THE IMPACT OF INFRASTRUCTURE ON FOREIGN DIRECT INVESTMENT IN DEVELOPING COUNTRIES

Abstract. The purpose of this paper is to determine relationship between the level of development of infrastructure and foreign direct investment (FDI). As a case study example this paper uses Kazakhstan data of infrastructure and FDI for the period of 1994 till 2018. For defining their correlation this paper used autoregressive distributed delay approach (ARDL). The results showed positive and significant impact of infrastructure on FDI.

Introduction

Maintaining a favorable investment climate and stimulating the flow of foreign direct investment (FDI) into the country's economy is an important government task. Developing nations have lack of access to modern technology, and short of capital, that's why the outcomes of foreign direct investment are more important for them, than for developed countries. FDI resolves these problems along with allocating benefits among foreign investors and host nation. The different articles on FDI investigate various determinants having the common benefit of investor as well as domestic nation. In my research, I examine the impact of infrastructure on foreign direct investments in Kazakhstan for the period (1994-2018). There are certain sectors of the economy that are of priority for the country. The development of these sectors will have an impact not only on economic growth but also on the social sphere, as well as on the integration of Kazakhstan into the international community. It is a capital-intensive industry the development of which requires both foreign capital and strategic control of the state hard. To develop the economy, the state should attract and effectively use foreign direct investment, which is the basis of one of the directions of mutually beneficial economic cooperation between the two countries. With the help of foreign investment can improve the production structure of Kazakhstan's economy, create new high-tech production, modernize fixed assets and technically re-equip many enterprises to prepare professionals and workers, to introduce cutting-edge management, marketing and know-how, to fill the domestic market with quality domestic products at the same time an increase in exports to foreign countries.

Literature review

While searching the literature review it was found that opinions regarding this research were divided on two sides. It was said that FDI has as positive as negative factors. The inflow of foreign investment can have a negative impact on domestic producers. In work Hanif A., and Jalaluddin, based on the econometric model proposed by the authors, was analyzed for statistics on foreign direct investment in Malaysia from 1970 to 2011 and their impact on domestic investment. It was shown that foreign investments exert a pushing effect on domestic investment. Moreover, the growth in the volume of foreign investment does not contribute to a significant increase in capital accumulated in the country. It should be noted that the investor's choice often changes, depending on external factors. Along with the positive effects of changes in yield, attracting additional investment in the industry, there are also negative, stimulating outflow of capital from the given branch of the country. Many researchers note that investments often go into a saturation phase and over time are greatly reduced for several developing countries. Analyzing the statistics on investment in the countries of Southeast Asia, many researchers pay attention to how the decisions of transnational corporations are gradually changing at the choice of countries to invest their assets. In work of Thorbecke W., and Salike N. it is noted that if the primary goal of investment of large Japanese companies were the countries attributed to the Newly industrialized countries 1 wave (Taiwan, Republic of Korea), then in the late 90s the main flow of investment switched to China and Second-wave countries included in the ASEAN: Thailand, Indonesia, Philippines, etc. Following Japan, TNCs from other countries also began to redirect their financial flows, ensuring rapid development of the high-tech sector in the ASEAN countries. However, based on international experience, attracting a foreign direct investment leads to risks. The most dangerous for an independent state is economic dependence on investors. According to researches below the positive effects are : Khadaroo and Seetanah claim the gains rendered by infrastructure growth are associated with greater accessibility and reduction in transportation costs. In addition, public goods reduce the cost of doing business for foreign enterprises, which leads to maximization of profits. Recent empirical studies also suggest that public goods have a significant impact on the cost structure and productivity of private firms. Ehrenberg believes that if such types of infrastructure are not publicly distributed among local and multinational enterprises, then these enterprises will work with less efficiency, as they will have to build their own infrastructure, which leads to duplication and depletion of resources, and therefore government spending reduces transportation cost. Haughwout objects to the fact that the availability of public goods reduces the cost of private firms, even if there is no direct role of infrastructure in the productivity of production and the structure of costs of private firms . Poor infrastructure increases transaction costs and limits access to local and global markets, which ultimately hinders FDI in developing countries. Greater efficiency can be achieved with the expansion of infrastructure facilities by considering the commercial principle and transferring responsibility for providing infrastructure, although management or leasing contracts such as construction-operation-transfer (COT), building management and full privatization. Based on the research of Akram, what he conducted for the country of Pakistan, one sees the interconnection of infrastructure on the inflow of FDI. . In a scientific paper, Muhammad Akram pointed to a strong positive relationship between infrastructure and FDI, arguing the results of the economic model. An important contribution to the study was the work of Root and Ahmed on the study and description of the critical role of infrastructure for domestic FDI. Using data for Pakistan, a developing country, Akram built an economic model and linked various variables. The study of Akram is very important for my work as it opens the question for developing countries. Shatz and Venables discussed two reasons for locating FDI in a foreign country. The first is ‘horizontal’ or ‘base expansion’ which extends the economies of transportation costs, tariffs and access to a new market. . The second reason is economies of production cost as lower labor, capital and other inputs cost to maximize the profits. Such FDI is termed as ‘vertical’ or minimizing production cost. Iwanow and Kirkpatrick argue that a significant contribution to improving the quality of infrastructure in export performance. In addition, the study indicates quantitative results, according to which improving infrastructure by 10% will lead to an improvement in exports by 8% in a developing country. Sekkat and Varoudakis estimate that infrastructure has a significant attractiveness for FDI, even if it differs from the openness and investment climate in developing countries.

Methodology

In this study, five variables are used: foreign direct investment (FDI), market size (MS), infrastructure (PR and RW), and exchange rate (EXR), which are explained below by their theoretical and empirical aspects. The data were collected for the entire explanatory and dependent variable from World Development Indicators online from the official World Bank website, as well as from the official website of the Republic of Kazakhstan for the provision of statistics with annual intervals for the period from 1994 to 2018.

The purpose of this study is to examine the impact of infrastructure on other important factors that determine the inflow of FDI. The hypothesis that a middle income country like Kazakhstan with better infrastructure is more attractive to foreign firms is analyzed by including the appropriate proxy server in the reduced specifications of specifications for the demand for foreign direct investment coming into the country. We indicate the following equation for studying the impact of infrastructure on foreign direct investment, along with the size of the market and the exchange rate.

ln(FDIt )= β0 + β1ln(MSt) + β2ln(PRt) + β3ln(RWt)+ β4ln(EXRt)

Where FDIt, MSt, PRt, RWt and EXRt represent foreign direct investment, market size, infrastructure and exchange rate. Whereas, ln represent natural logarithmic form of series. Parameters β1, β2, β3 and β4 are the long run elasticity of foreign direct investment net inflows with respect to MS, PR, RW and EXR respectively.

Many cointegration testing approaches have been used in the literature, such as those used by Phillips and Hansen; Johansen and Juselius; Engle and Granger. The Engle and Granger test and Johasen-Juselius are the most widely used methods of cointegration identification (long-term equilibrium relationship) between variables. These methods require that all variables be stationary at the first difference, i.e. I(1). In the case of a small sample size, the low efficiency of these methods is observed Chaudhry and Choudhary. However, these approaches do not estimate the structural breaks and the order of integration of variables should be unique. The autoregressive distributed delay approach (ARDL) for cointegration avoids this limitation. Pesaran, Shin and Smith developed this approach, while Pesaran popularized it. The ARDL approach is more suitable for this study compared with the three other approaches. Given the above advantages of the ARDL approach to cointegration, we specify the following model:

θln(FDIt)=β0+∑_(t=1)^q(β_1t〖θln(FDI〗_(t-1) ) +∑_(t=0)^q(β_2t 〖θln(MS〗_(t-1) ) + ∑_(t=0)^q〖(β_3t 〖θln(PR〗_(t-1) ) 〗+∑_(t=0)^q(β_4t 〖θln(RW〗_(t-1) )+∑_(t=0)^q(β_5t 〖θln(EXR〗_(t-1) )+ β6ln(FDIt-1)+ β7ln(MSt-1)+β8ln(PRt-1)+β9ln(RWt-1)+ β10ln(EXRt-1)+Ut            

(4.2)

Where, θ is the first difference operator, q is the optimal lag length, β1, β2, β3, β4 and β5 represents the short run dynamics and β6, β7, β8, β9 and β10 represents long run elasticity. Before applying ARDL approach, we test the level of integration of all variables because if any variable is I(2) or above, ARDL approach is not applicable. For this we use Augmented Dickey-Fuller (ADF) test statistic.

After all the series are confirmed to be stationary at I(1) or I(0) or mixed, and none of the series is stationary at I(2), then we can analyze the cointegration among the variables by employing the ARDL bound testing methodology, that will consist two-step procedure. The first step is to select the optimal lags of the variables by applying the Schwarz Bayesian criteria (SBC). The second step is running the regression to determine the existence of long run relationships between the variables by computing the ARDL F-statistic for the joint significance of lag-level variables. We derived two hypotheses from equation (4.2) for the long run relationships. The first is null hypothesis, H0: β6=β7=β8=β9=β10=0 and the second is the alternative hypothesis, H1: β6‡β7‡β8‡β9‡β10‡0.

We will compare F-statistics with critical values tabulated by Narayan.

If the ARDL F-statistic crosses the upper critical bound, then a cointegration is present, if the ARDL F-statistic lies before the lower critical bound, then no cointegration exist. If all related variable series are integrated at I(1), then our assessment regarding cointegration is based on the upper critical bound; otherwise, we follow the lower critical bound if all the series of related variables are integrated at I(0). After confirming the existence of a cointegration between the variables, we move to the second step of examining the long and short run relationships, expressed as follows:

θln(FDIt)=β0+∑_(t=1)^q1(β_1t 〖θln(FDI〗_(t-1) ) +∑_(t=0)^q2(β_2t 〖θln(MS〗_(t-1) ) +  ∑_(t=0)^q3〖(β_3t 〖θln(PR〗_(t-1) ) 〗+∑_(t=0)^q4(β_4t 〖θln(RW〗_(t-1) )  +∑_(t=0)^q5(β_5t 〖θln(EXR〗_(t-1) ) + µECMt-1 + Ut

(4.3)

Where, q1, q2, q3, q4 and q5 represent optimal lag length, µ is the speed of adjustment parameter and ECM represent error correction term derived from long run relationship as given in equation 4.2.

Results

Unit roots of variables are tested before applying the ARDL approach to cointegration. Results of unit roots under Augmented Dickey-Fuller test are summarized in Table 2. As per results, ln(PR) and ln(RW) is stationary at first difference form at one percent level of significance, whereas all other variables ln(FDI), ln(EXR) and ln(MS) are stationary at five percent significance level at first difference. ARDL approach to cointegration can be applied in this case.

Table 2.

Unit Root Test

Note: *-1% significance level

           **-5% significance level

           *** - 10% significance level

Table 3.

F-Statistics for Testing the Existence of Long-Run Relationship

Note:* indicates lag order selected by the criterion

AIC: Akaike information criterion

SC: Schwarz information criterion

HQ: Hannan-Quinn information criterion

The presence of cointegration between variables is checked by using the Wald test. The computed F-statistic is compared to the critical values of the F-statistic provided by Narayan (2005), to decide whether a long-run relationship between these two variables exists.

 Table 4.

Wald Test

Critical values are cited from Narayan for number of regressors (k) = 5 and number of time periods (n) = 30 [36]. The number of time periods in our study is 21 but we use the critical values computed for a sample size of 30 because it is the smallest sample size for which they are calculated by the author. However, this should have only a small impact on the results. The data is shown in Table 4.

Table 5.

Long Run Results of ARDL (1,0,0,0) Model Dependent Variable “FDI”

Note:* - 1% significance level

     ** - 5% significance level

     *** - 10% significance level

Table 5 shows the results of long run relationship of the selected ARDL model (1,0,0,0) using Akaike Information Criterion (AIC) and Schwarz Bayesian Information Criterion (SBC) to select optimal lag length in the model and we got the same output under these two criterion. Table 5 shows the results that EXR is significant at ten percent and its coefficient (-0.07). MS is significant at five percent. In variables of infrastructure PR is not significant, RW significant at one percent.

Table 6.

Short-run Error Correction Representation of the Selected ARDL (1,0,0,0) Model Dependent Variable “FDI”

R²=0.89, Adj. R²=0.87

The results of error correction representation of the selected ARDL model are summarized in Table 6. Coefficients of variables with θ sign represent short run elasticity. The results shows that according to results of probability in short run most significant are PR and RW variables (0.017 and 0.000). We can observe that RW is significantly positive impact on FDI inflows in short and as well as in long run. Highly significance of ECM (-1) evidenced long run relationship among the variables. The speed of adjustment from previous year’s equilibrium in foreign direct investment to current period’s equilibrium is 90 %.

Table 7.

Diagnostic test statics

The probabilities of the calculated test statistics are shown in Table 7. The results above indicate that the estimated model does not seem to have any serious diagnostic problems such as serial correlation, heteroskedasticity. Residual distribution is normal.

Figure 10. Plot of cumulative sum of recursive residuals

The plots of both CUSUM and CUSUM of Squares Tests, which shown in Figure 10 and Figure 11, that are used to check parameter stability suggest that the model is stable during the sample period.

Figure 11. Plot of cumulative sum of squares of recursive residuals

Conclusion

The purpose of this study is to prove the impact of domestic infrastructure on the flow of foreign direct investment in developing countries, such as Kazakhstan. Along with the infrastructure, which is the cardinal variable in this study, additional variables have also been included, such as the exchange rate and the size of the market. In the infrastructure, we included two variables: roadways and railway lines. The motivation behind this work is getting the positive impact of the infrastructure on foreign direct investment; also we can compare which of the two variables included in the infrastructure has a greater impact. Moreover, in carrying out this work, we could not find any exclusive research that would concern the influence of the infrastructure on foreign direct investment in the case of Kazakhstan. By using time series data from 1994 to 2015 and by applying autoregressive distributive lag (ARDL), we find out about the significant positive impact in short and long run of infrastructure on FDI inflows in Kazakhstan. In the short term, the increase in railway lines by one percent leads to an increase in the inflows of FDI by 8.07 %, and the influence of road routes is insignificant. In the long term, an increase in railway lines by one percent leads to an increase in FDI by 7.91 %, an increase in road ways by one percent will increase FDI by 0.6 %. Thus, we found that comparing the railways and roads, the first one is more significant for increasing the inflow of FDI. Discussing other variables, the market size also has a positive long-term relationship, but in the short term, the effect is negative. The exchange rate also has a negative impact in the short term and positive in the long term. These results are according to our expectations. The obtained result confirms the theory of the positive impact of infrastructure on the inflow of foreign direct investment into Kazakhstan. A variable, such as railways and roads, plays an important role in infrastructure. For successful development and progress in the country's economy, government should pay attention to the provision of services and the construction of new railway lines and roads.

References:

1. Hanif A., Jalaluddin: “FDI and Domestic Investment in Malaysia” Model of investment with FDI (2013) pp. 17-18.
2. Thorbecke, W., and N. Salike. “Foreign Direct Investment in East Asia.” RIET Policy Discussion Paper Series no. 13, (2013) P-003.
3. Graham, E.M., Krugman, P.R.: Foreign Direct Investment in the United States. Washington: Institute for International Economics. (1995).
4. Khadaroo, A. J., & Seetanah, B. Transport and economic performance: the case of Mauritius. Journal of Transport Economics and Policy, 42, (2008), 1-13.
5. Akram M.: “The Impact of Infrastructure on Foreign Direct Investment: The Case of Pakistan”, (2010), pp. 271-273.
6. Root, F.R., & Ahmed, A. Empirical determinants of manufacturing direct foreign investment in developing countries. Economic Development and Cultural Change, 27, (1979), 751-767.
7. Iwanow, T., & Kirkpatrick, C. Trade facilitation, regulatory quality and export performance. Journal of International Development, 19(6), (2006), 735-753.
8. Sekkat, K., & Varoudakis, A.M. Openness, Investment Climate and FDI in Developing Countries. Review of Development Economics, 11(4), (2007), 607-620.
9. Froot, K. A., & Stein, J. C. Exchange Rates and Foreign Direct Investment: An Imperfect Capital Markets Approach. Quarterly Journal of Economics, 106, (1991), 1191–1217.
10. Wheeler, D., & Mody A. International investment location decisions: the case of U.S. firms. Journal of International Economics, 33, (1992), 57-76.
11. Engle, R., & Granger, C. Co-Integration and Error Correction Representation, Test and Telling. Econometrica, 55(2), (1987), 251-276.
12. Pesaran, H. M., Shin, Y., & Smith, R. J. Bounds Testing Approaches to the Analysis of Long-Run Relationships. Journal of Applied Econometrics, 16, (2001), 289-326.
13. Narayan, P. K. Fiji’s tourism demand: the ARDL approach to cointegration. Tourism Economics. 2004, vol. 10, no.2, pp. 193-206.