growth in Developing Countries. The Pakistan Development Review, Vol.42(No.4), pp. 395-416.

Gusesh, J. S., 1997. Government Size and Economic Growth in Developing Countries: A Political-Economy Framework. Journal of Macroeconomics , Vol.19(No.1), pp. 175-192.

Hanushek, E. A., 2013. Economic growth in developing countries: The role of human capital. Economics of Education Review, Volume Vol.37, pp. 204-212.

Gocer I., Akin T., and Alatas S.,2016. The effects of saving-investment gap on economic growth in developing countries: A clustering and panel data analysis. Theoretical and Applied Economics, Vol.XXIII,No. 2(607), pp. 157-172.

Karras, G., 2006. Foreign Aid and Long-run Economic Growth: Emprical Evidence for a Panel of Developing Countries. Journal of International Development , Volume Vol.18, pp. 15-28.

Levine, R., 1991. Stock Markets, Growth, and Tax Policy. The Journal of Finance, Vol.46(No.4), pp. 1445-1465.

Knight, M., Loayza N., and Villanueva D., 1993. Testing the Neoclassical Theory of Economic Growth: A Panel Data Approach. Staff Papers (International Monetary Fund), Vol.40(No.3), pp. 512-541 .

Rahnama M., Fawaz F.,  and Gittings K., 2017. The effects of Foreign Aid on Economic Growth in Developing Countries. The Journal of Developing Areas, Vol.51(No.3), pp. 153-172.

Nwaogu G. U., and Ryan J. M., 2015. FDI, Foreign Aid, Remittance and Economic Growth in Developing Countries. Review of Development Economics, Vol.19(No.1), pp. 100-115.

Colombage N. R. S, Maslyuk S. and Taha R., 2015. Stock Market and Tax Revenue as Determinants of Economic Growth: Panel Data Evidence from Developing Asia. The Journal of Developing Areas, Vol.49(No.4), pp. 89-1.