CC BY
0ORIGINAL PAPER
DOI: 10.26794/2308-944X-2023-11-4-29-44 UDC 330:658.155[.67:332.6](045) JEL O16, O57, G15, G21
Foreign Direct Investment and Financial Development: Evidence from Selected Arab League Countries
H. Humta, i. E. §ahin
Selguk University, Konya, Turkiye
ABSTRACT
This study examined the impact of foreign direct investment on financial development in selected Arab League countries (Algeria, Comoros, Egypt, Jordan, Kuwait, Lebanon, Mauritania, Oman, and Oatar) for the period from 2010 to 2021. The study used one explanatory variable - financial development - and one explained variable - foreign direct investment. To obtain reliable and valid results, panel data were analyzed, and various tests were carried out, including the Chow test, Breusch-Pagan Lagrange multiplier test, Hausman test, Jarque-Bera normality test, Wooldridge test, generalized least squares, and ordinary least squares. The robust model of the study revealed a positive and significant relationship between foreign direct investment and the financial development index, human development index, and interest rate. Also, there is a negative and significant relationship between foreign direct investment and the consumer price index and domestic credit to private sector. Based on these findings, the study recommends that Arab League countries' policies and strategies should attract foreign investors to maintain and sustain economic developmental goals for healthy, literate, and wealthy lives.
Keywords: foreign direct investment; financial development index; panel data; Arab League countries; human development index; consumer price index; economic development; econometric models
For citation: Humta H., §ahin I. E. Foreign direct investment and financial development: Evidence from selected Arab League countries. Review of Business and Economics Studies. 2023;11(4):29-44. DOI: 10.26794/2308-944X-2023-11-4-29-44
ОРИГИНАЛЬНАЯ СТАТЬЯ
Прямые иностранные инвестиции и финансовое развитие: результаты исследований отдельных стран Лиги арабских государств
Х. Хумта, И. Э. Шахин
Университет Сельчук, Конья, Турция
АННОТАЦИЯ
В данном исследовании рассматривается влияние прямых иностранных инвестиций на финансовое развитие отдельных стран Лиги арабских государств (Алжир, Египет, Иордания, Катар, Коморские Острова, Кувейт, Ливан, Мавритания и Оман) за период с 2010 по 2021 г. В исследовании использовались одна объясняющая переменная - финансовое развитие, а также одна объясняемая переменная - прямые иностранные инвестиции. Для получения надежных и достоверных результатов был проведен анализ панельных данных и различные тесты, включая тест Чоу, тест Бреуша-Пагана на основе критерия множителя Лагранжа, тест Хаусмана, тест на нормальность Харке-Бера, тест Вулдриджа, а также методы обобщенных наименьших квадратов и обычных наименьших квадратов. Робастная модель исследования позволила выявить положительную и значимую связь между прямыми иностранными инвестициями и индексом финансового развития, индексом человеческого развития и процентной ставкой. Кроме того, существует отрицательная и значимая связь между прямыми иностранными инвестициями и индексом потребитель© НитЛа Н.,§аЫп КЕ., 2023
ских цен, а также внутренним кредитом частному сектору. Исходя из результатов исследования, авторы рекомендуют странам Лиги арабских государств привлекать иностранных инвесторов для поддержания и реализации целей устойчивого экономического развития здорового, грамотного и обеспеченного населения.
Ключевые слова: прямые иностранные инвестиции; индекс финансового развития; панельные данные; страны Лиги арабских государств; индекс человеческого развития; индекс потребительских цен; экономическое развитие; эконометрические модели
Для цитирования: Хумта Х., Шахин И. Э. Прямые иностранные инвестиции и финансовое развитие: результаты исследований отдельных стран Лиги арабских государств. Review of Business and Economics Studies. 2023;11(4):29-44. DOI: 10.26794/2308-944X-2023-11-4-29-44
1. Introduction
Foreign direct investment (FDInvestment) plays a crucial role in fostering an open and efficient international economic system, in contrast to economies that are characterized by restrictive laws and regulations. FDInvestment1 refers to the act of a person or firm making a direct investment in another country, specifically in production or for business purposes. FDInvestment may be undertaken via many methods, such as establishing a subsidiary, purchasing an already established foreign enterprise, or engaging in a merger or joint venture with a foreign corporation [1]. Foreign direct investment evidence indicates that FDInvestment inflows boost economic development through knowledge transfer and spillover efficiency. However, such a beneficial effect does not occur automatically; rather, it depends on the receiving country's absorptive ability. Absorption capacity has been the subject of several studies, and the success of FDInvest-ment focuses only on human capital and trade regimes [2].
According to the World Economic Forum's 2011 Financial Development Report, financial development is the sum of all elements — policy, institutions, and others — that support efficient financial markets and intermediation, as well as the provision of universally available capital and financial services [3].
The factors often used to measure financial development include financial depth, bank ratio, and financial activity, which may be generally characterized [4]. The prosperity of every economy is contingent upon the presence of a competent and efficient financial system since a robust financial system is essential for establishing a solid
1 FDInvestment denotes Foreign direct investment.
economic foundation. An enhanced financial system facilitates the provision of superior financial services, thereby enabling an economy to increase its GDP growth [5].
Foreign direct investment is an important source of growth and development, while financial development is a crucial input in developing nations for economics growth and poverty alleviation. Financial development fosters economic expansion and increases a nation's resilience. It enhances resource allocation, promotes information sharing and financial stability, mobilizes savings, and facilitates risk management and diversification [6].
The present study uses a relatively new measure of financial development proposed by the International Monetary Fund (IMF) [7]. The financial development index2 combines 20 indicators and includes both banking and nonbanking institutions, as well as markets. It assesses financial development across three dimensions: depth, access, and efficiency. In addition, the study offers important policy recommendations for the selected OIC member countries.
There are strong connections between foreign direct investment and financial development. To find out how important FDInvestment is for financial development, correlation and regression models were used to figure out the best way to attract FDInvestment to encourage economic development.
2. Literature review 2.1. Foreign direct investment
Foreign direct investment occurs when a company from one nation establishes a business operation in another nation, either by estab-
2 FDI indicates financial development index.
Fig. 1. Foreign direct investment of Arab League countries, millions of dollars
Source: The Organisation of Islamic Cooperation (OIC) Countries. OIC Statistics Database (OICStat) - Query 2023. URL: https://www.sesric.org/query.php.
lishing a new wholly owned affiliate, acquiring a local company, or organizing a joint venture in the host economy [8]. The potential advantages of foreign direct investment for the host economy are considerable, including several aspects such as the transfer of technology skills, support for capital development, assistance in fostering a competitive business environment, and facilitation of international trade integration. These advantages can stimulate the main economic sectors, including petroleum, mining, manufacturing, agriculture, transportation, communication, construction, and others. Numerous nations consciously seek to attract foreign direct investment due to their belief that multinational firms would significantly contribute to economic development via the creation of new employment prospects, added capital accumulation, and enhanced total factor productivity. Without a doubt, an extensive database of empirical data proves the notion that foreign direct investment often provides overall benefits for both the countries of ori-
gin and recipient nations. These sectors are crucial for attaining substantial levels of employment and fostering economic growth and development. However, the advantages of foreign direct investment for nations may vary based on economic conditions and resource availability [1]. Economic factors, infrastructure, technology, institutional-political factors, specific risks, legal integration, space factors, entrepreneurial matters, cultural factors, and para-cultural factors are believed to influence the attraction of foreign direct investment. Most studies prioritize examining the economic variables that influence FDIn-vestment flows while largely disregarding or briefly mentioning other categories of factors. Scientists have studied a variety of economic factors in relation to FDInvestment, with varying degrees of success [9]. These factors include income, exchange rate, economic freedom, economic stability, liquidity, market size, market size growth, inflation, trade, capital availability, wages, agglomeration, capital formation, financial market, and debt.
The ups and downs of foreign direct investment for some selected Arab League countries (Fig. 1) during the last decade revealed that factors having a relation to or effect on foreign direct investment could make it possible to attract investment, and as a result, economic development can be provoked.
2.2. Financial development
The advancement of the financial industry is a crucial element that contributes significantly to gaining a competitive edge. A resilient financial system encourages multinational corporations to make investments in the countries where they are based [10]. Financial development refers to the enhancement of services offered by the financial system, including the allocation of capital to profitable investments, savings, risk diversification and monitoring, and risk management of these investments, as well as facilitating the exchange of goods and services. Financial development has the potential to decrease the imbalance of knowledge between parties, promote the distribution of risk, and mitigate limitations on financial activities. Financial development enhances the capacity of the financial system to withstand economic shocks, mitigates the exacerbation of negative effects caused by such shocks, and reduces macroeconomic instability and social disparities.
The financial development index contains two sub-indices (the Financial Institution Index and the Financial Market Index). Each sub-index has its indicator in three (Depth, Access, and Efficiency) categories. Every category shows at least one indicator. The sum of the total indicator value is named the value of the financial development index. Financial development combines financial markets with financial institutions to extend the availability and access of debt or funds to clients or customers to maintain their organizational goals or profit. Financial markets are the channel through which FDInvestment may be beneficial for financial development and, most commonly, for growth. The theoretical model shows that improvements in financial markets increase output by increasing the marginal product of FDInvestment [7]. Table 1 explains the indicators of the financial development index.
2.3. Correlation between FDInvestment and financial development index
Orji et al. (2021) utilized autoregressive distributed lag (ARDL) and ordinary least squares to analyze the impact of FDI on Nigeria's GDP growth. The findings indicate that both foreign direct investment and human capital training contribute positively and significantly to economic development in Nigeria. The exchange rate and inflation have a negative impact on Nigeria's economic growth, while trade openness has an opposite effect [11]. Gokmenoglu et al. (2018) state that the impact of FDInvestment on the human development index (HDI) is a complex problem; thus, policymakers should be aware of and consider the pros and cons of FDInvestment inflows on all aspects of human development to achieve the best outcomes. The author's findings indicate that FDInvestment has a favorable impact on economic growth and educational advancement; therefore, they suggest that Nigerian policymakers should make attracting international investors a top priority. However, the dynamic ordinary least squares (DOLS) estimator shows that FDInvestment harms life expectancy in Nigeria due to FDInvestment-in-duced competition and insecurity [12].
In 2018, Saidi conducted an empirical study to examine the relationship between foreign direct investment, financial development, and economic growth in low-income countries. The study used data from 1990 to 2015. The findings of this study indicate that foreign direct investment operations can provide significant benefits for low-income countries in terms of technology acquisition, increased investment inflows, job generation, human capital development, and enhanced corporate growth. Moreover, the study revealed the presence of long-term cointegration and bidirectional causation between foreign direct investment and financial development in low-income nations [13].
According to Bayar and Gavriletea (2018), FDInvestment inflows do not significantly affect the level of financial development over the long and short terms. Nevertheless, it is evident that a one-way causal relationship exists between the growth of the financial sector and the influx of FDInvestment in the countries of the Central and Eastern European Union. Hence, based on theoretical analysis, no obvious influence, whether good or negative, can be identified [14].
Table 1
New broad-based index of financial development
Financial institution index Financial markets index
Categories-based indicators of financial Categories-based indicators of financial markets
institution index index
Private-sector credit to GDP (Depth) Pension fund assets to GDP (Depth) Mutual fund assets to GDP (Depth) Insurance premiums, life, and non-life to GDP (Depth)
Bank branches per 100000 adults (Access) ATMs per 100000 adults (Access) Net interest margin (Efficiency) Lending-deposits spread (Efficiency) Non-interest income to total income (Efficiency)
Overhead costs to total assets (Efficiency) Return on assets (Efficiency) Return on equity (Efficiency)
Stock market capitalization to GDP (Depth)
Stocks traded to GDP (Depth)
International debt securities of government to GDP
(Depth)
Total debt securities of financial corporations to GDP (Depth)
Total debt securities of nonfinancial corporations to GDP (Depth)
Percent of market capitalization outside of the top 10 Largest Companies (Access) Total number of issuers of debt (domestic and external, nonfinancial, and financial corporations), (Access)
Stock market turnover ratio (stocks traded to capitalization), (Efficiency)
Source: [7].
Majeed et al. (2021) analyzed 102 Belt and Road Initiative nations from Asia, Europe, Africa, and Latin America to determine the impact of foreign direct investment on financial development. They employed a suite of quantitative methods, including feasible generalized least squares and augmented mean group techniques, using data collected from 1990 to 2017. The results of the research indicate a statistically significant association between FDInvestment and trade openness, inflation, government consumption, and financial development. Asia, Europe, and Latin America experienced an increase in FDInvestment due to trade liberalization and government consumption, whereas Africa reported a reduction. Financial development is negatively impacted by inflation on every continent. Moreover, in Asia and Europe, the Dumitrescu-Harlin panel causality test validates a two-way causal link between FDInvestment, trade openness, and financial development. In contrast, FDInvestment and financial development in Latin America are interdependent in a unidirectional fashion. Due to high factor costs, low-income and middle-income countries attract more foreign direct investment than high-income countries, as indicated by income-based results [15].
Lestari et al. (2022) evaluated the impact of FDInvestment and corruption on financial development in developing nations. Furthermore, they examined the collective influence of financial development and corruption on FDInvestment. The study's findings indicated that financial development has a favorable and substantial influence on FDInvestment, but corruption does not exert a statistically noteworthy effect. This illustrates the significant role of financial development in fostering the expansion of foreign investment and serving as a crucial source of financing for emerging nations. Nevertheless, the correlation between financial development and corruption exerts an adverse impact on FDInvestment. Consequently, when FDInvestment experiences a decline due to an increase in corruption, these findings prompt policymakers to tackle concerns related to the combined influence of financial development and corruption on the inflow of foreign direct investment in developing nations [16].
Nguyen et al. (2023) examined the impact of the financial development index on the inflow of foreign direct investment in Vietnam between 1996 and 2021. The model used time series data to evaluate the impact of six factors representing financial development. Data on financial institu-
tion assessment, financial depth of institution, financial efficiency of the institution, financial assessment of the market, financial efficiency of the market, and domestic loans for the private sector were gathered from 1996 to 2021. The study reveals that a rise in financial institution assessment, financial efficiency of the institution, and financial efficiency of the market leads to a corresponding increase in Vietnam's FDInvestment inflow, whereas an increase in financial depth of the institution and financial assessment of the market, results in a fall in FDInvestment. The impact of domestic loans for the private sector is unreliable, and this study lacks any substantial correlation. This research all so reveals that the chosen financial development index has a substantial influence on attracting foreign direct investment to a country. Furthermore, the government and authorities must formulate suitable policies in the future [17].
3. Methodology
This section presents an overview of the study methods, econometric model, and tests employed to assess the research model results. Balanced panel data of selected countries (Algeria, Comoros, Egypt, Jordan, Kuwait, Lebanon, Mauritania, Oman, and Qatar) were collected from the website of the Statistical, Economic and Social Research and Training Centre for Islamic Countries (SESRIC) of the Organization of Islamic Countries (OICStat) database. A set of panel data was used in this investigation through Stata 17 and EViews 13. Panel data are a combination of cross-sectional and time-series data. For better measurement of financial development, the most general proxy, the Financial Development Index, is considered in the study, which consists of two different indexes named the Financial Institutions Index and the Financial Markets Index. Dependent, independent, and controlled variables are explained; descriptive statistics, correlation matrix, common, fixed, random, and robust models of effect were employed on the mentioned variables. Several tests (the Chow test, the Lagrange multipliers test, and the Hausman test) were performed to find the best model of effect. Once the model was chosen, other tests (the Breusch-Pagan Lagrange multiplier test for independence, the heteroskedasticity white test, the Wooldridge
test for autocorrelation, and the Jarque-Bera normality test) were carried out to find the appropriate model.
4. Results and Findings
4.1. Definition of the variables
Definitions of the variables are presented in Table 2.
In order to investigate the association and regression between foreign direct investment and financial development, we first consider the descriptive statistics and correlation analysis of the variables. Subsequently, we use three different types of panel regression models to further analyze the previously discussed relationship, as shown in Tables 3 and 4.
According to descriptive statistics, we have the means, standard deviation, min, and max for all three categories of variables named as dependent, independent, and control variables.
Correlation matrix Table 4 shows the type and degree of relationship between all three categories of variables (dependent, independent, and controlled variables).
4.2. Empirical models
This subsection covers the empirical models and techniques that provide the foundation for the subsequent estimation approach. These models and methods were used to estimate the impact of foreign direct investment on financial development of selected Arab League countries. The empirical model used to analyze the impacts of foreign direct investment of this type may be expressed in the following econometric equations:
FDInvestment = = f (FDI, HDI, CPI, IR, TO, PCGDP, DCtPS- %GDP), (1)
FDInvestment = P0 + P1 FDI u + P2HDIit + P3CPI it + + fi4IRit + P5TOu + P6PCGDPU + P7DCtPS - %GDPU + Uit, (2)
where, Foreign Direct Investment is a dependent variable, FDI is the financial development index, HDI is the human development index, CPI is the consumer price index, IR is the interest rate, TO is the trade openness, PCGDP is the per capita gross domestic product, and DCtPS%GDP stands for domestic credit to the private sector (% GDP). And P0 is the intercept, (P^..., P7) are the coefficients, and U it is the error term of the model.
Table 2 Definition of the variables
Full Name Variables Description Source
Dependent Variable Foreign Direct Investment The inflows of FDInvestment refer to transactions that raise the investment held by foreign investors in domestic firms, excluding transactions that reduce the investment of foreign investors in domestic companies, expressed in the current USD UNCTADST Database
Independent Variable Financial Development Index (Index Value) Evaluate the comparative placement of nations based on the extent, availability, and effectiveness of their financial institutions and financial markets. Combining the Financial Institutions Index and the Financial Markets Index yields the comprehensive financial development index IMF
Control Variables Human Development Index (Rank Value) The standard competition ranking, often known as the «1224» ranking, assigns the same ranking number to countries with equal Human Development Index scores, leaving a gap between the ranking numbers OlCStat
Trade Openness (Percent) Trade balance, calculated as the ratio of the sum of exports and imports of goods and services to the gross domestic product (GDP), is represented as a percentage WDI Database
Domestic Credit to the Private Sector (Percent),% GDP The private sector's share of financial resources provided by financial businesses, which includes loans, non-equity securities purchases, trade credits, and other accounts receivable, in relation to the Gross Domestic Product (GDP) IFS World
Interest Rate (percentage) Annual bank rate that usually meets the short- and medium-term financing needs of the private sector, expressed as a percentage OlCStat
Consumer Price Index (percentage) The final value of the consumer price index (CPI) at the end of a certain period. The Consumer Price Index (CPI) measures the fluctuations in the expenses associated with purchasing a standardized assortment of goods and services by the typical consumer OlCStat
Per Capita Gross Domestic Product Per capita real GDP, denominated in USD OlCStat
Source: Developed by the authors based on OlCStat Data Base.
Table 3 Descriptive statistics of variables
Variable Obs Mean Std.Dev. Min Max
FDInvestment 108 1670.449 2181.461 -2812.640 9010
FDI 108 0.284 0.152 0.040 0.600
HDI 108 0.721 0.107 0.510 0.860
CPI 108 132.635 57.616 16.600 452.500
IR 108 8.402 4.342 0.790 18.320
TO 108 79.708 28.306 29.860 163.980
PCGDP 108 15 057.205 20 571.258 1284.400 80 743.129
DCtPS-%GDP 108 51.312 35.710 -12.730 177.080
Source: Developed by the authors.
Table 4
Correlation matrix analysis
Variables (1) (2) (3) (4) (5) (6) (7) (8)
(1) FDInvestment 1.000
(2) FDI 0.086 1.000
(3) HDI 0.124 0.850 1.000
(4) CPI -0.407 -0.341 -0.244
(5) IR 0.117 -0.590 -0.746
(6) TO -0.236 0.588 0.453
(7) PCGDP -0.232 0.720 0.645
(8) DCtPS-%GDP -0.177 0.669 0.605
Source: Developed by the authors.
The general form of the panel data regression model is given by equation (3). The panel data regression model, which is briefly described in equations (4), (5), and (6), will be used in this study to estimate the impact of FDInvestment on financial development.
k
Y = at+YpjH X,t + U it, (3)
where: Yt is response variable of the (i) individual and t time period, at is constant value/intercept of the (i) individual and t time period,
P jt ( P lit, P2 it v-P kit ) are coefficients of (K) independent variables, X t is the independent vari-
j t
ables value of (i) individual and t time period, Utt is the error of (i) individual and t time period, and (i and t) are the number of individual (i; 1, 2, 3..., N) and time period (t;1,2,3..., T), respectively.
4.3. Estimation of the panel data regression model
For parameter estimation, the model depends on the intercept and slope coefficient assumptions. Using panel data permits unique intercept and slope coefficients for each individual and time period. The following three types of models are applicable to this concept: CEM (common effect model or pooled regression); FEM (fixed effects model); and REM (random effects models).
Common effect model. Pooled regression (CEM model) implies that the intercepts and slope coefficients for all individuals and time periods have the same value. This model does not consider
1.000
0.248 1.000
-0.010 -0.278 1.000
-0.166 -0.496 0.430 1.000
0.030 -0.432 0.393 0.405
both individual dimensions and time [18]. Equation (4) illustrates the CEM model:
k
Y =a + Epj Xjt + U It . (4)
j=1
A common effect model was employed using Stata 17 software. As a result, the financial development index, human development index and interest rate have a positive statistically significant relationship with foreign direct investment, whereas the consumer price index, trade openness, per capita gross domestic product and domestic credit to the private sector calculated as a percentage of GDP have a negative significant relationship with FDInvestment (Table 5).
Fixed effect model. This approach assumes that the intercepts represent the differences in individual characteristics. Thus, intercepts for each individual will vary, whereas slope coefficients will remain constant across all time intervals [18].
Y = a( + 5>j XJH + Utt . (5)
j=1
In this method, we used a dummy variable to estimate the intercept for each individual; thus, this approach is called least square dummy variable (LSDV) model.
n k
Yt = a + Jal D,, + xJtt +Uit, (6)
i=1 j=1
Table 5 Common effect
FD
Investment
Coef.
St. Err. t-value p-value [95% Conf Interval]
Sig
FDI
8828.658 2602.477
3.390 0.001 3665.417 13 991.898
HDI
14 652.036 3057.034 4.790 0.000
8586.968
20 717.105
CPI
-8.956
2.964
-3.020 0.003
-14.837
-3.075
IR
247.306
49.264
5.020 0.000
149.568
345.043
TO
28.142
6.345
4.440 0.000
40.731
15.554
PCGDP
-0.065
0.010
-6.380 0.000
-0.085
-0.044
DCtPS-%GDP
-25.245
5.810
-4.350 0.000
-36.771
-13.718
Constant
7785.323 2144.541
-3.630 0.000 -12 040.032
3530.613
***
***
***
***
Mean dependent var
1670.449
SD dependent var
2181.461
R-squared
F-test
Akaike crit. (AIC)
0.597 21.158
1883.903
Number of obs. Prob > F
Bayesian crit. (BIC)
108 0.000
1905.360
Source: Developed by the authors. Notes: p<0.01, "" p<0.05, " p<0.10.
where a0 is the mean of the intercept from the error terms cross-section and time series; uu =ц( +ги, where is the random error of the cross-sectional deviation, which explains the differences between individuals; and ей is the error term of the model.
As a result of employing the fixed effect model, it was determined that the human development index, per capita gross domestic product and interest rate have a positive statistically significant relationship with foreign direct investment, whereas the consumer price index has a negative significant relationship with FDInvestment (Table 6).
Random effect model. Error terms in REM take into account the variations in individual characteristics. Error terms may correlate between individuals and between time periods [18]. We can formulate it as follows:
Y =a + Tßj Xßt +W+
j=i
(7)
As a result of employing the random effect model, it was determined that the financial development index, human development index, and interest rate have a positive statistically significant relationship with foreign direct investment, whereas the consumer price index, trade openness, and domestic credit to private sector calculated as percentage of GDP have a negative significant relationship with FDInvestment (Table 7).
4.4. Selection of the panel data regression model
Multiple tests must be carried out to settle on the model that will be used to deal with panel data.
Table 6 Fixed effect
FD Investment Coef. St. Err. t-value P-value [95% Conf Interval] Sig
FDI 8528.507 5191.465 1.640 0.104 -1782.189 18 839.204
HDI 23 677.845 10402.984 2.280 0.025 3016.620 44339.070 **
CPI -8.979 3.532 -2.540 0.013 -15.993 -1.965 **
IR 214.528 65.93 3.250 0.002 83.585 345.471 ***
TO -6.366 12.3 -0.520 0.606 -30.794 18.062
PCGDP 0.124 0.067 1.860 0.066 -0.008 0.256 *
DCtPS-%GDP -16.002 12.053 -1.330 0.188 -39.94 7.936
Constant -18 983.483 7835.655 -2.420 0.017 -34 545.770 -3421.197 **
Mean dependent var 1670.449 SD dependent var 2181.461
R-squared 0.351 Number of obs 108
F-test 7.096 Prob > F 0.000
Akaike crit. (AIC) 1831.444 Bayesian crit. (BIC) 1852.901
Source: Developed by the authors. Notes: *** p<0.01, "" p<0.05, " p<0.10.
1. Chow Test. The Chow test is used to determine the most appropriate model for the study between the common effect and fixed effect models. Using this test, we test the null hypothesis (CEM is more appropriate than FEM). The hypothesis in the Chow test can be written as follows:
H0: Common Effect Model is more appropriate than Fixed Effect Model.
H1: Fixed Effect Model is more appropriate than Common Effect Model.
If the P-value is less than 0.05, then we can reject the H0 hypothesis. This means that the FEM is a better model than the CEM.
2. Lagrange multiplier (Bruch-Pegan) test. To decide whether the model, CEM or REM, is superior, researchers employ the Lagrange multiplier (Bruch-Pegan) test. The LM test hypothesis may be expressed as follows:
H0: The Common Effect Model is more appropriate than the Random Effect Model.
H1: The Random Effect Model is more appropriate than the Common Effect Model.
If the P-value is less than 0.05, then we can reject the H0 hypothesis. That means that the REM is a better model than the CEM.
3. Hausman Test. Finally, we performed the Hausman test to determine which of the two models (FEM or REM) would provide the most accurate results. In the Hausman test, one possible hypothesis states as follows:
H0: Random Effect Model is more appropriate than Fixed Effect Model.
H1: Fixed Effect Model is more appropriate than Random Effect Model.
If the P-value<0.05, then we can reject the null hypothesis, and conclude that FEM is a better model than REM. To select the best model from CEM, FEM and REM we will follow the diagram Fig. 2. As we can see, our base model is REM to finalize the most appropriate model.
Estimation of the parameter model in the CEM, FEM, and REM methodologies involves the use of ordinary least squares (OLS) and general least squares (GLS) methods. The classical assumption
Table 7 Random effect
FD Investment Coef. St. Err. t-value p-value [95% Conf Interval] Sig
FDI 12 389.857 4410.504 2.810 0.005 3745.428 21 034.285 ***
HDI 11 211.597 6395.127 1.750 0.080 -1322.622 23 745.816 *
CPI -9.238 3.394 -2.720 0.006 -15.889 -2.586 ***
IR 280.303 57.825 4.850 0.000 166.969 393.637 ***
TO -22.100 9.871 -2.240 0.025 -41.447 -2.754 **
PCGDP -0.037 0.031 -1.180 0.239 -0.097 0.024
DCtPS-%GDP -35.954 8.807 -4.080 0.000 -53.216 -18.693 ***
Constant -6907.591 4307.491 -1.600 0.109 - -15 350.118 1534.937
Mean dependent var 1670.449 SD dependent var 2181.461
Overall r-squared 0.493 Number of obs. 108
Chi-square 46.981 Prob > chi2 0.000
R-squared within 0.302 R-squared between 0.616
Source: Developed by the authors.
Notes: *** p<0.01, "" p<0.05, " p<0.10.
tests are conducted to obtain the optimal OLS or GLS estimator, often referred to as the best linear unbiased estimator (BLUE). Classical assumption tests include three specific tests: the Breusch-Pagan LM test of independence, the Modified Wald test for groupwise heteroskedasticity, and the Wooldridge test for autocorrelation. The Jarque-Bera normality test is a statistical test used to assess the normality of a given dataset. According to Gujarati (2022), in order to acquire the best estimator, it is necessary to meet the requirements of tests 2 and 3 among the four tests [20]. The results of the Breusch-Pagan LM test for independence, Heteroskedasticity White test, the Wooldridge test for autocorrelation, and the Jarque-Bera normality test are shown in the table below. These tests were conducted with random effect.
The LM test does not need to be performed if FEM is found to be the best in the Chow Test and Hausman Test. This test is only performed when the best model found in the Chow Test is FEM, whereas in the Hausman Test, the best model is REM. After selecting an appropriate model, we
apply the robustness tests to determine the validity of the model.
When selecting a model for managing panel data, it is necessary to conduct various tests. These tests include the Chow test, which is employed to determine the superior model between CEM and FEM [19]. In addition, the Breusch-Pagan Lagrange multiplier test was used to select a better model between CEM and REM, while the Hausman test was employed to choose the preferred model between FEM and REM.
The results of the Chow test, Breusch-Pagan Lagrange multiplier test and Hausman tests are illustrated in Table 8.
We used the Chow test, the Breusch, and Pagan Lagrange multiplier test, and the Hausman test to determine which model most accurately reflected the data. The Chow test was used to check whether the intercept in each regression model showed any differences across the various countries. According to the Chow test, the regression reliability was 95% accurate, with a p-value of 0.000 < 0.05. As a result, the random effect model is the correct choice and accepts the null hypothesis H0.
Hausman
Fig. 2. Diagram of selecting appropriate models of effects
Source: Developed by the authors. Table 8
Selecting the appropriate model for parameter estimation
Test name
Null hypothesis
P-value
Result
Chow Test
Breusch-Pagan Lagrange Multiplier Test
Hausman Test
H0: CEM is more appropriate than FEM
H0: CEM is more appropriate than REM
H0: REM is more appropriate than FEM
0.000
0.000
0.1780
Rejected Rejected
Accepted
Source: Developed by the authors.
Note: common effect model (CEM); fixed effect model (FEM); random effect model (REM).
Hence, the appropriate model for this scenario is REM. The use of the Chow and Hausman tests suggests that REM is suitable for this analysis. After REM is selected as the panel data regression model, classical assumption tests are performed to obtain the best OLS estimator, known as the best linear unbiased estimator (BLUE).
Estimation of the parameter model in the CEM, FEM, and REM methodologies uses OLS and GLS methods. The classical assumption tests are conducted to get the optimal OLS or GLS estimator, often referred to as the BLUE. The classical assumption tests include three specific tests: the Breusch-Pagan LM test of independence, the Modified Wald test for groupwise heteroskedas-
ticity, and the Wooldridge test for autocorrelation. The Jarque-Bera normality test is a statistical test used to assess the normality of a given dataset. According to Gujarati (2022), to acquire the best estimator, it is necessary to meet the requirements of tests 2 and 3 among the four tests [20]. The results of the Breusch-Pagan LM test for independence, the Heteroskedasticity White test, the Wooldridge test for autocorrelation, and the Jarque-Bera normality test are shown in Table 9. These tests were conducted with a random effect.
According to classical assumptions test, when residuals are not normally distributed and have dependency between cross sections, the model parameters (coefficient) will not be BLUE. In this
Table 9
Testing the validity of the selected model
Test Name
Null Hypothesis
Prob
Results
Breusch-Pagan LM test of independence Heteroskedasticity White test Wooldridge test for autocorrelation
H0: There is no dependency between
cross sections/individuals 0.000
H0: The residuals are homoscedastic 0.002 H0: There is no autocorrelation 0.229
between residuals
Rejected Accepted Accepted
Jarque-Bera normality test H0: Residuals are normally distributed 0.000
Rejected
Source: Developed by the authors. Table 10
Randon effect model robust SE regression
FD
Investment
Coef.
St. Err. t-value p-value [95% Conf [Interval] Sig
FDI
12 389.857 7034.772 1.760 .078 -1398.043 26 177.756
HDI
11 211.597 6028.547 1.860 .063 -604.138 23 027.332
CPI
-9.238
2.915 -3.170 .002 -14.951
-3.524
IR
280.303
110.393 2.540 .011
63.938
496.668
TO
22.100
13.864
1.590 .111
49.274
5.073
PCGDP
0.037
0.037
0.990 .324
0.109
0.036
DCtPS-%GDP
Constant
35.954
9.192
3.910 0.000
53.970
17.939
-6907.591 4130.732 -1.670 0.094 -15 003.677 1188.496
Mean dependent var 1670.449 Overall r-squared 0.493
Chi-square 512.249 R-squared within 0.302
SD dependent var Number of obs Prob > chi2 R-squared between
2181.461 108 0.000 0.616
Source: Developed by the authors. Notes: *** p<0.01, "" p<0.05, " p<0.10.
**
case, we can use robust standard error to correct the standard error of the model. The final model for this research. To test the validity of the model results, we use the following robustness (classical assumptions) tests. After correcting the standard error of the model, the effect regression model will be Randon Effect Model Robust SE Regression as shown in Table 10.
The results could be concluded as follows:
1. Foreign direct investment has a positive and significant effect (p-value < 0.10) on the financial development index in selected Arab League countries; this means a 1% increase in financial development will increase foreign direct investment by about 12,389 million USD.
2. Foreign direct investment has a positive and significant effect (p-value < 0.10) on the human development index in some selected Arab League countries; this means an increase in each percent of human development will increase foreign direct investment by about 11,211 million USD.
3. Foreign direct investment has a positive and significant relation to the interest rate, whereas the consumer price index and domestic credit to the private sector have a negative and significant impact on financial development. Overall, the model is significant at 95 percent, with a confidence level F (8, 92) = 7.19 and p-value = 0.000 < 0.05, which clarifies that the null hypothesis (H0: the predictor variable simultaneously does not affect the response variable) is rejected. As a result, the predictor variable simultaneously affects the response variable. The value of R2 = 0.615 shows that the independent variable (financial development index) and controlled variables (human development index, consumer price index, interest rate, and domestic credit to private sector as percentage of GDP) can affect the outcome variable (foreign direct investment) in the Arab League countries by 61.5 percent while assuming the other effecting factors are constant.
5. Discussion and conclusions
The paper explores the relationship between foreign direct investment and the financial development index. This relationship has been investigated using an econometric model with one independent, one dependent and six controlled variables. Panel data were analyzed using Stata 17 and EViews 13. Results indicate that there has been a positive and statistically significant rela-
tionship between foreign direct investment and the financial development index, human development index, and interest rates, whereas a negative relationship exists between foreign direct investment and the consumer price index and domestic credit to the private sector as percentage of GDP. In recent years, some studies have shown that foreign direct investment has an impact on financial development and vice versa, but only a few studies have identified that there is no effect between foreign direct investment and financial development. While our research has established a positive influence of FDIn-vestment on financial and human developments, this result agrees with the studies [14-17].
Furthermore, the results of the study indicate that the contribution of domestic investment is essential and more significant than the foreign investment of every country, especially in Organisation of Islamic Cooperation (OIC) member countries, such as Malaysia, where a 1% increase in foreign direct investment contributes 0.02% to GDP, while local investment contributes up to 0.025%. For greater efficiency and contribution to GDP regarding foreign investment, certain economic, political, and cultural structures must be changed according to the national strategies of countries. The selected OIC member countries should pay attention to monetary policy based on interest rates, domestic credit to the private sector and inflation, where the interest rate positively affects foreign direct investment. In contrast, inflation and domestic credit to the private sector negatively affect foreign direct investment. The limitation of this study was the focus on foreign direct investment instead of considering other types of investment, such as portfolio investment or financial assets. A further constraint of this analysis was the use of aggregate foreign direct investment instead of sector-specific foreign direct investment. The authors suggest that the future investigation maintain the controlled variables of the study as main variables in the context of OIC member countries and could compare one or a group of the member countries to the rest of the countries to find a solution to the problem. Several factors, including initial income, human capital, population growth, government consumption, black market premium, institutional quality, rate of inflation, and trade volume, may influence the
effectiveness of FDInvestment. In future research, these variables could be considered as moderated variables between foreign direct investment and financial development in the context of different continents. The study also recommended that the
government and policymakers implement financial system quality standards, and that country authorities should also aim to boost the banking sector and financial markets by enacting market-friendly policies.
REFERENCES
1. Johnny N., Timipere E. T., Krokeme O., Markjackson D. Impact of foreign direct investment on unemployment rate in Nigeria (1980-2015). International Journal of Academic Research in Business and Social Sciences. 2018; Feb8(3):57-69. URL: http://dx.doi.org/10.6007/IJARBSS/v8-i2/3905
2. Ang J. B. Foreign direct investment and its impact on the Thai economy: the role of financial development. Journal of Economics and Finance. 2009; Jul33:316-323. URL: https://doi.org/10.1007/s12197-008-9042-6
3. Streimikiene D. World economic forum 2012. Intelektine ekonomika. 2012;6(1):806-810. URL: https://www. ceeol.com/search/article-detail?id=129960
4. Irsova Z., Havranek T. Determinants of horizontal spillovers from FDI: Evidence from a large meta-analysis. World Development. 2013;Feb1;42:1-5. URL: https://doi.org/10.1016/j.worlddev.2012.07.001
5. Ang J. B. Economic development, pollutant emissions and energy consumption in Malaysia. Journal of Policy Modeling. 2008;Mar1;30(2):271-278. URL: https://doi.org/10.10Wj.jpolmod.2007.04.010
6. Irandoust M. FDI and financial development: Evidence from eight post-communist countries. Studies in Economics and Econometrics. 2021;Apr3;45(2):102-116. URL: https://doi.org/10.1080/03796205.2021.1978859
7. Svirydzenka K. Introducing a new broad-based index of financial development. International Monetary Fund. 2016;Jan12. URL: https://www.elibrary.imf.org/view/journals/001/2016/005/article-A001-en.xml
8. Lenzen M., Moran D., Kanemoto K., Foran B., Lobefaro L., Geschke A. International trade drives biodiversity threats in developing nations. Nature. 2012;Jun7;486(7401):109-112. URL: https://doi.org/10.1038/na-ture11145
9. Tocar S. Determinants of foreign direct investment: A review. Review of Economic and Business Studies. 2018;11(1):165-196. DOI: 10.1515/rebs-2018-0069
10. Sahin S., Ege I. Financial development and FDI in Greece and neighbouring countries: A panel data analysis. Procedia Economics and Finance. 2015;Jan1;24:583-588. URL: https://doi.org/10.1016/S 2212-5671(15)00640-1
11. Orji A., Nwagu G. U., Ogbuabor J. E., Anthony-Orji O.I. foreign direct investment and growth nexus: Further evidence from Africa's largest economy. Journal of Infrastructure Development. 2021; Jun13(1):65-78. URL: https://doi.org/10.1177/09749306211024433
12. Gokmenoglu K. K., Apinran M. O., Ta^pinar N. Impact of foreign direct investment on human development index in Nigeria. Business and Economics Research Journal. 2018;9(1):1-4. DOI: 10.20409/berj.2018.90
13. Saidi K. Foreign direct investment, financial development and their impact on the GDP growth in low-income countries. International Economic Journal. 2018; Jul3;32(3):483-497. URL: https://doi.org/10.1080/10168737. 2018.1529813
14. Bayar Y., Gavriletea M. D. Foreign direct investment inflows and financial development in Central and Eastern European Union countries: A panel reintegration and causality. International Journal of Financial Studies. 2018; May29;6(2):55. URL: https://doi.org/10.3390/ijfs6020055
15. Majeed A., Jiang P., Mahmood A., Khan M. A., Olah J. The impact of foreign direct investment on financial development: new evidence from panel reintegration and causality analysis. Journal of Competitiveness. 2021; Mar1;13(1):95. URL: https://doi.org/10.7441/joc.2021.01.06
16. Lestari D., Lesmana D., Yudaruddin Y. A., Yudaruddin R. The impact of financial development and corruption on foreign direct investment in developing countries. Investment Management and Financial Innovations. 2022;19(2):211-220. URL: http://dx.doi.org/10.21511/imfi.19(2).2022.18
17. Nguyen C. T., Nguyen C. D., Do Bui V. Index of financial development with foreign direct investment: case of Vietnam. Contemporary Economics. 2023;Sep1;17(3). DOI: 10.5709/ce.1897-9254.513
18. Jauhari A. F., Suprapto Y. K., Mauludiyanto A. Data panel modelling with fixed effect model (FEM) approach to analyze the influencing factors of DHF in Pasuruan Regency. ICASESS 2019. 2020:224. URL: http://repository.
umy.ac.id/bitstream/handle/123456789/35320/ICASESS_2019.pdf?sequence=1&isAllowed=y%23page=186# page=241
19. Greene W. Reconsidering heterogeneity in panel data estimators of the stochastic frontier model. Journal of Econometrics. 2005; Jun1;126(2):269-303. URL: https://doi.org/10.1016/j.jeconom.2004.05.003
20. Gujarati D. N. Basic econometrics. Prentice Hall; 2022;Jan1. URL: http://portal.belesparadisecollege.edu. et:8080/library/bitstream/123456789/2098/1/Basic%20Economics-The%20McGraw-Hill%20Series.pdf
ABOUT THE AUTHORS / ИНФОРМАЦИЯ ОБ АВТОРАХ
Humayun Humta — Ph. D. Candidate in Accounting and Finance, Faculty of Economics and Administrative Sciences, Sel^uk University, Konya, Turkiye
Хумаюн Хумта — аспирант банковского дела и финансов, Институт социальных наук, экономики и административных наук, Университет Сельчук, Конья, Турция https://orcid.org/0000-0002-2921-6265 Corresponding Author / Автор для корреспонденции: h.humta@gmail.com
ibrahim E. gahin — Associate Professor of Accounting and Finance and Coordinator of Faculty Member Training Program, Faculty of Economics and Administrative Sciences, Sel^uk University, Konya, Turkiye
Ибрагим Э. Шахин — доцент кафедры финансов и координатор программы подготовки преподавателей факультета экономики и административных наук/бухгалтерского учета и финансов, Университет Сельчук, Конья, Турция https://orcid.org/0000-0002-0442-8499 eremsahin@selcuk.edu.tr
Conflicts of Interest Statement: The authors have no conflicts of interest to declare.
The article was submitted on 09.11.2023; revised on 07.12.2023 and accepted for publication on 11.12.2023.
The authors read and approved the final version of the manuscript.
