Acta Universitatis Danubius. Œconomica, Vol 14, No 4 (2018)
Investment of Foreign Capital in Commercial Banks in Kosovo and Unpredicted Profits from High Interest Rates
Nakije Kida^{1}
Abstract: Kosovo as a transition country has made continuous efforts to attract foreign capital, being confident that it could be a means of capital growth in Kosovo. The Government of Kosovo since 2000 has opened especially the banking sector for foreign investment. Even economic reforms have given special emphasis to this sector by supporting it with internal policies. This paper analyzes the relationship between FDI, interest rates and human capital in Kosovo. The model is evaluated using the Ordinary Least Squares (OLS) approach. The results show that statistically, the predictive variables, Human Capital (HC) and the high interest rates (IR), have significantly contributed to FDI in the service sector. This may be one of the youngest human capital factors with high interest rates in Europe. The fact that expansion of investments from this sector is missing is a sign that profits are repatriated to the home country. The “Crowdingout” effect of foreign capital invested in commercial banks exists. The implication of trade policy and FDI should be revised so that FDI, from the banking sector, is expanded with new banks, a descending factor of interest rates and a facilitating factor for businesses and employment in Kosovo.
Keywords: Kosovo; FDI; the services sector; interest rates; Crowdingout; human capital
JEL Classification: F21; G21; J24; E43
1. Introduction
Despite the positive growth rates (2016 with 3.5%), inequality in income distribution reflects high levels of total poverty of 29.7%, unemployment rate of 32.9%, export growth of 11.1% and imports by 3.5% in 2016, while the trade deficit continues with 11.2% in March 2017. (KSA, 2017) However, the bank lending increase, increase of remittance flow and increase in compensation by employees resulted in the increase of private consumption in Kosovo during 2015 and 2016. Whereas, FDI suffered decrease by 30% in 2016 (215.9 million euros) compared with the previous year (308.8 million euros). There is a dominance in real estate, rent and business activities (183.6 and 167.6 million euros), followed by construction and trade services with a fall in the processing industry at the country level of 2.8%. The FDI services sector, at the beginning (2000), replaced the existing economy structure and went to the industry where banks and insurance companies (local creditors) were formed, while increase of competition led to the “Crowdingout” effect, Kida N. (2017, p 7^{2}). Despite the liquid banking system, Kosovo enterprises have difficulty in accessing the finance sources, while 10 commercial banks, of which 8 with foreign capital, distributed in 258 units with 3341 employees, do not create free financial breathing. Annual credit growth of 10. 5%, deposits of 8.4%, interest on loans of 14% (in 2001) to 7.7% in 2017, take more than half of the profit of local businesses. The 16 microfinance institutions also have the same action strategy.
A Defined Line of Growth Theories, International Trade and FDI
Growth theories and the theories of International Trade and FDI have a defined line, ranging from a Ricardian economics which focuses on the concepts of production factors, and where Smith is of special contribution. The treatment of theories also continues in the 1950s with the SollowSwan model as a model of exogenous growth, since it considers the investment rate and technological progress as exogenous.
The contribution to the growth theory is unstoppable. To authors Paul Romer and Robert Lucas (19801990), technological progress and the role of human capital are the central point to their theoretical model. It is Robert Barro (1995, 1898), Barro J.R., SalaiMartin (2003) that outlined these models, emphasizing not only the technological process, but also the human capital and public policies as a contributor to growth in continuity.
As seen, the relationship of growth theories with international trade theories and FDI is mutual, since the authors and their contribution to the economic growth is mutual, based on Hymer’s contribution (1960), who presented the Industrial Organization Theory (IOT), or the Imperfection Theory of Market for FDI study, which explains the performance of firms according to different market structures.
Buckley and Casson (1976), Dunning and Rugman (1985), Caves (1996), etc., dealt with the expansion of his contribution. As a result of this cooperation derives Buckley and Casson’s Internalization (Internationalization) Theory, (1976, 1985) which explains that the full value of the owner's wealth is understood through the Interaction. Whereas, Dunning's Theory of International production known as OLI paradigm (1980), a combination of ownership advantages (O), location (L) and internalization (I), is also the motive of foreign investors.
Purpose, Objective, Research Question, Hypothesis, and Study Importance
The purpose of this study  is to provide response to the research question related to the relationship and impact of interest rates and human capital on the Foreign Direct Investment Services Sector in Kosovo based on endogenous theory.
Study Objectives – Study of the effect of interest rates, human capital alone and in cooperation with FDI_Services sector, how attractive FDI are from the same sector.
Hypothesis  H_{1}: There are positive relationships between the rates of interest and the FDI Services sector in Kosovo, while the trained Human Capital is attractive.
Research question  Are there relationships and positive effects between real interest rates, human capital and Foreign Direct Investment Services sector in Kosovo?
Reference Resources  This paper is based on secondary sources for the period 20072016. The main data sources are from the Kosovo Agency of Statistics (https://ask.rksgov.net/), the Central Bank of Kosovo (http://www.bqkkos.org/) and the Ministry of Education Science and Technology of Kosovo (www.mashtgov.net/).
Opportunities and limitations – Besides the contribution provided by the study that is increasing the time horizon of the study, compared to previous ones (finding the determinants of the FDI services sector), there are limitations, where the secondary data have not included a wide range of variables and years.
Importance of the Study  If the variables selected in the study as determinants of FDI services sector are positive or negative, but not zero, it is an important prognosis to guide government policies for or against these macroeconomic variables.
2. Study Metodology
The basis of the regression analysis is the sensitivity measurement of attracting process of FDI in the services sector (FDI_Services sector), as a dependent variable towards real interest rates (IR), human capital (HC) and FDI services sector, in cooperation with human capital (FDI*HC) in Kosovo as independent variables. In our case, through the descriptive statistical analysis, multiple regression and Pearson correlation, and the approach of the generalized least squares model OLS (Ordinary Least Squares), the following equation is specified.
HD_S.Sh_{t} = βₒ+ b β_{1} (IR_{t1}) + β_{2}( IHD *HC_{t1}) + β_{3} (HC_{t1}) + ξ_{t}……Eku.(1)
Where, β_{0} = constant; β_{1, }β_{2 }and β_{3} = regression coefficients that measure the change in percentage of the flow value of the FDI_Services sector towards IR, HC and FDI*HC from one period to the next. ξ_{t} = the term of error, while the term t_{1} implies a time lag of one year.
3. Data Analysis and Result of Study
Descriptive statistics – Table (1) the first best variable is FDI_services sector (FDI services sector_{t}), more concentrated around the average than other variables, because the distribution of values is 1.79 units from the average of 3.77 units, which is the lowest distribution.
Table 1. Descriptive statistics
Variables 
Average 
Standard Deviation 
Years  N 
FDI_services sector 
3.77E8 
1.79E8 
10 
HC (secondary level of high education) 
96.40 
14.175 
10 
FDI * HC (FDI in cooperation with HC) 
6.828412600000 
2.5250988088700 
10 
IR (Real Interest Rates) 
11.64726138210 
5.052225392178 
10 
Source: CBK, monthly data from time series, 20071016 ( by the author)
The second best variable, by distribution after FDI_services sector, is the FDI*HC logarithm when interacting with each other, having an average of 6.82 units with a distribution by average of 2.52 units, which shows that even in this case the distribution is concentrated during our period of analysis.
The third variable, better by distribution after the FDI_services sector, is the real interest rate logarithm with an average of 11.64 units with a distribution by average of 5.05 units, which shows that even in this case the distribution is concentrated during our period of analysis.
And finally, the fourth variable is Human Capital (HC), where the distribution of values is 14.17 units by average of 96.40 units. This indicates that this variable has no distribution concentrated throughout the period of analysis (20072016). In Table 2, there is the R value that represents the simple correlation that is 98.6 percent, which indicates a high degree of correlation. R^{2} can be explained as a measure of the model’s explanatory power 97.1%, which in this case is very large.
Table 2. Model Summary^{b}
Model 
R 
R Square 
Adjusted R Square 
Std. Error of the Estimate 
DurbinWatson 
1 
.986^{a} 
.971 
.957 
37.86266 
3.343 
a. Independent variable in %: IR, FDI * HC, HC
b. Dependent variable: FDI in %
Table (3), is ANOVA which shows that the regression model predicts the dependent variable in statistically significant way.
Table 3. ANOVA^{b}
Model 
Sum of Squares 
Df 
Mean Square 
F 
Sig. 
Regression 
2.821E17 
3 
9.404E16 
67.697 
.000^{a} 
Residual 
8.335E15 
6 
1.389E15 


Total 
2.905E17 
9 



a. Independent variable in %: IR, FDI * HC, HC
b. Dependent variable: FDI % of growth
The statistical significance of the regression model, p = .000^{a} is less than 0.05, and indicates that the regression model statistically predicts the result. Table 4 “Coefficients” provides us with the necessary information to predict whether the predictive variables HC, FDI*HC and IR have statistically contributed significantly to FDI_services sector.
Table 4. Coefficients ^{a}
Model 
Unstandardized Coefficients 
Standardized Coefficients 
T 
Sig. 
95.0% Confidence Interval for B 
Correlations 
Collinearity Statistics 

B 
Std. Error 
Beta 
Lower Bound 
Upper Bound 
Zeroorder 
Partial 
Part 
Tolerance 
VIF 

1 
(Consta nt) 
3.144E8 
1.567E8 

2.006 
.092 
6903.369 
6.979E8 






HC 
0.386 
1.223 
.075 
0.315 
.479 
4107.320 
2172.547 
.624 
.294 
.052 
.477 
2.094 

IHD*HC 
4.610 
5.459 
.759 
0.844 
.000 
316.566 
602.655 
.957 
.955 
.544 
.513 
1.949 

IR 
1.786 
3.046 
.356 
0.586 
.018 
2227.752 
3057672.820 
.820 
.796 
.223 
.392 
2.549 
a) Dependent variable: FDI_ Services sector in % of growth
Natural flow equation 1, is:
FDI_Services sector_{t} = 3.144  0.386+ 4.610+1.786 = 3.638
Thus, the FDI_services sector is expected to be 363 percent over the next period if a favorable macroeconomic climate for the services sector continues. We evaluate the coefficients through regression analysis for three independent variables:
i) Human capital (secondary level of education)  The estimation of the slope coefficient for human capital (HC) is around 0.386, suggesting that any negative change of human capital by 1 percent is accompanied by a change of 38.6 percent (in the same direction) in FDI_servicessector in Kosovo. Since the standard error of the slope coefficient is around 1.223, the value of t is (0.386/1.223) = 0.315. This implies that there is a fragile negative relationship between the FDI services sector and human capital. So, the human capital alone has a trivial effect (p = .478>0.05) in attracting the FDI services sector. Their training in the services sector creates enough knowledge about the use of technology and work machinery.
ii) IHD in cooperation with human capital (FDI*HC)  The estimation of the slope coefficient for the human capital (HC) is around 4.610, suggesting that any change in human capital of 1 percent when it cooperates with FDI is accompanied by a change of 461 percent (in the same direction) in FDI_services  sector in Kosovo. The value t is also estimated and calculated to determine whether the slope coefficient is significantly from zero (4.610/5,459 = 0.844). This implies that there is a strong relationship of 84.4 per cent between the FDI_services sector and Human Capital (FDI*HC). As it is seen, even in the case when FDI cooperates with Human Capital, there is a positive and significant effect (p = 0.00 < 0.05), in attracting the FDI_services sector.
iii) Real Interest Rates – The estimation of the slope coefficient is around 1.786, suggesting that any 1 percent change in interest rates estimated in euros in Kosovo is associated with a change of 1.786 percent (in the same direction) in the FDI services sector. We also evaluated the t value to determine if the slope coefficient is significantly different from zero. Since the standard error of the slope coefficient is around 3.046, the t value is (1.786 / 3.046) = 0.586 or 58.6 percent, this means that there is a strong positive relationship between interest rates and the FDI services sector in Kosovo. So, the Real Interest Rates have a positive and significant effect (p = 0.018<0.05) in attracting the FDI services sector, but they stifle Kosovo businesses that have little room to find funding sources for expanding their businesses.
Correlation between all variables  While Table 5 shows the correlation among the four variables this relationship is statistically significant and a positive p = 000 < 0.05, strong link (.957 **) between the FDI services sector and FDI*HC, a strong relationship (.820 **) also between the FDI services sector and IR and statistically significant p = 0.012 > 0.05. While the relationship between FDI services sector and the unskilled HC is fragile 0.027. In the table 6, after the diagnosis of collinearity, is confirmed that there are no serious problem with multicollinearity.
The eigenvalues are not close to 0, indicating that predictive variables are not correlated with each other and that small changes in data values cannot lead to major changes in coefficient estimations.
Table 5. Correlation

FDI  services sector 
HC 
FDI * HC 
IR 

Pearson Correlation 
FDI net inflow 
1.000 
.624 
.957 
.820 

HC 
.624 
1.000 
.589 
.708 

FDI * HC 
.957 
.589 
1.000 
.681 

IR 
.820 
.708 
.681 
1.000 

Sig. (1tailed) 
FDI  net inflow 
. 
.027 
.000* 
.002* 

HC 
.027 
. 
.036 
.011 

FDI * HC 
.000* 
.036 
. 
.015 

IR 
.002* 
.011 
.015 
. 

N 
FDInet inflow 
10 
10 
10 
10 

HC 
10 
10 
10 
10 

FDI * HC 
10 
10 
10 
10 

IR 
10 
10 
10 
10 
*. Correlation is significant at the 0.05 level (1tailed).
**. Correlation is significant at the 0.01 level (1tailed).
Table 6. Diagnostics of Colinearity^{a}
Model 
Dimension 
Eigenvalue 
Condition Index 
Variance Proportions 

(Constant) 
Human Capital 
FDI*Human Capital 
Real Interest Norm % 

1 
1 
3.708 
1.000 
.00 
.00 
.01 
.00 
2 
.257 
3.796 
.00 
.00 
.13 
.12 

3 
.032 
10.845 
.02 
.10 
.86 
.32 

4 
.004 
32.428 
.98 
.90 
.01 
.55 

Dependent variable: FDI (net flow) 
4. Conclusions and Recommendations
The following hypothesis is confirmed that "there are positive relationships between the rates of interest and the FDI services sector in Kosovo, while human capital is attractive when it interacts with this sector."
Strong correlation (r = .986 or 98.6 percent) and statistically significant (p = 0.00 < 0.05), and high explanatory power (R^{2} = .971 or 97.1 percent, and the corrected R .957 or 95.7 percent), result that 97.1 percent of the level of FDI services sector are dependent on the interest rates and human capital if it is trained and uses the advanced technologies that FDI bring (if FDI*HC).
This means that attracting investments from the services sector is not accidental, foreign firms have advanced technology and expertise to train Kosovo workers for using technology and management, this is also indicated by values p = .000 < 0.05. Banking sector has also shown high performance thanks to high interest rates for 17 years (in 2001 by 14%  to 7.7% in 2017). The measurement of capital itself has a statistically insignificant effect (p = .478 > 0.05) in attracting the FDI services sector since the general knowledge of education only, without its training and its adequate practice in advanced technology, does not suffice. Investing at high rates of interest from local businesses has created great risk of return. The research found that interest rates had a statistically significant and positive impact (p = 0.018<0.05) on FDI inflows in the services sector that opt to be funded by the parent society, while for the local businesses the effect is negative. The Kosovo government should offer fiscal incentives and free land allocation to serious Greenfied Investment, manufacturing sector and various types of industries that increase employment, while to the local firms, easier access to finance at lower interest rates, creating easier breathing opportunities.
5. References
Alfaro, L. (2003). Foreign Direct Investment and Growth: Does the Sector Matter?*. Harvard Business School, 2003. Available here: http://www.people.hbs.edu/lalfaro/fdisectorial.pdf.
Boreinsztein, E.; Gregory, J.D. & Lee, J.W. (1998). How does foreign direct investment affect economic growth ? Journal of International Economics, 45(1), pp. 115135. DOI: 10.1016/S00221996(97)000330.htpp:www.sciencedirect.com/science/article/pii.
Dunning, John H. (2006). Towards a new paradigm of development: implication for the determinants of international. Transnational Corporations, Vol. 15, No. 1 (April 2006). http://unctad.org/en/docs/iteiit20061a7_en.pdf.
Dunning, John H. & Fortanier, F. (2006). Multinational Enterprises and the New Development Paradigm: Consequences for Host Coutry Development. Georgia Tech Center for International Business Education and Research Working Paper Series 20072008 Working Paper 01107/08 https://pdfs.semanticscholar.org/69e6/6c5ed9391d18f007b44e98076813b4dd407f.pdf.
Barro J.R. & SalaiMartin, X. (2003). Economic Growth. Second Edition. MIT Press, 3003, pp.16, 50, pp. 6465. SBN13: 9780262025539 ISBN10: 0262025531. 2^{nd} Edition; Cambridge, Mass, The MIT.
Press (2003). http://mitpress.mit.edu/books/chapters/0262025531pref1.pdf.
CBK (2017). The financial system monthly information. http://bqk kos.org/repository/Jana/02017.
Jagi, S. (2007). Descriptive Statistics and exploratoty data analysis. pp. 118. htt://www.iasri.res.in/ebook/EB_SMAR/ebook_pdf%20files/Manual%20II/.
Kida, N. (2016). The Impact of Foreign Direct Investment on Gross Domestic Product in Kosovo.EUT. Albania. http://www.uet.edu.al/index.php/sq/publikimet.
Kida N. (2017). The impact of FDI on the structure of the Kosovo Economy. European Journal of Economics and Business Studies. 2017.http://ejes.euser.org.
Peter J. & Buckley et al. (2002). FDI, regional differences economic growth: panel data evidence from China. Transnational Corporations, 11(1), pp. 128.
1 Associate Professsor, PhD, Albanian, Address: Str, Esat Mekuli, Republic of Kosovo, Tel.: +38344205392, +386 49 603 666, Corresponding author: nakije.kida@universitetiaab.com, nakijekida@gmail.com.
AUDŒ, Vol. 14, no. 4, pp. 99106
2 Crowdingout occurs when FDI distorts the activity of domestic firms and other foreign associates from investment activity, removing them from businesses.
Refbacks
 There are currently no refbacks.
This work is licensed under a Creative Commons Attribution 4.0 International License.