Corruption and Foreign Direct Investment: A Conceptual Review

This paper carefully reviews the empirical literature on the relationship of corruption and FDI inflow from 2000 to 2019 with a focus on more recent studies. Though corruption is not a new phenomenon, its pervasiveness in the contemporary era, as well as its role in modern economy, has continued to attract the interest of scholars and development institutions in recent times. Theoretically, there are two conflicting views on corruption. The “grabbing hand” hypothesis maintains that corruption raises the cost of transactions, risks and uncertainties which inhibits FDI inflow, while the “helping hand” hypothesis believes that corruption eases bureaucratic as well as regulatory hurdles and greases the wheels of commerce, thereby facilitating the inflow of FDI. Although the methodologies employed and the context of the various empirical investigations reviewed differ considerably, the paper has largely shown that there is no consensus on the nature of corruption-FDI nexus. While the outcomes of the studies are dominated by the validation of the “grabbing hand” hypothesis, some have supported the grease the view of the wheel, whereas few concluded on insignificant relationships between corruption and FDI inflow. These contradictory outcomes from various studies have prompted some scholars to examine the connection between some specific aspects of FDI with corruption in a more dynamic framework. Accordingly, the areas of concern from the review have been pointed out in the discussion of the findings. The paper has also documented the need for further studies in this area with relevant suggestions.


INTRODUCTION
Foreign direct investment (FDI) has continued to play the foremost role in global wealth creation and has remained the largest source of external financing for developing countries (UNCTAD, 2018). The low level of savings capacity associated with capital deficient developing economies has compelled them to look up to FDI inflow as a viable means of augmenting domestic capital to boost productive resources and support their developmental aspiration (Asiedu, 2002). The growing rate of industrialization in the newly emerging developing economies has been attributed to the trend of rising inflow of FDI to these countries over the last few decades. FDI is a cross country investment involving an exercise of control or significant influence by a resident of a country in an enterprise situated in another economy (World Bank, 2018). Basically, it involves the transfer of capital by an investor across the border to the host economy or the shifting of the assets of an enterprise or aspect of its operations involving production, distribution or sales to the host country. FDI play a significant role in the growth process of the host country through transfer of technology and managerial skills as well as stimulating domestic competition, generation of employment and facilitation of access to global market. Besides bridging the savings-investment gap, FDI also contributes to the inflow of foreign exchange and tax revenue generation (Quazi, Vemuri & Soliman, 2014).
The cyclical global financial and economic crisis which have continue to entangle economies across the world has provoked decrease in the flow of foreign aid to developing countries, leading to a new vigor for exploring alternative means of supporting sustainable growth. Consequently, the quest towards realizing the actual and potential benefits of FDI has prompted most developing economies to make conscious efforts to entice FDI inflow through policy initiatives around their macroeconomic environment.
Accordingly, there has been relative boom in FDI inflow in the past few decades with global FDI flow rising from USD204 billion in 1990 to USD1,429 in 2017(UNCTAD, 2018. Although, the inflow of FDI to developing economies has also risen considerably in response to the global FDI surge, the inflow of FDI to different countries and regions differ substantially due to varying capacity to fascinate FDI. Among the key elements identified as obstacle to FDI inflow, is the prevalent of corruption in the host country along with other institutional and macroeconomic factors. The increasing consideration being accorded to efficiency factors by Multinational Corporations (MNCs) has culminated in the growing significance of institutional features in encouraging FDI inflow as well as the attention of scholars in this direction (Dunning, 2003). Furthermore, OECD (2002) even went to the extent of arguing that when there is strong and conducive institutional environment, there is no need for special incentives to attract inflow of FDI. Though, there is a growing interest of scholars on the link between institutional variables and FDI inflow in recent times, the FDI literature is inherently dominated by studies on the relationship of macroeconomic related factors and FDI with less attention accorded to institutional variables (Tun, Azman-Saini & Law, 2012). Thus, the focus of this paper is a deliberate attempt to contribute in filling this gap. Moreover, the review of empirical investigations in respect of the connection between key institutional variables such as corruption and FDI will provide insights on the key issues around FDI inflow for policy makers, researchers as well as other relevant stakeholders.
The aim of this study is essentially to complement prior review studies on association of traditional macroeconomic variables with FDI by mainly focusing on review of recent empirical studies on the link between corruption and FDI inflow from 2000 to 2019 as well as indicate other directions for further exploration. This paper largely departs from the prevailing literature reviews that concentrated on the connection of economic growth with FDI such as the work of Wan (2010), Ozturk (2007) and Almfraji and Almsafir (2014). Hence, this paper dwelt on more recent empirical investigations in respect of the nexus between corruption and FDI inflow. The motivation behind this paper is the growing importance of institutional environment in boosting FDI inflow and the role of corruption in contemporary economy as well as the increasing interest of scholars in this regard.
The remaining part of this paper is structured thus; Section 2.0 reviews recent empirical studies on corruption and FDI inflow, whereas Section 3.0 explains the emerging trend in the study of corruption-FDI nexus. Section 4.0 provides the summary of the reviews and conclusion of the paper with relevant suggestions on the direction for further research.

CORRUPTION AND FOREIGN DIRECT INVESTMENT
Predominantly, corruption is defined as 'the misuse of public office for private benefit' (Kaufman, Kraay & Mastruzzi, 2005). Although, corruption as a phenomenon predates the beginning of modern civilization, it has however become widespread globally in modern era especially in developing economies. Issues such as entrusting public officers with huge discretionary power for dispensing favor, bureaucratic bottlenecks, poor remunerations for public officials and weak judicial system are some of the factors identified as lubricants for corrupt practices (Castro & Nunes, 2013). Though, there is no consensus on the nature of the relationship of corruption with FDI, it is widely believed that corruption retards FDI inflow, hinders domestic investment, inhibits entrepreneurship, dampens public infrastructure quality, affects public expenditure and undermines economic growth.
Theoretically, there are two main conflicting views on corruption trending in the literature. The dominant view is that corruption serves as grabbing hand by raising the cost of operations for foreign investors (Bardhan, 1997). These extra costs are associated with payments of commissions or bribes to local officials, bureaucrats and politicians for contracts, permits, licenses, provision of utility services, assessment of tax and security services. All these tend to "sand the wheels of commerce" by increasing operational costs and ruining investment profitability. In contrast, the alternative view is that corruption serves as helping hand by "greasing the wheels of commerce" (Leff, 1964). This arises from the use of corrupt practices to circumvent bureaucratic hurdles in a weak institutional environment, supplement low wages of public officials which enables the tax burden to be kept low, avoid excessive regulations and facilitate the acceleration of decision making process which stimulates efficiency and economic growth. Sekuła (2016), Subasat and Bellos (2013), Fung and Gracia-Herrero (2012) and Helmy (2013) have submitted that corruption is a helping hand which facilitates FDI inflow. On their part, Anyanwu (2012), Gobinder and Haider (2014), Gani and Al-Abri (2013), Saidi, Ochi and Ghandri (2013), Okafor (2015), Bannaga et al. (2013), Jadhav (2012, Hoa and Lin (2016) and Cleeve (2012) have concluded that the relationship between corruption and FDI inflow is insignificant.
In terms of methodological approach, virtually all the studies reviewed employed panel data approach with the exception of Kasasbeh, Mdanat and Khasawneh (2018), Abala (2014) and Osabutey and Okoro (2015) who used time series data of Jordan, Kenya and Nigeria, respectively. In addition, most of the studies utilized Ordinary Least Squares (OLS) technique, while some deployed Generalized Method of Moments (GMM) and other econometrics methods. For example, Osabutey and Okoro (2015), Nnadi and Soobaroyen (2015), De Beule and Duanmu (2012) Among the studies that employed GMM dynamic technique in their investigations of the link between corruption and FDI inflow are Aziz and Mishra (2015), Dauti (2015), Morrissey and Udomkerdmongkol (2012) Amongst the contemporary works which show that corruption impedes FDI inflow is the work of Amarandei (2013). The researcher used OLS regression to examine the role of corruption on FDI inflow in 10 selected European countries during the period 2000 -2012 and reported that corruption affects inflow of FDI negatively in the selected countries. On their part, Júlio, Pinheiro-Alves and Tavares (2013) employed AGM and Poisson Pseudo-Maximum Likelihood (PPML) techniques to investigate the impact of institutional, economic and geographic factors on FDI inflow in European countries to validate the grabbing hand hypothesis on corruption. Though, the two studies focused on European countries, they failed to extend their studies to the inflow of FDI into the specific sectors of the economy such as manufacturing, services and extractive industries.
In another study on 15 European Union (EU) member countries, Dauti (2015) utilized AGM and GMM models in empirical analysis of the relationship of institutional variables with FDI inflow. The author concluded that corruption discourages FDI inflow and control of corruption has significant positive influence on inflow of FDI. Although, the study applied a dynamic model, it did not consider different forms of FDI to the EU countries. Nevertheless, the exploration of Fung and Gracia-Herrero (2012) using AGM on the locational factors attracting Chinese and Indian FDI to various countries between 1991 and 2008 came out with diverging outcomes. The analysis reveals that FDI from India is enticed by control of corruption, whereas it is insignificant for FDI from China. In the same vein, the investigation of De Beule and Duanmu (2012) with Fixed Effects Model (FEM) on FDI from India and China corroborated the findings of Fung and Gracia-Herrero, though both Indian and Chinese FDI in the mining sector were found to be attracted to countries with weak corruption control. Although, both studies concentrated on Indian and Chinese FDI, Fung and Gracia-Herrero (2012) failed to examine FDI inflow to the specific sectors of the host economies, whereas De Beule and Duanmu (2012) have overcome this limitations by considering FDI inflow to the mining sector. Nonetheless, the findings indicate that FDI into the extractive industries is largely resource seeking and therefore less sensitive to institutional environment.
Al-Khouri and Khalik (2013)  Similarly, studies by Aziz and Mishra (2015) which deployed GMM estimation technique to investigate the locational determinants of FDI involving institutional, economic and policy variables in 16 Arab countries from 1984 to 2012 supported the "sand the wheels" view on corruption. Despite the varying techniques applied in their analysis by the two studies, the outcomes basically show that institutional environment matters for FDI inflow in the Arab region. In addition, a recent study by Kasasbeh, Mdanat and Khasawneh (2018) employed VAR analytical framework in examining the relationships of institutional, financial and economic factors with FDI in Jordan between 1990 and 2015 to demonstrate the existence of positive significant relationship between control of corruption and inflow of FDI. Although, a dynamic model was engaged in the study, the link of corruption with different forms of FDI was not explored.
In contrast, Bellos and Subasat (2012) used REM in the study of institutional factors and FDI inflow in 14 transition economies from 1990 to 2003 and found that corruption encourages inflow of FDI. Also, Khan and Akbar (2013)  Among the studies that focused on African region is the work of Nnadi and Soobaroyen (2015) which used OLS regression model to investigate the role of corruption perception index and other economic and institutional variables in 34 African economies spanning the period 1990 -2014.The authors affirmed the grabbing hand view on corruption and importance of corruption control in encouraging FDI inflow. On their part, Osabutey and Okoro (2015) utilized OLS regression to explore the effect of corruption control along with other political factors on inflow of FDI in the Nigerian telecommunications industry between 2002 and 2011, while Abala (2014) employed 2SLS estimation technique to examine the key drivers of FDI inflow and economic growth in Kenya from 1970 to 2010. They all concluded that corruption undermines FDI inflow in the countries. Although, the studies did not employed dynamic model in their respective context, they were able to highlight the significance of corruption control in attracting FDI to Africa. Besides, the investigation of Ferreira and Ferreira (2016) (2018) recently reinvestigated the effect of institutional quality in 132 host developing economies on FDI inflow from China with system GMM approach. The studies concluded that corruption facilitates inflow of FDI. It is interesting that apart from Subasat and Bellos who used GLS method, the other authors employed GMM technique and reported same verdict, though the context of their investigations varies. However, Anyanwu (2012) who employed OLS and GMM techniques to explore the effect of corruption control on FDI in 53 African host countries as well as Okafor (2015) who examined the determinants of FDI from United States (US) to 23 SSA economies between 1996 and 2010 submitted that corruption control has no significant influence on FDI in the African countries. This is an indication that US FDI inflow to SSA is not driven by efficiency motive, but other strategic considerations.
Several other studies have been conducted on developing economies generally. For instance, Azam and Ahmad (2013) and Jeong (2014) employed multiple regression analysis in their investigation of FDI inflow in 33 selected developing countries from 1985 to 2011 and 34 sampled developing economies between 2002 and 2006, respectively; whereas Quazi (2014) utilized GLS estimation method to examine the effect of corruption on inflow of FDI in seven South Asia and Nine East Asian countries between 1995 and 2011. Though, the studies utilized various analytical approach, they failed to consider heterogeneity across the panels in terms of economic, cultural and socio-political variations.
In a related study, Mathur and Singh (2013) interrogated the relationship of corruption and democracy with FDI inflow using GLS technique in 29 emerging markets between 1980 and 2000 to support the sand the wheels view, while Castro and Nunes (2013) deployed GLS estimation method on 73 selected developing economies between the period of 1998 and 2008 to agree with Mathur and Singh that corruption discourage FDI inflow. Likewise, Hossain and Rahman (2017) recently used OLS to explore the relationship of six institutional variables with FDI inflow in 80 developing countries spanning the period 1998 -2014 and concluded that corruption is an impediment to inflow of FDI. However, the study of Saidi, Ochi and Ghandri (2013) on 20 developed and developing economies from 1998 to 2011 with FEM produced mixed results. While corruption control was found to have positive influence on FDI inflow in developed economies, it turn out to be insignificant in developing countries. Nevertheless, the studies have failed to consider disaggregation of FDI into various sectors with a view to explore whether there are variations in the outcome.
Furthermore, Morrissey and Udomkerdmongkol (2012) examined the impact of different institutional indicators on FDI inflow as well as domestic private investment in 46 developing economies from 1995 to 2009. The researchers applied dynamic GMM technique in their empirical analysis to report that control of corruption facilitates FDI inflow in the selected countries. Similarly, the analytical investigation of Hayakawa, Kimura and Lee (2013) on the connection between different components of financial as well as political risks with inward FDI in a panel of 89 developing countries covering the period 1985 -2007 utilized GMM technique to support the "sand the wheels" view on corruption. Additionally, Kurul and Yalta (2017) in a recent study reexamined the link between the various institutional variables and FDI inflow in 113 developing and emerging economies between 2002 and 2012 by employing system GMM estimation approach. The authors documented that corruption undermines inflow of FDI, thereby authenticating the grabbing hand hypothesis. Although, the researchers have all used GMM dynamic model to investigate different panels of developing economies, the outcomes are the same indicating that the sand the wheels view is real and institutional environment of the host countries is significant for inflow of FDI.
Within the framework of panel data analysis, Alemu (2012), Udenze (2014), Erdogan and Unver (2015), Jalil, Qureshi and Feridun (2016) and Hossain (2016) have also explored the corruption-FDI nexus using various econometric approaches. For instance, Alemu (2012) and Hossain (2016) employed REM and GLS techniques to examine 16 selected Asian economies between 1995 and 2009 and 48 countries from three regions of Asia, Africa and Latin America from 1998 to 2014, respectively. The outcomes were inherently similar. Alemu (2012) reported that an increase in the level of corruption triggered significant decrease in FDI inflow and even countries with high level of corruption as well as remarkable inflow of FDI could further enhance their FDI inflow substantially through reduction in the existing level of corruption in their countries, whereas Hossain (2016) concluded that corruption is a grabbing hand in the 48 countries covered in the study and a reduction in the level of corruption will boost investors' confidence and accelerate inflow of FDI. Nonetheless, studies by Udenze (2014) and Jalil, Qureshi and Feridun (2016) have established varying impact of corruption on FDI in different regional context. Jalil, Qureshi and Feridun (2016) used PMG dynamic heterogeneous panel to investigate 42 developing economies across Latin America, Africa and Asia between 1984 and 2012. They found that corruption impedes FDI inflow in Latin American countries, but exhibit positive relationship with FDI in the case of Africa and Asian countries. A related outcome was reported by Udenze (2014) who employed OLS to examine 73 low and middle income countries drawn from various regions across the globe. The author submitted that corruption retards FDI inflow in all the countries covered by the study, except in SSA where the outcome shows that corruption is an efficient grease facilitating FDI inflow.
On their part, Erdogan and Unver (2015) interrogated the role of corruption control among other determinants of FDI with FEM and GMM techniques in 88 selected countries spanning the period 1985 -2011. The authors reported that corruption restrains FDI inflow and control of corruption exert positive and significant influence on inflow of FDI. On the contrary, the study of Kolstad and Wiig (2013) which employed FEM on corruption and FDI inflow in the extractive sectors of 81 selected countries spanning the period 1996 -2009 reported that FDI is attracted to host countries that have less corruption control. However, Bannaga et al. (2013) concluded from a study of 18 Arab countries using GLS technique during the period 2000 -2009 that FDI inflow is insensitive to corruption control in the Arab economies.

EMERGING TREND
Arising from the contradictory outcomes from different studies using numerous methodologies in various contexts, some empirical works have employed different approach and frameworks by focusing on specific aspects of the relationships with a view to provide additional insights on corruption-FDI nexus (Luu et al., 2019). For example, Uhlenbruck et al. (2006) and Smarzynska and Wei (2000) employed regression analysis in exploring the role of FDI ownership structure as foreign investors entry adjustment strategy in corrupt environments. The study of Uhlenbruck et al. (2006) focused on 64 emerging economies, whereas the work of Smarzynska and Wei (2000) concentrated on former Soviet Union and Eastern Europe. They all found that investors are inclined to joint venture partnership with local partners than undertaking wholly owned investment in corrupt economies. However, this submission did not hold for firms from US who are inclined to having fully owned investments in corrupt locations.
In addition, Cuervo-Cazurra (2006) utilized AGM in the study of corruption distance with FDI data of 183 source countries and 106 host countries to submit that corruption in host country attracts FDI from economies with higher degree of corruption. On their part, Qian and Sandoval-Hernandez (2016) analyzed the relationship of corruption distance with bilateral FDI inflow using AGM with data of 45 selected developing economies during the period 1997 -2007. They endorsed the "grabbing hand" hypothesis, by submitting that corruption impair the inflow of FDI. However, the study of Godinez and Liu (2015) which investigated corruption distance and inflow of FDI to Latin America with REM reported that FDI from countries with low level of corruption is retarded by corruption, whereas FDI from countries with high degree of corruption is indifference to corruption distance. In a similar study on corruption distance, Lucke and Eichler (2016) scrutinized the impact of cultural and institutional factors using system GMM approach on FDI from 29 source countries to 65 host economies between 1995 and 2009 and concluded that investors are more disposed to investing in developed countries with higher degree of corruption than their home countries.
From another perspective, Hakkala et al. (2008) separated vertical from horizontal FDI in order to examine the association between corruption and different forms of FDI. They used OLS regression with Swedish firm-level data in their analysis and concluded that corruption deters inflow of horizontal FDI, but its impact on vertical FDI inflow is insignificant. Furthermore, Luu, Nguyen, Ho and Nam (2019) approached the study of corruption-FDI connection from different viewpoint by exploring the impact of corruption on FDI inflow generally as well as different modes of FDI flow involving Greenfield investment and cross border mergers and acquisitions (M&A). They utilized GMM dynamic technique on a panel of 131 countries from all over the world using datasets for the period 2003 -2015 and concluded that corruption retards FDI inflow generally. Though, investigations into different components of FDI inflow reveal that corruption impedes cross border M&A, but facilitates Greenfield investment. This implies that the two modes of FDI inflow are driven by different motives and considerations.

SUMMARY AND CONCLUSION
The studies reviewed in this paper on the relationship of corruption and FDI inflow can generally be classified into three groups on the basis of their reported outcomes. The mainstream group consist of studies that validated the "grabbing hand" hypothesis, whereas the second group relates to studies that affirmed the "helping hand" hypothesis on corruption and the third group comprise of few studies that documented insignificant connection between corruption and FDI inflow. The varying results from different empirical investigations may not be unconnected with dissimilar methodological approaches employed by the researchers in various context and periods. Although, the emerging trend in the framework of empirical investigation is adding value to the understanding of the link between corruption and FDI, what is obvious is that there is the need for additional empirical studies on the link between corruption and FDI inflow with adequate consideration to heterogeneity in terms of political, social and economic attributes. In addition, most of the studies employed panel data analysis, future studies could consider time series approach in their empirical investigation of corruption and FDI inflow in specific context. Although, studies such as Hayakawa, Kimura and Lee (2013), Aziz and Mishra (2015), Jalil, Qureshi and Feridun (2016), Kurul and Yalta (2017), and Kasasbeh, Mdanat and Khasawneh (2018) utilized dynamic models in their analysis, other dynamic approaches for time series and macro panel analyses could be explored as well. Furthermore, future investigations could also be extended to explore non-linear relationship and inflow of various forms of FDI into specific sectors of the economy as well as explore causal relationships between corruption and FDI inflow. All these could generate additional insights on corruption-FDI nexus leading to specific recommendations that would stimulate FDI inflow particularly to the capital deficient developing economies.