Impact of FDI and Its Absorption Capacity on the National Innovation Ecosystems: Evidence from the Largest FDI Recipient Countries of the World

Abstract

Foreign direct investment (FDI) improves economic growth by stimulating native investment, facilitating technology transfers in the recipient country and increasing human capital development, thus playing a vital role in economic development. On the other hand, innovation is also considered one of the major drivers for the economic growth of a country. This study empirically investigates the impact of FDI and its absorption capacity on the national innovation system of the world’s top five largest FDI recipient countries for the period of 1990–2016. Using two-stage analysis (DEA and Tobit regression), we found that research and development expenditures, researchers in the host country and the number of patents, trademark and industrial design applications are positive drivers of the national innovation systems. Moreover, the FDI inflows positively impact the innovation efficiency in the host countries. However, the strength of this relationship depends on the availability of the absorption capacity of FDI in the host country. The result shows that the global financial crisis and inflation negatively impact the FDI inflows and innovation efficiency in the sample countries. It concludes that FDI inflows and the country’s strength of domestic absorption capacity are essential drivers for developing national innovation ecosystems.

JEL Codes: F1, F4, G10, M21

Keywords

FDI inflows national innovation systems absorption capacity of FDI innovation complementary assets FDI recipient countries

Introduction

Foreign direct investment (FDI) plays a significant role in developing the economic growth of its recipient country. Over the last two decades, numerous developing countries have devoted significant attention to encouraging the inflows of FDI to boost the economy (Zaman et al., 2012). The theoretical prospects for FDI inflows suggest that FDI is a combination of capital stocks, technology and technical know-how, which expands the market access, provides positive technical spillovers and improves human capital, hence promoting the economic growth of a country (Blomström et al., 1994; K. H. Zhang, 1999). FDI inflows play a significant role in the economic development of its recipient country. Whether local or foreign, investment in a county depends on different factors, and any change or move in the direction of these factors results in a significant change in investment. Investors always prefer to invest in a healthy and safe environment. Any unfavourable move in this environment makes the investor too much conscious. The investment becomes risky, and investors lose confidence in the host country due to political instability (Khachoo & Khan, 2012). FDI has been recognised mainly as a growth-enhancing factor in developing countries because it offers many advantages to the host country, that is, increases the stock of human capital, used for maintaining the current account deficit, eases the access of new technologies, introduces new modern technologies and, ultimately, boosts the national innovation system (Falki, 2009). The impact of FDI on the host country has been the empirical investigation in the economics literature.

The economic theory suggests two significant impacts of the FDI on the host country. First, inward FDI is considered an essential channel of technology transfer because multinational enterprises are more productive and innovative, and they invest more in research and development (R&D) activities than domestic enterprises. However, technology transfer cannot be taken for granted, and it will depend on the capacity of the firms to adopt and apply new technologies. The economic environment mainly allows knowledge transmission among domestic and foreign firms (Antonietti et al., 2015). The economic theory also suggests that FDI may have a pro-competitive effect on its host country. The entry of multinational firms boosts the competition in the domestic market and pushes domestic firms to search for new technologies, productivity improvements and utilisation of resources efficiently to compete with foreign firms (Kiriyama, 2012).

In order to gain the supposed advantages from the multinational firms, the governments of many countries offer different types of financial supports, that is, financial assistance, tax reliefs and some other types of benefits to attract foreign firms, since these policies infer high costs for public finance and they are justified only if the positive externalities on the host economy from FDI are significant. However, the impact of FDI on the host economies remains an open question.

In today’s world, the technological capabilities of a country are the key factors at the national and firm levels. The progress of technological abilities is the outcome of a multifaceted collaboration of incentive structure with institutional factors, technology efforts and human resources. The incentive includes incentives from competition and factor markets and macroeconomic incentives. The institutions include market and non-market institutions, training and technology institutions, and industrial institutions. The technological capabilities of a country or region are developed from all these interplay factors. These are the efforts of the agents in a country, that is, the Triple Helix Model that combines the strengths of government, industry and academia and the strengths of the linkages between these agents that determine the performance of a national innovation system (Fu et al., 2006; Lall, 1992).

The United States has been the world’s largest recipient of FDI since 2006 because multinational firms from all over the world recognise it as an innovative and stable market and the world’s largest economy. One of the main reasons for the more prominent FDI recipient is that the United States has implemented an Open Investment Policy since 1983, beginning with President Ronald Reagan. The policy treats all investors fairly and equitably under the law and encourages all countries to pursue such a policy (SelectUSA, 2018). The United States has the world’s largest consumer market, a growing energy sector, predictable regulatory environment, appropriate legal protections, highly innovative environment, and skilled and productive workers. This enables the United States to offer an attractive investment climate from firms around the globe.

The United Kingdom received $299.7 billion in FDI in 2016 and secured the second position among the largest recipient of FDI countries. The policies and schemes of the United Kingdom encourage foreign investors to start businesses in the United Kingdom. For this purpose, the United Kingdom Guarantees Scheme (UKGS) is implemented in the United Kingdom, which supports private investments in the United Kingdom infrastructure projects. UKGS works by offering a government-backed guarantee to help projects access debt finance where there have been unable to raise finance in the markets. The UKGS can issue up to £40 billion of guarantees and is open to at least 2026. As a result of UKGS, the United Kingdom was the second-largest recipient of the FDI in 2016 (GOV.UK, 2018).

China is the largest recipient of FDI after the United States and the United Kingdom. It received $176.6 billion in FDI in 2016. China has received a large part of the FDI since it launched economic reforms and called for foreign capital formation in its economy in 1979. After 1992, China’s speedy growth in FDI inflows has started after Deng Xiaoping visited China’s southern coastal areas and Special Economic Zones. After this China adopted a new approach, which opens the special regimes towards more nationwide implementations of different policies for the growth of FDI.

Netherland was the fourth largest FDI recipient country in the world in 2016, having an FDI inflow of $80.8 billion. The country is also the third global investor in terms of investment flows behind the United States and China. A robust international orientation towards FDI characterises the Dutch investment policy. Most Dutch companies are multinational by nature, whereas some are listed on the foreign stock markets. There are no regulatory restrictions on the FDI into Netherland. The FDI inflow reached the top in 2006 before the crisis, but it significantly fell and was affected during the financial crisis. Netherland’s innovative top sectors are among the world’s best sectors. It has top nine sectors, that is, horticulture and propagation materials, agri-food, water, life sciences and health, chemicals, high-tech, energy, logistics and creative industries. The government of the Netherland with the help of universities are working together in the top sectors for knowledge creation and innovation. As a result, Netherland is amongst the top innovative countries in the world.

The Republic of Ireland has been the most proactive in fostering economic development using different tools related to the industry policy, free trade environment and FDI promotion. Ireland was the fifth-largest recipient of FDI in 2016 with $79.2 billion FDI inflows. Ireland has been named the best country globally for attracting high-value FDIs for the sixth year in a row (The Irish Times, 2017). In Ireland, the business climate is favourable and it encourages the new start-ups, which are the reason this country is ranked among the top 20 countries (18th out of 190 countries) in the World Bank’s Doing Business 2017 ranking. In recent years, Ireland has been seen as a more innovative country than in previous years. Ireland was ranked in 13th place out of 50 countries in the latest Bloomberg Innovation Index. It was ranked first place in value-added manufacturing and productivity.

The main objective of this study is to empirically investigate the impact of FDI and its absorption capacity on the national innovation system of the world’s largest FDI recipient countries with special emphasis on the role of absorption capacity and complementary assets in defining the strength of integration and modification. The study is organised in the following sections. The next section presents the literature review and theoretical and conceptual framework for understanding the relationship between FDI, the absorption capacity of FDI and national innovation systems, followed by hypothesis development. The third section discusses the research methodology, which includes sample data, data analysis techniques, operational model and variables description. The fourth section presents the empirical evidence, and finally, the last section concludes the study with policy implications.

Literature Review

The contributions of FDI inflows in the national innovation systems fall into four categories. First, R&D expenditures, R&D labs and other forms of innovation created by multinational firms directly increase innovation output. The increase in FDI is found to be a factor that contributes to the emergence of economies with sophisticated technology creation (Athreye & Cantwell, 2007). In recent decades, multinational firms’ globalisation of R&D activities has played a significant role in international business. Multinational firms generally have three global R&D programmes, internationally independent laboratories, support laboratories, and locally integrated laboratories (Pearce, 1999). The operations of foreign firms in host countries increase the supply of high-value parts, resulting in the development of locally integrated laboratories.

Second, spillovers of technological innovation from foreign countries may affect the national innovation performance. There are many channels of knowledge spillovers from foreign firms to domestic firms, such as skilled labour turnover, knowledge transfers through the supply chain and demonstration effects (Görg & Greenaway, 2004). The knowledge transfers require effective quality linkages among foreign and domestic firms. Similarly, the local firms may take benefit from the multinational firms when cross-regional labour mobility is low. The demonstration effects may also favour local firms only if it closely observes and replicates in the host country.

Third, the inflow of FDI may affect the national innovation capacity through the competition effect. Competition in the market is essential for boosting the innovation capabilities of the local firms. For instance, Geroski (1990) argues that lack of competition in the markets increases the inefficiency of the markets and results in sluggish innovation activity in markets. Similarly, Hu and Jefferson (2002) found significant productivity rather than positive spillover impact of FDI inflows on local Chinese firms.

Finally, in addition to the enormous spending on R&D activities by multinationals firms and their affiliates, FDI may contribute to the national innovation systems by advanced practices and experiences in R&D management and innovation systems. Innovation efficiency cannot be achieved through the simple linear transformation of basic science and other inputs and commercialisation, but it requires efficient R&D, coordination, management techniques, motivation, long-term planning and high-quality decision-making (Helinska-Hughes & Hughes, 2003). Therefore, national innovation efficiency is determined not only by R&D investments and workforce but also by certain soft factors such as management practices, motivational level and governance structures of the countries. The multinational firm plays a major role in national innovation because they are more experienced in innovation management. The local firms may take benefit from the managerial know-how spillovers effect by foreign firms.

The present study bridges the gap between two streams of past studies. For instance, Cheung and Ping (2004) conducted a study in China to find the spillover effects of innovation using the provincial data from 1995–2000. They found a positive relationship between FDI and innovation. Similarly, Girma et al. (2006) investigated whether FDI inflows impact product innovation in state-owned enterprises of China. The sample data used in the analysis consists of firm-level panel data from 1999 to 2003. The findings show that foreign capital participation is positively associated with innovation activity. The FDI inflow has a negative impact on the innovation of these sectors. Furthermore, they found that the state-owned enterprises are successful innovators with their internal R&D activity and human capital development.

Kuemmerle (1999) investigated the drivers of FDI into R&D laboratories using a sample data set of 32 electronics and pharmaceutical companies’ R&D laboratories. Econometric results of 136 laboratories show that the strength of a country’s science base and its relative market size determine the FDI role in R&D in Japanese, European and US firms. Barrell and Pain (1997) analysed the factors behind the growth of FDI and its consequences in the home and host countries. The findings suggest that the extent to which technology transfer from multinational firms has affected the rate of technological progress within the United Kingdom and German economies. The growth of FDI largely depends upon the acquisition of firm-specific knowledge. Potterie and Lichtenberg (2001) empirically investigated whether FDI transfers technology across borders. The results indicate that FDI transfers technology across borders but only in one direction. The results suggest that countries’ productivity is increased if it invests in the R&D intensive countries.

There are two main conditions for substantial spillovers from FDI suggested by the literature. First, the absorption capacity of the domestic firms and organisations, and second, effective linkages generated between foreign and local economic activities (Cohen & Levinthal, 1989; Fu, 2004; Girma, 2005). The absorption capacity refers to the ability of a country, organisation or firm to adapt, identify and exploit knowledge from the environment (Cohen & Levinthal, 1989). The absorption capacity is usually proxied by human capital in local firms, R&D intensities and the technology gap between foreign and domestic firms. Absorption capacities are considered the essential elements of national innovation systems because it determines the overall absorption capacities of a host country. In addition to the absorption capacity of the host country, its complementary assets also play a significant role in overall countries innovation capabilities that are required to convert technological opportunities into an efficient manner and to gain a competitive market advantage (Cosh et al., 2007; Hughes & Morton, 2006). Hence, a countries’ responsiveness to innovative ideas, the strength of information and communication structure, and the strength of entrepreneurship may enhance its capability to assimilate innovative ideas and technology generated locally or transferred from abroad.

Conceptual Framework

Based on the literature, the conceptual framework of this study is presented in Figures 1 and 2. Figure 1 shows Model 1 of the study in which we measure the impact of FDI on the innovation of the largest FDI recipient countries of the world, with control variables which include country effect, time effect and financial crisis effect. Similarly, Figure 2 shows Model 2 of the study, which incorporates the impact of FDI inflows with its absorption capacity and complementary assets for innovation on the innovation efficiency of the largest FDI recipient countries of the world.

Figure 1.

Source: Author estimation.

Figure 2.

Source: Author estimation.

In this study, in order to determine the impact of FDI and its absorption capacity on the innovation and innovation efficiency of the largest FDI recipient countries of the world, the following hypotheses have been proposed.

FDI and Innovation

The relationship between FDI and innovation is well defined in the literature and was among the interest of researchers around the world. FDI can benefit the innovation activity in the host country via different spillover channels. First, the local firms can learn the different technologies and innovative products experience brought in by the investors from abroad. Second, these spillovers can be beneficial in terms of labour skills and technological know-how from foreign investors, and finally, the inward FDI shows the practical demonstration effect on the local R&D activity (Aitken & Harrison, 1999; Cheung & Ping, 2004). The current study is closely related to the spillover theory. The spillover theory assumes the conditions under which the spillover effect among microsystem and family microsystem is negative or positive (E. J. Hill et al., 2003). The spillover effect is closely related to commons in the economist’s concept of externalities. An externality is a benefit or cost imposed on others due to some economic activity, and its results may be positive or negative. The social cost is another economic-related concept, in which when a negative externality occurs, the agent that causes the externality incurs a personal cost (Levin, 2013). Based on these theories and theoretical framework, this study constitutes following hypothesis for relationship between FDI and innovation.

H₁: There is a positive relationship between FDI and innovation of the largest FDI recipient countries of the world.

Absorption Capacity of FDI and Innovation Efficiency

Khordagui and Saleh (2016) explain that FDI can influence the host country’s economy with the necessary factors such as complementary assets to absorb the FDI in the host country. This raises the issue of absorption capacity that is the ability of the country to absorb the benefits that FDI can offer. They found that the FDI spillover effect exists in the Middle East and North Africa (MENA) economies and is more evident when controlling the schooling factor as the absorptive capacity factor.

The two stages of absorbability of FDI are (a) to bring FDI projects to the practice and (b) to convert the benefits of FDI into the host country’s competencies (Cohen & Levinthal, 1990). The second-most mentioned factor is the labour force described by human capital and education, which are essential for absorbing and adapting foreign technology and generating sustainable long-run growth (Blomström & Kokko, 2003). For utilising the capacity of FDI, there is a need to have enough complementary assets in the host country. In order to generate profits in the markets, the host countries need both codified and tacit technical know-how to be utilised in conjunction with other capabilities or assets (Teece, 1986). Hence, transitions in institutional and market environments will interact and shape how firms acquire and control their resources and complementary assets. Therefore, the hypotheses for absorption capacity of FDI and complementary assets to improve the innovation efficiency are as follows:

H_2a: FDI improves the innovation efficiency of the largest FDI recipient countries of the world.

H_2b: There is a relationship between the absorption capacity of FDI of host country and innovation efficiency of the largest FDI recipient countries of the world.

H_2c: There is a relationship between the complementary assets of the host country and the innovation efficiency of the largest FDI recipient countries of the world.

Research Methodology

Data and Methods

The target population of this study is the top FDI recipient countries of the world. The selected sample is the top five largest FDI recipient countries. Table 1 shows the sample countries’ detailed overview and rankings according to the Best Countries 2016 by US News and World Report. The US News and World Report is a multifaceted digital media company dedicated to helping policymakers, professionals and business leaders make crucial decisions. The platform includes cars, travel, money, health, education and 360 reviews. Best countries is a ranking provided by US News and World Report created to capture how countries are perceived on a global scale. The ranking evaluates 80 countries across 24 rankings from a survey of more than 21,000 global citizens. The rationale behind choosing the sample countries was that the United States (4,084,000 million USD), the United Kingdom (2,027,000 million USD), China (1,514,000 million USD), Netherland (4,888,000 million USD) and Ireland (1,477,000 million USD) hold the most extensive stock of FDI at home, by the end of 2016 (CIA, 2016).

Table 1.

Top Five Countries Receive Most of the FDI, 2016 Ranking.

Country	Rank	FDI Inflow ($) Billion
United States	1	479.4
United Kingdom	2	299.7
China	3	170.6
Netherland	4	80.8
Ireland	5	79.2

Source: Author estimation.

This study uses a sample period from 1990 to 2016. The reason for choosing this time frame was to capture the impact of FDI from 1990 to 2016 because in 2016, three-quarters of FDI ($1.8 trillion) went to these countries, and this was the highest level of FDI recorded since 2009. However, the worldwide inflows of FDI were slightly decreased by 1% in 2016 (Bank, 2018). The study used secondary data for data analysis, and the sources of the data were the World Bank Database. This study differs from the other studies because we use data for R&D expenditures, R&D staff, government expenditures on tertiary education and technicians in R&D rather than using data for investment in innovation.

In order to achieve the objectives of this study, we applied different statistical and econometric techniques such as descriptive statistics, correlation matrix, Data Envelopment Analysis (DEA) and Tobit regression. This study uses a two-stage analysis approach. In the first stage, DEA analysis was performed to find out the innovation efficiency using multiple inputs and outputs of the innovation. A common assumption is that different countries face certain environmental and technical factors for innovation (Simar & Wilson, 2007). Therefore, it is essential to examine the determinants of innovation efficiency first. The second stage of this study involves the regression analysis using the censored Tobit regression model. The Tobit regression model is more appropriate because the dependent variables innovation and innovation efficiency are bound censored having values in a defined range (Tobin, 1958). The calculated innovation efficiency score from DEA analysis is limited and constrained to the intervals 0 and 1.

Innovation Efficiency Using Data Envelopment Analysis (DEA)

The innovation efficiency is calculated using DEA analysis, allowing the model’s multiple outputs and multiple inputs. DEA analysis is suitable for organising and analysing complex such as innovative production systems (Guan & Chen, 2010). DEA analysis is non-parametric because all deviations from the frontier are assumed to be the result of technical inefficiency. DEA analysis is an effective way to utilise multiple outputs and inputs with different units of measurement simultaneously (Fritsch & Slavtchev, 2006). We use DEA analysis to estimate the national innovation efficiency of sample countries because this method allows us to evaluate efficiency in innovation against best practices by the countries and allows us to estimate innovation efficiency with multiple inputs and multiple outputs. In this approach, for a sample of n country, if X and Y are the observations on innovation inputs and outputs, assuming variables returns to scale, the country’s innovative efficiency score θ is the solution of the linear problem.

\begin{matrix} \max_{θ, λ} \\ s t . - θ y_{i} + y_{λ} \geq 0 \\ x_{i} - X λ \geq 0 \\ λ_{i} \geq 0 \\ \sum^{​} λ_{i} = 1 i = 1, \dots n, \end{matrix}

(1)

where θ is the scalar and λ is an n × 1 vector of constraints. The efficiency score of innovation ranges from zero to 1. Similarly, if θ k = 1 and all slacks are zero, the kth country is deemed to be technically efficient (Cooper et al., 2011). The innovation output is measured by the number of patents, trademark and industrial design applications each year by a country. Inputs in our model include R&D expenditure by the country, researchers in R&D and technicians in R&D. We use this approach because this method allows us to evaluate efficiency in innovation against the best practice (Charnes et al., 1978).

Tobit Regression Model

Recently, econometric models with censored error terms are used widely in different types of studies (Aldieri et al., 2021; Guo et al., 2021; J. Zhang et al., 2021). The Tobit regression model assumes that the dependent variable has a number of its values clustered at a limiting value, usually zero. The dependent variables, that is, innovation efficiency and innovation, have limiting censored values in this study. In order to overcome the problem of endogeneity, where independent variables are correlated with the error term, we applied the Tobit regression model by estimating two equations (Equations 3 and 4). The proposed Tobit model performs well under the endogeneity problem (Ngo & Tsui, 2020). We run Equations 3 and 4 of this study under two parameters, that is, (a) full sample analysis and (b) individual country-wise analysis.

The general equation for the Tobit regression model is as follows:

Y_{i}^{*} = X_{i} β + ε_{i},

(1)

where Yi* is the latent variable that is observed for the values greater than Ʈ and censored otherwise.

X_iβ is the independent variable, and its coefficient and ε_i is the error term.

The observed y is defined as follows:

Y_{i} (\begin{matrix} y * i f y * > τ \\ T y * i f y * \leq τ \end{matrix}) .

(2)

Equations 3 and 4 are the operational model of this study.

\begin{array}{l} L_I N N O V_{i}^{*} = β_{1} (L_F D I) + β_{2} (L_G R O W T H) + \\ β_{3} (L_I N F L A T I O N) + β_{4} (C R I S I S) + ε_{i} . \end{array}

(3)

\begin{array}{l} I N N O V_E F F_{i}^{*} = β_{1} (L_F D I) + β_{2} (F D I \times A B C) + β_{3} (F D I \times C A + \\ β_{4} (L_G R O W T H) + β_{5} (L_I N F L A T I O N) + β_{6} (C R I S I S) + ε_{i} . \end{array}

(4)

The details of the variables used in Equations 3 and 4 with their measurements are presented in Table 2.

Table 2.

Measurement of Variables.

Variable Name	Abbreviation	Measurement
Foreign direct investment inflow	FDI Foreign direct investment, net inflows(% of GDP)
Innovation	INNOV No. of patent applications + No. of trademark application + No. of industrial design applications
Research and development	R&D R&D expenditure (% of GDP)
Researchers	RES Researchers in R&D (per million people)
Technicians in R&D	TEC_R&D Technicians in R&D (per million people)
Start-up cost	COST COST is the Cost of business start-up procedures cost is measured as Cost of business start-up procedures (% of GNI per capita).
Industry value-added	IVA IVA is the industry value added % of GDP.
Labour force quality	LAB_QUALITY LAB_QUALITY is the school enrolments at the tertiary level.
Absorption capacity of FDI	ABC ABC is the product of R&D and LAB_QUALITY.
Complementary assets	CA CA is the product of IVA,TEC_R&D and COST.
An interaction term for foreign direct investment and absorption capacity	FDI × ABC Product of FDI and absorption capacity (ABC).
An interaction term for foreign direct investment and complementary assets	FDI × CA Product of FDI and complementary assets (CA).
Innovation efficiency	INNOV_EFF Is measured by DEA analysis with inputs of R&D, RES,TEC_R&D and output of INNOV.
Economic growth	GROWTH GROWTH is the gross domestic product, current USD.
Inflation	INFLATION INFLATION is measured as the consumer price index.
Global financial crisis	CRISIS Financial crisis dummy, equal to 1 in financial crisis period 2008–2011 otherwise 0.

Source: Author estimation.

Results and Discussion

Results of Descriptive Statistics

Descriptive statistics analysis is used to describe the various features of the data set. In general, as compared to inferential statistics, descriptive statistics are not based on the probability theory. These types of analysis are generally presented to describe the central tendency and variability of the data set. Measures of central tendency include mean, median and mode, whereas measures of variability include standard deviation, range, variance, maximum and minimum values in the data. This type of analysis is necessary before running any econometrics model because it provides the overall first look at the data and the extreme values included in the data set. Tables 3–8 describe the results of descriptive statistics of data and variables used in this study. Table 3 displays the summary of descriptive statistics of the full sample. CRISIS variables identify the dummy variable used in the study in which one is for crisis period that is 2008 to 2011 and zero otherwise. The results show that the variables used in the study are normally distributed. Initially, most of the variables were not normal, so we converted them into natural logarithms to make them normal. Similarly, Tables 4–8 represent the individual sample countries descriptive statistics.

Table 3.

Descriptive Statistics of Full Sample.

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
Mean	10.6708	0.7259	17.3607	27.8772	86.4854	70.4909	167.7010	0.1481
Maximum	15.0550	1	20.4838	30.5620	117.2206	89.2238	286.8137	1
Minimum	7.3809	0.0259	14.1139	24.6215	40.4403	56.3158	77.7030	0
Standard Deviation	2.3245	0.2965	2.1147	1.6049	16.6395	8.8537	56.3205	0.3566
Skewness	0.0414	–1.1582	0.0663	–0.0827	–0.3031	0.7010	0.1692	1.9809
Kurtosis	1.5958	3.0288	1.7487	2.1193	2.5478	2.4863	1.8327	4.9239

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Table 4.

Descriptive Statistics of United States Sample.

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
Mean	12.9675	0.7834	18.8088	30.0432	85.8521	73.6679	226.9986	0.1481
Maximum	13.4772	1	18.9070	30.5620	110.0670	76.5032	238.1500	1
Minimum	12.3048	0.4919	18.6671	29.4166	59.9198	70.3628	212.5839	0
Standard Deviation	0.3463	0.1649	0.0742	0.3590	16.0384	2.0477	8.4664	0.3620
Skewness	–0.5227	–0.3878	–0.5469	–0.2716	0.0281	–0.3053	–0.3591	1.9809
Kurtosis	2.2753	2.0005	2.0306	1.7945	1.6761	1.7791	1.7602	4.9239

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Table 5.

Descriptive Statistics of United Kingdom Sample

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
Mean	10.9797	0.8508	17.2291	28.2962	87.4744	67.1651	168.6087	0.1481
Maximum	11.4039	1	17.3328	28.7603	112.0769	70.1118	171.9486	1
Minimum	10.8164	0.5465	17.1626	27.6906	61.0967	63.8277	165.3196	0
Standard Deviation	0.1393	0.1263	0.0574	0.3641	14.9397	2.2868	1.7429	0.3620
Skewness	1.5665	–0.7271	0.4083	–0.3127	0.1740	–0.2973	0.1442	1.9809
Kurtosis	4.9269	2.5353	1.6885	1.6401	1.9710	1.3671	2.2394	4.9239

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Table 6.

Descriptive Statistics of Netherland Sample.

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
Mean	7.7000	0.8789	15.9205	27.0730	88.2544	64.7479	122.6007	0.1481
Maximum	7.8575	1	16.0183	27.5776	109.5290	68.5212	125.7893	1
Minimum	7.3809	0.7311	15.7488	26.4864	64.4304	59.9752	116.3622	0
Standard Deviation	0.1269	0.0935	0.0915	0.3718	14.2988	2.5787	2.5069	0.3620
Skewness	–1.0532	0.0690	–0.5736	–0.0693	–0.0534	–0.3254	–1.1158	1.9809
Kurtosis	3.3950	1.5217	1.8375	1.4324	1.7487	1.7361	3.4018	4.9239

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Results of Correlation Matrix

When multiple variables are included in the model, there is a possibility of the problem of multicollinearity in the model. The major problem with multicollinearity is that the least squares estimators of coefficients of variables involved in the linear dependencies have significant variances, and the results become biased (Alin, 2010). Multicollinearity may adversely affect the estimated coefficients in the model in multiple regression; therefore, it is essential to detect its existence (Mansfield & Helms, 1982). A correlation matrix is one of the methods to detect this problem. Tables 9–14 represents results of correlation matrix of sample countries. The results confirm that there is no problem of multicollinearity in the estimated models of the study. Table 13. Correlation Matrix of Ireland Sample.

Table 7.

Descriptive Statistics of Ireland Sample.

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
Mean	8.5142	0.8471	14.4272	25.6345	86.3038	61.5325	94.1038	0.1481
Maximum	9.0180	1	14.6354	26.4256	105.6623	65.8712	100.0670	1
Minimum	8.1259	0.2810	14.1139	24.6215	60.9548	56.3158	77.7030	0
Standard Deviation	0.2680	0.2178	0.1875	0.6386	16.1368	3.0016	6.1541	0.3620
Skewness	0.0445	–1.6522	–0.4181	–0.3134	–0.1700	–0.4199	–1.4488	1.9809
Kurtosis	1.8367	4.3382	1.6147	1.5440	1.4314	1.9383	4.0562	4.9239

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Table 8.

Descriptive Statistics of China Sample.

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
Mean	13.1927	0.2693	20.4178	28.3393	84.5422	85.3413	226.1933	0.1481
Maximum	15.0550	1	20.4838	30.0499	117.2206	89.2238	286.8137	1
Minimum	11.1059	0.0259	20.2769	26.6118	40.4403	78.6784	175.8233	0
Standard Deviation	1.2021	0.2795	0.0669	1.1388	21.7497	3.2159	35.8709	0.3620
Skewness	–0.0606	1.4443	–0.7676	0.1069	–0.5306	–0.7275	0.2498	1.9809
Kurtosis	1.7194	3.8427	2.2568	1.7275	2.6652	2.3210	1.6488	4.9239

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Table 9.

Correlation Matrix of Full Sample.

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
L_INNOV	1	–	–	–	–	–	–	–
INNO_EFF	–0.3975	1	–	–	–	–	–	–
L_FDI	0.0883	–0.6126	1	–	–	–	–	–
L_GROWTH	0.1003	–0.0890	0.6414	1	–	–	–	–
INFLATION	0.0845	0.1930	–0.0097	0.3376	1	–	–	–
FDI_CA	0.0154	–0.6269	0.0107	0.5293	–0.1166	1	–	–
FDI_ABC	0.0552	–0.3370	0.0270	0.0872	0.1212	0.0312	1	–
CRISIS	0.0399	0.0512	0.0175	0.1378	0.3467	–0.0833	0.0599	1

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Table 10.

Correlation Matrix of United States Sample.

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
L_INNOV	1	–	–	–	–	–	–	–
INNO_EFF	0.0854	1	–	–	–	–	–	–
L_FDI	0.0855	0.0823	1	–	–	–	–	–
L_GROWTH	0.0814	0.0837	0.0926	1	–	–	–	–
INFLATION	0.0594	0.0661	0.0707	0.0899	1	–	–	–
FDI_CA	–0.3826	–0.4136	–0.4018	–0.4762	–0.5905	1	–	–
FDI_ABC	0.0768	0.0834	0.0899	0.0939	0.0790	–0.4535	1	–
CRISIS	0.2810	0.2649	0.3649	0.3438	0.3767	–0.4290	0.3270	1

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Table 11.

Correlation Matrix of United Kingdom Sample.

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
L_INNOV	1	–	–	–	–	–	–	–
INNO_EFF	–0.6590	1	–	–	–	–	–	–
L_FDI	0.5841	–0.609	1	–	–	–	–	–
L_GROWTH	0.3576	–0.2690	0.1846	1	–	–	–	–
INFLATION	0.5753	–0.0758	0.3544	0.1171	1	–	–	–
FDI_CA	–0.3206	0.4241	–0.5643	–0.0754	–0.3055	1	–	–
FDI_ABC	–0.4070	0.2783	–0.4431	–0.5252	–0.5844	0.6579	1	–
CRISIS	–0.3172	–0.2106	0.3667	0.3441	0.3358	–0.5423	–0.3732	1

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Table 12.

Correlation Matrix of Netherland Sample.

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
L_INNOV	1	–	–	–	–	–	–	–
INNO_EFF	0.6689	1	–	–	–	–	–	–
L_FDI	0.6184	0.0000	1	–	–	–	–	–
L_GROWTH	0.5607	–0.0155	0.3266	1	–	–	–	–
INFLATION	0.5470	–0.0833	0.4651	0.3379	1	–	–	–
FDI_CA	–0.1509	0.1297	–0.5093	–0.6716	–0.6746	1	–	–
FDI_ABC	0.1753	0.5365	0.0765	0.2102	0.6097	–0.2656	1	–
CRISIS	0.4504	0.3455	0.3866	0.5057	0.3391	–0.4134	0.4722	1

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Table 13.

Correlation Matrix of Ireland Sample.

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
L_INNOV	1	–	–	–	–	–	–	–
INNO_EFF	0.6156	1	–	–	–	–	–	–
L_FDI	–0.6603	–0.4876	1	–	–	–	–	–
L_GROWTH	–0.3236	–0.5219	0.3896	1	–	–	–	–
INFLATION	–0.4653	–0.5756	0.2843	0.1865	1	–	–	–
FDI_CA	0.5196	0.5412	–0.5089	–0.5092	–0.6020	1	–	–
FDI_ABC	0.4326	0.6819	–0.2436	–0.3165	–0.3629	0.3942	1	–
CRISIS	–0.3852	–0.0175	0.4290	0.3840	0.4209	–0.5696	0.1765	1

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Table 14.

Correlation Matrix of China Sample.

	L_INNOV	INNOV_EFF	L_FDI	L_GROWTH	INFLATION	FDI_CA	FDI_ABC	CRISIS
L_INNOV	1	–	–	–	–	–	–	–
INNO_EFF	0.5776	1	–	–	–	–	–	–
L_FDI	0.6452	0.6697	1	–	–	–	–	–
L_GROWTH	0.5728	0.2650	0.4322	1	–	–	–	–
INFLATION	0.0169	0.3010	0.0407	0.1544	1	–	–	–
FDI_CA	0.3564	0.4233	0.3454	0.0901	0.8110	1	–	–
FDI_ABC	0.2699	0.5866	0.1022	0.6863	0.6046	0.5566	1	–
CRISIS	0.3854	0.1128	0.3610	0.3871	0.3026	0.3575	0.3552	1

Source: Author estimation.

Note: Table includes values of descriptive statistics and correlation matrix, and it comprises of its values based on sample data.

Results of Tobit Regression

Tables 15–26 summarise the results of the Tobit regression. To test this study’s hypotheses, H₁, H_2a, H_2b and H_2c, we regress FDI, FDI × ABC and FDI × CA on innovation and innovation efficiency while including economic growth, inflation and financial crisis as control variables. In probability values, * indicates significance level at 10%, ** indicates significance level at 5% and *** indicates significance level at 1%. Tables 15 and 16 show the results of Models 1 and 2 of the full sample of this study. In Model 1 of the full sample, H₁ (β = 0.707, p < .01) proves the significant positive relationship between FDI and innovation in the largest FDI recipient countries of the world, whereas controlling variable economic growth also has a significant positive impact on the innovation. In Model 2 of the full sample, H_2a (β = 0.272, p < .01) proves the significant positive relationship and states that FDI improves innovation efficiency. H_2b (β = 0.009, p < .01) proves the existence of a significant positive relationship between absorption capacity of FDI and innovation efficiency. H_2c (β = 0.011, p < .1) also confirms the significant positive relationship between complementary assets and innovation efficiency; however, the strength of the relationship is relatively weak and control variables are also insignificant. The results suggest that inward FDI in these countries is effectively utilised in R&D to drive national innovation ecosystems.

Table 15.

Model 1 Tobit Regression Results of Full Sample.

Dependent Variable: INNOV_EFF
Method: Tobit Regression
Full Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-statistics	Probablility
L_FDI	0.272	0.024	–11.181	0.000***
FDI_ABC	0.009	0.001	6.199	0.000***
FDI_CA	0.011	0.006	–1.824	0.068*
L_GROWTH	–0.003	0.036	–0.091	0.928
INFLATION	–0.001	0.001	–1.071	0.284
CRISIS	–0.023	0.042	–0.555	0.579
C	4.892	1.012	4.835	0.000***
Mean dependent variable	0.726	SD dependent variable		0.297
SE of regression	0.144	Akaike info criterion		–0.375
Sum squared residual	2.622	Schwarz criterion		–0.203
Log likelihood	33.345	Hannan–Quinn criterion		–0.306
Average log likelihood	0.247

Source: Author estimation.

Note: *means significant at 90% confidence interval

Table 16.

Model 2 Tobit Regression Results of Full Sample.

Dependent Variable: INNOV_EFF
Method: Tobit Regression
Full Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-statistics	Probablility
L_FDI	0.272	0.024 –11.181		0.000***
FDI_ABC	0.009	0.001 6.199		0.000***
FDI_CA	0.011	0.006 –1.824		0.068*
L_GROWTH	–0.003	0.036 –0.091		0.928
INFLATION	–0.001	0.001 –1.071		0.284
CRISIS	–0.023	0.042 –0.555		0.579
C	4.892	1.012 4.835		0.000***
Mean dependent variable	0.726	SD dependent variable		0.297
SE of regression	0.144	Akaike info criterion		–0.375
Sum squared residual	2.622	Schwarz criterion		–0.203
Log likelihood	33.345	Hannan–Quinn criterion		–0.306
Average log likelihood	0.247

Source: Author estimation.

Note: *means significant at 90% confidence interval and ***means significant at 99% confidence interval.

In order to perform the individual country-wise analysis, we regress Models 1 and 2, using each country’s data separately. Tables 17 and 18 show the results of Models 1 and 2 of the United States sample. In Table 17 Model 1, H₁ (β = 5.485, p < .01) proves the significant positive relationship between FDI and innovation in the United States. The global financial crisis has a negative impact on the innovation of the United States; however, economic growth and inflation are insignificant factors. In Table 18 Model 2, H_2a (β = 1.581, p < .1) proves the relationship between FDI and innovation efficiency. Whereas H_2b (β = 0.005, p > .1) and H_2c (β = 0.010, p > .1) provide evidence of the rejection of H_2b and H_2c because there is no significant relationship between absorption capacity of FDI and complementary assets and innovation in the United States. However, the financial crisis has a significant negative impact on the innovation efficiency of that country.

Table 17.

Model 1 Tobit Regression Results of United States Sample.

Dependent Variable: L_INNOV
Method: Tobit Regression
United States Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-statistics	Probability
L_FDI	5.485	1.798	3.050	0.002***
L_GROWTH	–0.416	0.646	–0.644	0.520
INFLATION	0.006	0.007	0.852	0.394
CRISIS	–0.103	0.036	–2.859	0.004***
C	–78.214	16.719	–4.678	0.000***
Mean dependent variable	12.968	SD dependent variable		0.346
SE of regression	0.056	Akaike info criterion		–2.742
Sum squared residual	0.065	Schwarz criterion		–2.454
Log likelihood	43.010	Hannan–Quinn criterion		–2.656
Average log likelihood	1.593

Source: Author estimation.

Note: ***means significant at 99% confidence interval.

Tables 19 and 20 show the results of Models 1 and 2 of the United Kingdom sample. In Table 18 Model 1, H₁ (β = 2.047, p < .01) proves the significant positive relationship between FDI and innovation in the United Kingdom. The control variables inflation has positive and economic growth and the financial crisis have a negative impact on the innovation in the United Kingdom. In Table 18 Model 2, H_2a (β = 0.470, p > .1) proves no relationship between FDI and innovation efficiency in the United Kingdom, hence rejecting the hypothesis. Whereas H_2b (β = 0.048, p < .01) supports the hypothesis that there is a positive significant relationship between absorption capacity of FDI and innovation efficiency in the United Kingdom. We reject H_2c (β = –0.007, p > .1); there is no relationship between complementary assets and innovation efficiency in the United Kingdom. However, control variables inflation has negative, and the crisis has a positive impact on the innovation efficiency of the country.

Table 18.

Model 2 Tobit Regression Results of United States Sample.

Dependent Variable: INNOV_EFF
Method: Tobit Regression
United States Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-statistics	Probability
L_FDI	1.581	0.907	1.744	0.081*
FDI_ABC	0.005	0.006	0.887	0.375
FDI_CA	0.010	0.010	0.976	0.329
L_GROWTH	–0.464	0.462	–1.003	0.316
INFLATION	0.012	0.008	1.471	0.141
CRISIS	–0.058	0.019	–3.074	0.002***
C	–17.956	14.265	–1.259	0.208
Mean dependent variable	0.783	SD dependent variable		0.165
SE of regression	0.027	Akaike info criterion		–3.436
Sum squared residual	0.014	Schwarz criterion		–3.052
Log likelihood	54.381	Hannan–Quinn criterion		–3.321
Average log likelihood	2.014

Source: Author estimation.

Note: *means significant at 90% confidence interval and ***means significant at 99% confidence interval.

Table 19.

Model 1 Tobit Regression Results of United Kingdom Sample.

Dependent Variable: L_INNOV
Method: Tobit Regression
United Kingdom Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-statistics	Probability
L_FDI	2.047	0.692	2.959	0.003***
L_GROWTH	–0.356	0.079	–4.533	0.000***
INFLATION	0.008	0.003	2.532	0.011**
CRISIS	–0.223	0.035	–6.425	0.000***
C	–14.835	11.666	–1.272	0.204
Mean dependent variable	10.980	SD dependent variable		0.139
SE of regression	0.067	Akaike info criterion		–2.368
Sum squared residual	0.095	Schwarz criterion		–2.080
Log likelihood	37.964	Hannan–Quinn criterion		–2.282
Average log likelihood	1.406

Source: Author estimation.

Note: **means significant at 95% confidence interval and ***means significant at 99% confidence interval.

Table 20.

Model 2 Tobit Regression Results of United Kingdom Sample.

Dependent Variable: INNOV_EFF
Method: Tobit Regression
United Kingdom Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-statistics	Probability
L_FDI	0.470	0.507	0.926	0.354
FDI_ABC	0.048	0.007	7.437	0.000***
FDI_CA	–0.007	0.006	–1.107	0.268
L_GROWTH	–0.021	0.047	–0.439	0.661
INFLATION	–0.006	0.002	–3.483	0.000***
CRISIS	0.054	0.020	2.676	0.007***
C	–13.797	8.903	–1.550	0.121
Mean dependent variable	0.851	SD dependent variable		0.126
SE of regression	0.033	Akaike info criterion		–2.972
Sum squared residual	0.021	Schwarz criterion		–2.588
Log likelihood	48.125	Hannan–Quinn criterion		–2.858
Average log likelihood	1.782

Source: Author estimation.

Note: ***means significant at 99% confidence interval.

Tables 21 and 22 show the results of Models 1 and 2 of the Netherland sample. In Table 21, Model 1, H₁ (β = 1.605, p < .05) proves the significant positive relationship between FDI and innovation in Netherland. The control variables inflation, economic growth and financial crisis have no impact on innovation in the Netherland. Table 22 Model 2, H_2a (β = 1.344, p < .1) proves that FDI and innovation efficiency has a positive relationship in the Netherland. The H_2b (β = 0.060, p < .01) supports to accept the hypothesis that there is a significant positive relationship between absorption capacity of FDI and innovation efficiency in Netherland. Whereas H_2c (β = –0.036, p < .01), provides evidence of a significant negative relationship between complimentary assets and innovation efficiency in the Netherlands. Among the control variables, only inflation has significant and negative relation with innovation efficiency in Netherland.

Table 21.

Model 1 Tobit Regression Results of Netherland Sample.

Dependent Variable: L_INNOV
Method: Tobit Regression
Netherland Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-statistics	Probability
L_FDI	1.605	0.774	2.073	0.038**
L_GROWTH	–0.079	0.170	–0.466	0.641
INFLATION	–0.004	0.006	–0.676	0.499
CRISIS	0.095	0.065	1.470	0.142
C	–15.363	12.148	–1.265	0.206
Mean dependent variable	7.700	SD dependent variable		0.127
SE of regression	0.103	Akaike info criterion		–1.510
Sum squared residual	0.224	Schwarz criterion		–1.222
Log likelihood	26.380	Hannan–Quinn criterion		–1.424
Average log likelihood	0.977

Source: Author estimation.

Note: **means significant at 95% confidence interval.

Table 22.

Model 2 Tobit Regression Results of Netherland Sample.

Dependent Variable: INNOV_EFF
Method: Tobit Regression
Netherland Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-Statistics	Probability
L_FDI	1.344	0.779	1.724	0.084*
FDI_ABC	0.060	0.007	8.114	0.000***
FDI_CA	–0.036	0.011	–3.165	0.001***
L_GROWTH	–0.081	0.087	–0.935	0.350
INFLATION	–0.019	0.006	–3.148	0.001***
CRISIS	–0.048	0.044	–1.093	0.274
C	–21.721	11.089	–1.959	0.050*
Mean dependent variable	0.879	SD dependent variable		0.094
SE of regression	0.047	Akaike info criterion		–1.879
Sum squared residual	0.043	Schwarz criterion		–1.495
Log likelihood	33.368	Hannan–Quinn criterion		–1.765
Average log likelihood	1.236

Source: Author estimation.

Note: *means significant at 90% confidence interval and ***means significant at 99% confidence interval.

Tables 23 and 24 describe the results of Models 1 and 2 of the Ireland sample. In Table 23 Model 1, H₁ (β = –2.364, p < .1) proves the significant negative relationship between FDI and innovation in Ireland. Economic growth has a positive impact, whereas inflation has a negative impact on the innovation in Ireland. In Table 24 Model 2, H_2a (β = 0.257, p > .1), H_2b (β = 0.024, p > .1) and H_2c (β = 0.013, p > .1) provide the evidence to reject the hypotheses, meaning that FDI, the absorption capacity of FDI and complementary assets have no relationship with innovation efficiency in Ireland. Whereas the control variables, economic growth, inflation and crisis, also have no impact on the innovation efficiency in Ireland.

Finally, Tables 25 and 26 summarise the results of Models 1 and 2 of the China sample. In Table 25, Model 1, H₁ (β = 8.074, p < .01) proves the significant positive relationship between FDI and innovation in China and the strong relationship. Growth has a positive impact, and inflation has a negative impact on innovation in China. In Table 26 Model 2, H_2a (β = –0.713, p > .1) provides evidence of the rejection of the hypothesis, meaning that FDI has no relationship with innovation efficiency in China. The H_2b (β = 0.012, p < .01) supports the hypothesis that there is a significant positive relationship between the absorption capacity of FDI and innovation efficiency in China. Whereas H_2c (β = –0.053, p < .01) provides evidence of a significant negative relationship between complementary assets and innovation efficiency in China. The control variables growth and crisis have a negative impact and inflation has a positive impact on the innovation efficiency in China sample.

Table 23.

Model 1 Tobit Regression Results of Ireland Sample.

Dependent Variable: L_INNOV
Method: Tobit Regression
Ireland Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-statistics	Probability
L_FDI	–2.364	1.226	–1.928	0.053*
L_GROWTH	0.994	0.396	2.511	0.012**
INFLATION	–0.025	0.012	–2.042	0.041**
CRISIS	0.032	0.097	0.329	0.742
C	19.275	12.330	1.563	0.118
Mean dependent variable	8.514	SD dependent variable		0.268
SE of regression	0.172	Akaike info criterion		–0.495
Sum squared residual	0.618	Schwarz criterion		–0.207
Log likelihood	12.680	Hannan–Quinn criterion		–0.409
Average log likelihood	0.470

Source: Author estimation.

Note: *means significant at 90% confidence interval and **means significant at 95% confidence interval.

Table 24.

Model 2 Tobit Regression Results of Ireland Sample.

Dependent Variable: INNOV_EFF
Method: Tobit Regression
Ireland Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-statistics	Probability
L_FDI	0.257	1.845	0.139	0.889
FDI_CA	0.024	0.022	1.103	0.270
FDI_ABC	0.013	0.008	1.532	0.126
L_GROWTH	0.028	0.452	0.061	0.951
INFLATION	–0.010	0.023	–0.412	0.680
CRISIS	0.167	0.127	1.313	0.189
C	–5.446	21.092	–0.258	0.796
Mean dependent variable	0.847	SD dependent variable		0.218
SE of regression	0.152	Akaike info criterion		0.269
Sum squared residual	0.440	Schwarz criterion		0.653
Log likelihood	4.374	Hannan–Quinn criterion		0.383
Average log likelihood	0.162

Source: Author estimation.

Table 25.

Model 1 Tobit Regression Results of China Sample.

Dependent Variable: L_INNOV
Method: Tobit Regression
China Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-statistics	Probability
L_FDI	8.074	1.953	4.133	0.000***
L_GROWTH	0.999	0.135	7.397	0.000***
INFLATION	–0.022	0.008	–2.964	0.003***
CRISIS	–0.071	0.129	–0.552	0.581
C	–178.088	38.616	–4.612	0.000***
Mean dependent variable	13.193	SD dependent variable		1.202
SE of regression	0.239	Akaike info criterion		0.168
Sum squared residual	1.199	Schwarz criterion		0.456
Log likelihood	3.733	Hannan–Quinn criterion		0.254
Average log likelihood	0.138

Source: Author estimation.

Note: *** means significant at 99% confidence interval.

Table 26.

Model 2 Tobit Regression Results of China Sample.

Dependent Variable: INNOV_EFF
Method: Tobit Regression
China Sample: 1990–2016
Independent Variables	Coefficient	Standard Error	Z-statistics	Probability
L_FDI	–0.713	1.007	–0.708	0.479
FDI_ABC	0.012	0.002	5.592	0.000***
FDI_CA	–0.053	0.011	–4.629	0.000***
L_GROWTH	–0.162	0.096	–1.690	0.091*
INFLATION	0.010	0.002	4.219	0.000***
CRISIS	–0.093	0.025	–3.759	0.000***
C	20.454	19.774	1.034	0.301
Mean dependent variable	0.269	SD dependent variable		0.280
SE of regression	0.037	Akaike info criterion		–3.163
Sum squared residual	0.026	Schwarz criterion		–2.779
Log likelihood	50.695	Hannan–Quinn criterion		–3.048
Average log likelihood	1.878

Source: Author estimation.

Note: * means significant at 90% confidence interval and *** means significant at 99% confidence interval.

Conclusion and Policy Implications

This study endeavours to explore the relationship between FDI and necessary innovation factors on the innovation efficiency of the national innovation systems in the largest FDI recipient countries of the world. The findings of this study suggest that considerable investments in R&D, a more significant number of technicians in R&D, more spending on tertiary education by governments and quality labour force are associated with the increasing national innovation capacity of the whole sample, that is, the United States, the United Kingdom, China, Netherland and Ireland. The inward FDI has played a significant role in promoting the national innovation systems and building the capacity for innovation in these countries throughout 1990–2016. The significant positive impact of these innovation drivers is because these countries spend more than average on R&D activities. According to the Global Innovation Index 2016 rankings, the United Kingdom and the United States are in the 3rd and 4th positions, respectively, after Switzerland and Sweden among the top innovative countries globally (WIPO, 2016).

In the United States, the United Kingdom, Netherland and China, the inward FDI is positively associated with innovation. However, the inward FDI is only associated with the innovation efficiency of the United States and Netherland. The findings suggest that FDI can contribute significantly to the overall innovation efficiency of the top FDI recipient countries of the world and thus may lead to an increase the innovation capacity. The strength of this significant positive effect depends on the absorption capacity and the presence of complementary assets that are necessary for innovation in the host country. FDI is insignificant for increasing the innovation efficiency of the United Kingdom, China and Ireland. Figure 3 shows the trend of innovation efficiency of the country-wise sample, that is, the United Kingdom, the United States, China, Ireland and Netherland, from 1990 to 2016. We can see that China has a tremendously increasing trend of innovation efficiency throughout the sample period. At the same time, other countries have some downfalls in innovation efficiency at some points.

Figure 3.

Source: The author.

The selected sample period of this study includes the era of increasing R&D globalisation followed by the financial crisis; thus, the global financial crisis has a negative impact on the inward FDI and also on the innovation efficiency of the United States, the United Kingdom and China (H. Hill & Jongwanich, 2009). This increased innovative capability of the country and innovation efficiency contributes further to the fastest economic growth of these countries. In the United Kingdom, the Netherland and China, the absorption capacity of FDI positively impacts innovation efficiency; these findings are consistent with (Fu, 2008). This evidence provides an intense support for the role of the absorption capacity of a country in the adaptation for knowledge spillovers from FDI. The evidence may also suggest that the number of technicians in R&D, better industry value-added services and lower cost of new business start-ups will complement the advanced technology exemplified in FDI, which will ultimately facilitate the innovation process of a country and lead to greater innovation capacity of the nation.

Our findings suggest valuable insights for the policymakers that the lack of absorption capacity of FDI of innovation is the main factor that may prevent inward FDI boost the national innovation systems. The lack of absorption capacity of FDI and complementary assets may become a bottleneck that hinders the local economy towards the higher end of the value chain and the knowledge-based economy. The attention of placing more focus on the R&D investments may be the one crucial factor for the innovation, but government spending on tertiary education and technicians in R&D may also play a significant role in developing the innovation economy. The globalisation of R&D may allow developing countries to catch up on the technology edge. Multinational firms can become an embedded driver for the knowledge-based economy; however, these firms have been very different in technology, incentive systems and operations. Attention must be paid to enhancing the local absorption capacity and the necessary complementary assets to assimilate knowledge and technology spillovers from inward FDI effectively. Therefore, a role remains with the government policies to strengthen these conditions for a successful techno goal achievement from FDI.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author received no financial support for the research, authorship and/or publication of this article.

References

Aitken

B. J.

, & Harrison

A. E.

(1999). Do domestic firms benefit from direct foreign investment? Evidence from Venezuela. American Economic Review, 89(3), 605–618.

Aldieri

, Brahmi

, Chen

, & Vinci

C. P.

(2021). Knowledge spillovers and technical efficiency for cleaner production: An economic analysis from agriculture innovation. Journal of Cleaner Production, 320(4), 128830.

Alin

(2010). Multicollinearity. Wiley Interdisciplinary Reviews: Computational Statistics, 2(3), 370–374.

Antonietti

, Bronzini

, & Cainelli

(2015). Inward greenfield FDI and innovation. Economia e Politica Industriale, 42(1), 93–116.

Athreye

, & Cantwell

(2007). Creating competition? Globalisation and the emergence of new technology producers. Research Policy, 36(2), 209–226.

Bank

T. W.

(2018). Globally, foreign direct investment decreased in 2016. Retrieved from https://datatopics.worldbank.org/world-development-indicators/stories/global-fdi-inflows-decreased-in–2016.html

Barrell

, & Pain

(1997). Foreign direct investment, technological change, and economic growth within Europe. The Economic Journal, 107(445), 1770–1786.

Blomström

, & Kokko

(2003). Human capital and inward FDI (Vol. 3762). Centre for Economic Policy Research.

Blomström

, Kokko

, & Zejan

(1994). Host country competition, labor skills, and technology transfer by multinationals. Review of World Economics, 130(3), 521–533.

10.

Charnes

, Cooper

W. W.

, & Rhodes

(1978). Measuring the efficiency of decision making units. European Journal of Operational Research, 2(6), 429–444.

11.

Cheung

K.-Y.

, & Ping

(2004). Spillover effects of FDI on innovation in China: Evidence from the provincial data. China Economic Review, 15(1), 25–44.

12.

CIA. (2016). Country comparison: Stock of direct foreign investment—At home. The World Factbook. Retrieved from https://web.archive.org/web/20201224025625/https://www.cia.gov/library/publications/the-world-factbook/rankorder/2198rank.html

13.

Cohen

W. M.

, & Levinthal

D. A.

(1989). Innovation and learning: The two faces of R & D. The Economic Journal, 99(397), 569–596.

14.

Cohen

W. M.

, & Levinthal

D. A.

(1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 128–152.

15.

Cooper

W. W.

, Seiford

L. M.

, & Zhu

(2011). Handbook on data envelopment analysis. Springer.

16.

Cosh

, Fu

, & Hughes

(2007). Management characteristics, managerial ownership and innovative efficiency in high-technology industry. Oxford University.

17.

Falki

(2009). Impact of foreign direct investment on economic growth in Pakistan. International Review of Business Research Papers, 5(5), 110–120.

18.

Fritsch

, & Slavtchev

(2006). Measuring the efficiency of regional innovation systems: An empirical assessment. Retrieved from https://econpapers.repec.org/paper/zbwtufwps/200608.htm

19.

(2004). Exports, foreign direct investment and economic development in China. Springer.

20.

(2008). Foreign direct investment, absorptive capacity and regional innovation capabilities: Evidence from China. Oxford Development Studies, 36(1), 89–110.

21.

, Cosh

, Yang

, & Hughes

(2006). World innovation frontier and the EU-US innovation gap [Paper presentation]. The 9th North America Workshop of Efficiency and Productivity Analysis, Stern Business School, New York University, New York.

22.

Geroski

P. A.

(1990). Innovation, technological opportunity, and market structure. Oxford Economic Papers, 42(3), 586–602.

23.

Girma

(2005). Absorptive capacity and productivity spillovers from FDI: A threshold regression analysis. Oxford bulletin of Economics and Statistics, 67(3), 281–306.

24.

Girma

, Gong

, & Görg

(2006). Can you teach old dragons new tricks? FDI and innovation activity in Chinese state-owned enterprises (DP No. 2267). IZA.

25.

Görg

, & Greenaway

(2004). Much ado about nothing? Do domestic firms really benefit from foreign direct investment? The World Bank Research Observer, 19(2), 171–197.

26.

GOV.UK. (2018). Attracting foreign direct investment: UK government success. Retrieved from https://www.gov.uk/government/speeches/attracting-foreign-direct-investment-uk-government-success

27.

Guan

, & Chen

(2010). Measuring the innovation production process: A cross-region empirical study of China’s high-tech innovations. Technovation, 30(5–6), 348–358.

28.

Guo

, Li

, & Lin

(2021). The impact of economic growth, FDI, and innovation on environmental efficiency of the logistics industry in provinces along the belt and road in China: An empirical study based on the panel Tobit model. Science Progress, 104(2), 00368504211018054.

29.

Helinska-Hughes

, & Hughes

(2003). Joining the competition: Central and Eastern European challenge to established FDI destinations. Edward Elgar.

30.

Hill

E. J.

, Ferris

, & Märtinson

(2003). Does it matter where you work? A comparison of how three work venues (traditional office, virtual office, and home office) influence aspects of work and personal/family life. Journal of Vocational Behavior, 63(2), 220–241.

31.

Hill

, & Jongwanich

(2009). Outward foreign direct investment and the financial crisis in developing East Asia. Asian Development Review, 26(2), 1.

32.

A. G.

, & Jefferson

G. H.

(2002). FDI impact and spillover: Evidence from China’s electronic and textile industries. The World Economy, 25(8), 1063–1076.

33.

Hughes

, & Morton

M. S. S.

(2006). The transforming power of complementary assets. MIT Sloan Management Review, 47(4), 50.

34.

Khachoo

A. Q.

, & Khan

M. I.

(2012). Determinants of FDI inflows to developing countries: A panel data analysis (MPRA Paper No. 37278). Retrieved from https://mpra.ub.uni-muenchen.de/37278/1/MPRA_paper_37278.pdf

35.

Khordagui

, & Saleh

(2016). Absorptive capacity factors that mediate foreign direct investment spillovers: A sector-level analysis from emerging economies. International Journal of Business and Globalisation, 16(2), 188–201.

36.

Kiriyama

(2012). Trade and innovation: Synthesis report. Retrieved from https://www.oecd-ilibrary.org/trade/trade-and-innovation_5k9gwprtbtxn-en

37.

Kuemmerle

(1999). The drivers of foreign direct investment into research and development: An empirical investigation. Journal of International Business Studies, 30(1), 1–24.

38.

Lall

(1992). Technological capabilities and industrialization. World Development, 20(2), 165–186.

39.

Levin

S. A.

(2013). Encyclopedia of biodiversity. Elsevier.

40.

Mansfield

E. R.

, & Helms

B. P.

(1982). Detecting multicollinearity. The American Statistician, 36(3a), 158–160.

41.

Ngo

, & Tsui

K. W. H.

(2020). A data-driven approach for estimating airport efficiency under endogeneity: An application to New Zealand airports. Research in Transportation Business & Management, 34, 100412.

42.

Pearce

R. D.

(1999). Decentralised R&D and strategic competitiveness: Globalised approaches to generation and use of technology in multinational enterprises (MNEs). Research Policy, 28(2–3), 157–178.

43.

Potterie

B. V. P. D. L.

, & Lichtenberg

(2001). Does foreign direct investment transfer technology across borders? Review of Economics and Statistics, 83(3), 490–497.

44.

SelectUSA. (2018). Why FDI, learn about the role of FDI in the U.S. economy. Retrieved from https://www.selectusa.gov/why-fdi

45.

Simar

, & Wilson

P. W.

(2007). Estimation and inference in two-stage, semi-parametric models of production processes. Journal of Econometrics, 136(1), 31–64.

46.

Teece

D. J.

(1986). Profiting from technological innovation: Implications for integration, collaboration, licensing and public policy. Research Policy, 15(6), 285–305.

47.

The Irish Times. (2017). Ireland named best country for high-value FDI for sixth year in a row. Retrieved from https://www.irishtimes.com/business/economy/ireland-named-best-country-for-high-value-fdi-for-sixth-year-in-a-row–1.3204594

48.

Tobin

(1958). Estimation of relationships for limited dependent variables. Econometrica: Journal of the Econometric Society, 26(1), 24–36.

49.

WIPO. (2016). Global innovation index 2016. Retrieved from https://www.wipo.int/edocs/pubdocs/en/wipo_pub_gii_2016.pdf

50.

Zaman

, Shah

I. A.

, Khan

M. M.

, & Ahmad

(2012). Macroeconomic factors determining FDI impact on Pakistan’s growth. South Asian Journal of Global Business Research, 1(1), 6.

51.

Zhang

, Ouyang

, Ballesteros-Pérez

, Li

, Philbin

S. P.

, Li

, & Skitmore

(2021). Understanding the impact of environmental regulations on green technology innovation efficiency in the construction industry. Sustainable Cities and Society, 65, 102647.

52.

Zhang

K. H.

(1999). How does FDI interact with economic growth in a large developing country? The case of China. Economic Systems, 23(4), 291–304.