**2. Infrastructure and FDI inflow in India**

India, among the worldwide five major countries in the emerging economy, has significant potential to improve rapidly and thus made up to be an appropriate endpoint for FDI inflow. Despite all difficulties, India has attracted reasonable FDI inflow comparatively with other developing economic developed issues. In the 20th century's, policy reforms, Indian performed better and received comparatively greater FDI inflow [18]. The FDI inflow for the financial year 2009–2010 was 37,745 US million dollars, and over the ensuing period of 10 years, the FDI inflows have recovered stable growth in each year. For the year 2019–2020, the inflow was \$27.1 billion higher than the annual inflows in previous years from 2010 to date. The FDI inflows reduced during the year 2011–2012 and 2013–2014 where they fell by 8 per cent and 26 per cent respectively due to some reason like slowdown of economic development issues in India. **Table 1** has explained the input of specific sector FDI out of total FDI inflow in the Indian sector. The table illustrates the evocative variations in the infrastructure sectoral composite of FDI inflow in India from the last 20 years. An inclusive investigation of sector-wise FDI discloses that external investors favored in Manufacturing sectors throughout 2000–2020. The manufacturing sector has accounted for more than US\$ 89.40 billion from April 2000 to March 2020. During 2020 the Government of India increased FDI in manufacturing under the automatic route from 49–74%.

Similarly, the FDI inflow has significantly contributed to the above-mentioned three sectors and has shown the reforms of sectoral-FDI also account for significant variation for the time-to-time period. The distribution of the service sector out of total FDI inflow has increasingly and expansively improved in the same period. FDI equity inflow amount for services sector India FY 2015–2020. In the financial year 2020, the foreign direct investment equity inflow in the services sector in India was worth approximately 7.86 billion U.S. dollars. The foreign investment inflows have been consistently increasing over the last five years in this sector. To assess the post- and pre-reform of sector-wise FDI performance in the economy of India, this paper calculated the FDI performance index sector level that indicates the share of FDI sector-wise, comparative to its influence to aggregates of the India GDP. A value higher than 1 presents that the specific sector has recognized additional FDI inflow than its comparative economic size while a value less than 1 suggests that the specific sector received less FDI- inflow than its relative contribution; the below method was also used by Shah et al. [19].


#### **Table 1.** *Index of sector-level FDI performance.*

#### **2.1 Rural infrastructure**

Rural infrastructure in the country is crucial for agriculture, agro-industries and poverty alleviation in the rural areas. Rural infrastructure provides essential production conditions which are required for social and economic growth and for promoting the quality of life in rural areas. As per the government statistics clean tap water is available to only 24% rural households. About 56% of rural households had electricity connections. Centre and state government have over all estimated a total capital expenditure of Rs. 7,73,915 crore between fiscals 2020 and 2025 on rural infrastructure development in India.

According to the Department for Promotion of Industry and Internal Trade (DPIIT), the Indian food processing industry in rural has cumulatively attracted Foreign Direct Investment (FDI) equity inflow of about US\$ 10.24 billion between April 2000 and December 2020.

In the year 2021 infrastructure activities accounted for 13% share of the total FDI inflows of US\$ 81.72 billion. The government invested US\$ 1.4 trillion in infrastructure development as of July 2021.

Department of Drinking Water and Sanitation will be implementing the Jal Jeevan Mission to provide functional household tap connection to every rural household i.e., "Har Ghar Nal se Jal" by 2024. The program will be implemented at an estimated total capex of Rs. 3,60,000 crore shared between states and center as follows: Rs. 2,48,626 crore would be invested in rural housing under PMAY Gramin and about Rs 162,329 crore would be invested to improve rural roads under PMGSY. Improving the rural road connectivity by providing all-weather roads to connect eligible habitations in rural areas. As on December 31, 2019, road length worth Rs. 2.9 lakh crore had been sanctioned and expenditure of Rs. 2.17 lakh crore incurred. World Bank sanctioned about INR 2462 billion (US\$ 37 billion) through its Country Assistance Strategy committed to a series of loans/credits to support "Pradhan Mantri Gram Sadak Yojana (PMGSY) to complete 165,411 Road projects in rural areas. The total projected rural infrastructure investment from 2020 to 2025 is given in the Table below.

From the table given above it can be understood that, the rural infrastructure investment is 7% in the total infrastructure investment in India. The projected cumulative investment from 2020 to 2025 is 773,915 million rupees.

Sector-wise break-up of capital expenditure of Rs. 111 lakh crore during fiscals 2020–2025.


**Table 2.**

*Table shows the projected investment in rural infrastructure in India from 2020 to 2025. (rupees in crores).*

from the above diagram it can be understood that, energy sector 24%, roads sector 18%, railways 12%, ports 1%, Airports 1%, urban infrastructure 17%, digital infrastructure 3%, irrigation 8%, rural infrastructure 7%, agriculture & food processing 2%, social infrastructure 4% and industrial infrastructure 3%. Hence, it is concluded that, the total share of the rural infrastructure in total FDI is 7%.

$$\text{Present FDI inflow} = \frac{\text{FDI}\_{\text{t}}/\text{FDI}\_{\text{i}}}{\text{GDP}\_{\text{t}}/\text{GDP}\_{\text{i}}}$$

From the above equation used for the determination of present FDI inflow, whereas, FDIi inflow in the infrastructure sector I; FDIt is cumulative FDI inflow, GDPi&t indicates GDP of the infrastructure sector I and overall value of GDP is t.

**Table 2** has shown the variance between the infrastructure project performance during pre-and post-reforms periods of sector-wise FDI growth and better performance indices. The performance indexes illustrate that during the pre-and postreforms era, the major and important sectors are gas, oil, power sector, transportation, construction and mining sectors, which attracted FDI inflow and contributed to GDP growth. In the present situation, the Indian industries have overcome the shortage of electricity and the deficiency of proper infrastructure facilities. Both private and public manufacturing sectors are facing low-level problems against the lack of infrastructure issue, it looks like the latter is winning. Based on the literature review, the paper has tested the following two hypotheses.

Null Hypothesis (H0): There is no significant difference in FDI equity inflows to Infrastructure projects.

Alternative Hypothesis (H1): There is a significant difference in FDI equity inflows to Infrastructure projects.

### **3. Data and methodology**

#### **3.1 Explanation of variables and data gathering**

To assess the relationship between infrastructure and FDI inflow from 2000 to 2020, the paper depends on the global infrastructure index 2020 (GII-2020) a compound index and also sub-sector of infrastructures such as transportation (TI),

telecommunication (CI), power sector (PI) and energy sector (EI) and financial sector (FI) recognized on data collected from various sources (RBI, world bank and Global infrastructure index and CMIE reports).

The global infrastructure index 2020 (GII-2020) encompasses different quality and quantity magnitudes of infrastructure for India. The GII-2020 is created every year on a comprehensive range of infrastructure development parameter datasets of the accessibility and quality of infrastructure throughout 2000–2020. Besides, the paper used the institutional quality component, trade openness and human capital factors as control variables.

#### **3.2 Research methodology**

The present research investigates the two functional short and long-run causal dynamic relationships between infrastructure and FDI inflow, employing granger causality, ARDL (autoregressive distributed lag), and NARDL (Nonlinear autoregressive distributed lag) estimators to cointegration. This method is recognized in the case when the carefully chosen indicator is integrated either at the 1(0) level or the first difference I (1).

Moreover, from the simple linear transformation, the error ECM (�1) correction model easily may originate [16, 17]. To calculate the relationship between FDI and infrastructure the autoregressive distributed lag model assesses the following unlimited error correction model:

$$\begin{aligned} \Delta \text{FDI}\_{t} &= \alpha\_{0} \text{fdi} + \sum\_{\text{i}=\text{t}}^{\text{p}} \boldsymbol{\vartheta}\_{\text{fdi}} \text{FDI}\_{\text{-n}} \text{inf}\_{\text{t}-\text{i}} + \sum\_{\text{i}=1}^{\text{p}} \pi\_{\text{fdi}} \Delta \text{HC}\_{\text{-n}} \text{inf}\_{\text{t}-\text{i}} + \sum\_{\text{i}=1}^{\text{p}} \boldsymbol{\vartheta}\_{\text{fdi}} \Delta \text{TO}\_{\text{-n}} \text{inf}\_{\text{i}-\text{t}} \\ &+ \sum\_{\text{i}=1}^{\text{p}} \gamma\_{\text{fdi}} \Delta \text{IQ}\_{\text{-n}} \text{inf}\_{\text{t}-\text{i}} + \sum\_{\text{i}=1}^{\text{p}} \mathcal{Q}\_{\text{fdi}} \Delta \text{GI}\_{\text{-n}} \text{inf}\_{\text{I}-\text{I}} + \mu\_{\text{1d}\text{iii}} \text{FDI}\_{\text{-n}} \text{inf}\_{\text{t}-\text{i}} + \mu\_{\text{2d}\text{ii}} \Delta \text{HC}\_{\text{-n}} \text{inf}\_{\text{t}-\text{i}} \\ &+ \mu\_{\text{3d}\text{ini}} \Delta \text{TO}\_{\text{-n}} \text{inf}\_{\text{i}-\text{t}} + \mu\_{\text{4f}\text{ini}} \Delta \text{IQ}\_{\text{-n}} \text{inf}\_{\text{t}-\text{i}} + \mu\_{\text{5d}\text{iii}} \Delta \text{GI}\_{\text{-n}} \text{inf}\_{\text{I}-\text{I}} + \mathcal{D}\_{\text{t}} + \forall\_{\text{1}} \end{aligned} \tag{1}$$

$$\begin{aligned} \Delta \text{HC\\_inf\\_t} &= \alpha\_0 \text{fdi} + \sum\_{i=1}^{p} \in\_{\text{fdi}} \Delta \text{HC\\_inf\\_t} + \sum\_{i=t}^{p} \beta\_{\text{fdi}i} \text{FDI\\_inf\\_t} + \sum\_{i=1}^{p} \beta\_{\text{fdi}i} \Delta \text{TO\\_inf\\_t} \\ &+ \sum\_{i=1}^{p} \gamma\_{\text{fdi}i} \Delta \text{IQ\\_inf\\_t} + \sum\_{i=1}^{p} \mathcal{Q}\_{\text{fdi}i} \Delta \text{GI\\_inf\\_1} + \mu\_{\text{fdi}i} \text{FDI\\_inf\\_t} + \mu\_{\text{fdi}i} \Delta \text{HC\\_inf\\_t} \\ &+ \mu\_{\text{fdi}i} \Delta \text{TO\\_inf\\_t} + \mu\_{\text{fdi}i} \Delta \text{IQ\\_inf\\_t} + \mu\_{\text{fdi}i} \Delta \text{GI\\_inf\\_t} + \mathcal{D}\_t + \forall\_{\text{lt}} \end{aligned} \tag{2}$$

The measuring the long-run relationship between FDI and infrastructure this paper employs the bound testing techniques. The process of bound testing technique analysis of the hypothesis of no cointegration between the chosen indicator and the existence of cointegration between the indicators of study interest. The lower and upper bound critical values are significant role-plays as a determinant for the cointegration test [20]. If the calculated F-statistic value is higher than the upper bound critical value, then the H0 (Null hypothesis) is rejected. If the F-statistic value is lower than the lower bound critical value,

The Granger causality model using the I(I) of variables all over a VAR may cause uncertainty in the results in the existence of cointegration among selected variables. Hence, an advanced form of traditional Granger causality model relating the error correction method (ECM) is articulated in VECM as follow:

*The Linear and Nonlinear Relationship between Infrastructure and FDI in India DOI: http://dx.doi.org/10.5772/intechopen.101612*

$$\begin{aligned} \Delta \text{FDI}\_{t} &= \mathfrak{a}\_{0} \text{fdi} + \sum\_{i=t}^{p} \mathfrak{d}\_{\text{fdi}i} \text{FDI\\_inf}\_{t-i} + \sum\_{i=1}^{p} \mathfrak{e}\_{\text{fdi}i} \Delta \text{HC\\_inf}\_{t-1} + \sum\_{i=1}^{p} \mathfrak{f}\_{\text{fdi}i} \Delta \text{TO\\_inf}\_{i-t} \\ &+ \sum\_{i=1}^{p} \mathfrak{q}\_{\text{fdi}i} \Delta \text{IQ\\_inf}\_{t-i} + \sum\_{i=1}^{p} \mathfrak{g}\_{\text{fdi}i} \Delta \text{GI\\_inf}\_{i-1} + \mathfrak{q} \text{ECM}\_{t-1} + \text{D}\_{t} + \mathfrak{p}\_{\text{3t}} \end{aligned} \tag{3}$$

$$\begin{aligned} \Delta \text{HC\\_inf\\_t} &= \alpha\_0 \text{fdi} + \sum\_{i=1}^{\text{p}} \mathsf{e}\_{\text{fdi}} \Delta \text{HC\\_inf\\_t} + \sum\_{i=t}^{\text{p}} \partial\_{\text{fdi}} \text{FDI\\_inf}\_{t-i} + \sum\_{i=1}^{\text{p}} \beta\_{\text{fdi}} \Delta \text{TO\\_inf}\_{i=t} \\ &+ \sum\_{i=1}^{\text{p}} \chi\_{\text{fdi}} \Delta \text{IQ\\_inf}\_{t-i} + \sum\_{i=1}^{\text{p}} \mathfrak{g}\_{\text{fdi}} \Delta \text{GI\\_inf}\_{t-1} + \dots + \mathfrak{g} \text{ECM}\_{t-1} + \text{D}\_{t} + \mu\_{\text{g}} \end{aligned} \tag{4}$$

#### **3.3 The non-linear auto-regressive distributive lag model (NARDLM)**

According to Pesaran et al. [20] the cointegration test makes available proof of a linear relationship among the chosen variables. The current research paper also uses the NARDL [19] model to examine the existence of an association between FDI inflow and infrastructure in India. The non-linear auto-regressive distributive lag model [21] is a nonlinear extended form of the autoregressive distributive lag model for consistent impeding both short and long-run irregularity in the autoregressive distributive lag model.

The non-linear auto-regressive distributive lag model is calculated in the current paper that determines the short run and long run of the positive and negative partial sums. Thus, the non-linear auto-regressive distributive lag model contemplates the form of the resulting equation:

$$\begin{aligned} \Delta \text{FDI}\_{t} &= \alpha\_{0} \text{fdi} + \sum\_{i=t}^{p} \beta\_{\text{fdi}} \text{GII\\_inf}\_{t-i} + \sum\_{i=1}^{p} \mathsf{e}\_{\text{fdi}} \Delta \text{HC\\_inf}\_{t-1} + \sum\_{i=1}^{p} \beta\_{\text{fdi}} \Delta \text{TO\\_inf}\_{i-t} \\ &+ \sum\_{i=1}^{p} \gamma\_{\text{fdi}} \Delta \text{IQ\\_inf}\_{t-i} + \sum\_{i=1}^{p} \mathcal{Q}\_{\text{fdi}} \Delta \text{GI\\_inf}\_{i-1} + \mu\_{\text{fdi}} \text{FDI\\_inf}\_{t-i} \\ &+ \mu\_{\text{fdi}} \Delta \text{GII\\_inf}\_{t-1} + \mu\_{\text{fdi}} \Delta \text{TO\\_inf}\_{i-t} + \mu\_{\text{fdi}} \Delta \text{HC}\_{t-1} + \mathcal{D}\_{t} + \forall\_{\text{lt}} \tag{5} \end{aligned} \tag{6}$$

$$\begin{aligned} \Delta \text{GII}\_{\text{t}} &= \mathfrak{a}\_{0} \text{fdi} + \sum\_{i=t}^{\mathfrak{p}} \beta\_{\text{fdii}} \text{GII}\_{\text{-n}} \text{inf}\_{t-i} + \sum\_{i=1}^{\mathfrak{p}} \mathsf{e}\_{\text{fdii}} \Delta \text{FDI}\_{t-1} + \sum\_{i=1}^{\mathfrak{p}} \beta\_{\text{fdii}} \Delta \text{FDI}\_{i \text{-t}} \\ &+ \sum\_{i=1}^{\mathfrak{p}} \gamma\_{\text{fdii}} \Delta \text{IQ\\_inf}\_{\text{-n}} + \sum\_{i=1}^{\mathfrak{p}} \mathfrak{g}\_{\text{fdii}} \Delta \text{HC}\_{\text{l} = 1} + \sum\_{i=1}^{\mathfrak{p}} \beta\_{\text{fdii}} \Delta \text{TO}\_{\text{-n}} \text{inf}\_{i \text{-t}} \\ &+ \mathfrak{p}\_{\text{fdii}} \Delta \text{GII}\_{\text{-n}} \text{inf}\_{t-1} + \mu\_{\text{3fdii}} \Delta \text{TO}\_{\text{-n}} \text{inf}\_{i \text{-t}} + \mu\_{\text{5fdii}} \Delta \text{HC}\_{\text{l} = 1} + \mathfrak{D}\_{\text{t}} + \forall\_{\text{lt}} \end{aligned} \tag{6}$$

### **4. Empirical outcomes and argument**

#### **4.1 Descriptive statistics and unit-root testing**

**Table 3** explains descriptive statistics value, this table helps to highlight the how data descriptive statistics like. Descriptive statistics values comprise of several

observations with determined values, mean, minimum, maximum, central value and standard deviation point with corresponding variables to transportation infrastructure (T\_inf), telecommunication infrastructure (Te\_inf), energy sector infrastructure (E\_inf), Financial sector infrastructure (F\_inf), global infrastructure index 2020 (G\_inf-2020), quality in institutional approach (IQ\_inf), the primary sector of FDI inflow (PFDII), FDI inflow service sector (FDIIS), export and import to GDP or trade openness in infrastructure (T\_inf) and human capital (H-inf).

Ouattara (2004) illustrated that the level of stationary among all the chosen variables of the study was of interest to observe the probable variables of FDI inflow in infrastructure sectors wise during the long run and short run. Due to the circumstance that if the factors of the study interest are stationary at I (2) the estimated F-test value will not be significant. In the current paper, use the two types of tests are structural break analysis which is (1) and (1) contemplate the structural break in the given timer series data to examine the order of integration among the selected variables.

**Table 4** explore that each variable is integrated either at I(1) OR I(0) order and none of the indicators is stationary at I(2) order, According to (1) in this condition, the auto-regressive distributive lag model is suitable moderately another cointegration process. To assess the existence of a long-run relationship among chosen variables,


#### **Table 3.**

*Shares' of different economic groups in % of cumulative FDI inflow in India.*

