Do countries' endowments of non-renewable energy resources matter for FDI attraction? A panel data analysis of 125 countries over the period 1995–2012

Author’s Accepted Manuscript Do countries’ endowments of Non-Renewable Energy Resources matter for FDI attraction? A panel data analysis of 125 countries over the period 1995–2012 Aurora A.C. Teixeira, Rosa Forte, Susana Assunção www.elsevier.com/locate/inteco

PII: DOI: Reference:

S2110-7017(16)30043-9 http://dx.doi.org/10.1016/j.inteco.2016.12.002 INTECO112

To appear in: International Economics Cite this article as: Aurora A.C. Teixeira, Rosa Forte and Susana Assunção, Do countries’ endowments of Non-Renewable Energy Resources matter for FDI attraction? A panel data analysis of 125 countries over the period 1995–2012, International Economics, http://dx.doi.org/10.1016/j.inteco.2016.12.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Do countries’ endowments of Non-Renewable Energy Resources matter for FDI attraction? A panel data analysis of 125 countries over the period 1995-2012 Aurora A.C. Teixeiraa1, Rosa Forteb2, Susana Assunçãoc a

Research areas/interests: Human Capital; Economic Growth; Innovation; Bibliometrics

b

c

Research areas/interests: Foreign direct investment (FDI); International Trade; Multinationals

Faculdade de Economia do Porto

[email protected] [email protected] [email protected] *

Correspondence to: 4200 – 464 Porto. Tel.: +00351225571100.

Abstract Empirical studies on FDI location determinants have neglected the role of natural resources. Panel data estimations for 125 host countries over the period between 1995 and 2012 show that a country’s endowment of Non-Renewable Energy Resources (NRERs) matters for FDI attraction, when measured by the share of oil, coal and gas exports in total exports but not when measured by oil, coal and gas ‘proven reserves’. Thus, although to possess a vast amount of proven NRERs is not a sufficient condition for FDI inflows, countries with low export diversification, highly dependent on the exports of mineral fuel, tend to succeed in attracting FDI. This evidence supports the content that resource seeking FDI targets mainly economically feeble countries. Moreover, our results firmly indicate that regardless NRERs endowments, FDI attraction is fostered when countries make convincing efforts to open up their economies to international trade and devote resources to the enhancement of their human capital, control of corruption, and have more beneficial tax rates.

Keywords: FDI, Determinants of FDI; Non-Renewable Energy Resources; Panel data 1

Associate Professor with Habilitation (Agregação) at Faculdade de Economia do Porto (Universidade do Porto) and researcher at CEF.UP and OBEGEF. 2 Assistant Professor at Faculdade de Economia do Porto (Universidade do Porto) and researcher at CEF.UP. 1

JEL-Codes: F21, C19, O13

1. Introduction Foreign direct investment (FDI) is regarded as a driving force behind economic growth (Wang, 2009; Aziz and Mishra, 2016). Many governments see FDI as a way of dealing with stagnation and even the poverty trap (Brooks et al., 2010; Gohou and Soumaré, 2012).

The growing search for natural resources, in particular for non-renewable energy resources (NRERs), namely by economies that are growing rapidly, has led to an increase in commodity prices, resulting in a renewed interest by governments in exploiting energy resources and a redirection of FDI towards the mining sector (UNCTAD, 2007). Over the last twenty years, economic development policies have tended to neglect investment in the mining sector (UNCTAD, 2007; Betz et al., 2015). However, in an age when energy security is a global concern and countries, such as China, are attempting to take positions in mining companies around the world to ensure future supply and thereby continued economic growth (Moran, 2010; Henson and Yap, 2016), it is important to understand how far the endowment in NRERs, specifically ‘proven reserves’ (the economically extractable fraction of a resource using current technology – see Grafton et al., 2004), is a factor that attracts inward FDI.

Despite the enormous amount of literature on FDI (see Faeth, 2009; Mohamed and Sidiropoulos, 2010; Chanegriha et al., 2016) and non-renewable energy resources (for example, Crawford et al., 1984; Mitchell, 2009; Zhang et al., 2013), considered separately, not many studies have looked at the two topics together, establishing and appraising a (possible) correlation and causality between these two variables. The few studies there are in this domain focus on a limited number of regions and countries including Sub-Saharan Africa (SSA) (Asiedu, 2006; Ezeoha and Cattaneo, 2011), the Middle East and North African (MENA) countries (Mohamed and Sidiropoulos, 2010; Aziz and Mishra, 2016), African countries (Sanfilippo, 2010),

2

the Economic Community of West African States (ECOWAS) (Ajide and Raheem, 2016), China (Cheung and Qian, 2009), India (Kumar and Chadha, 2009), Eurasia (Poland, Hungary and the Baltic states) (Deichmann et al., 2003), the Southern African Development Community (SADC) (Mhlanga et al., 2010), the nations from the ex-Soviet Union (Ledyaeva, 2009), the BRICS (Brazil, Russia, India, China & South Africa) (Jadhav, 2012), and a group of developing countries (Asiedu and Lien, 2011; Asiedu, 2013). Furthermore, except the works of Asiedu (2006), Asiedu and Lien (2011), Ezeoha and Cattaneo (2011) and Asiedu (2013), such studies neither tend to look specifically at the possible correlation and causality between FDI and NRERs nor at the particular relevance of the latter as determining the former. The present study sets out to add evidence to this research area by analysing the role of NRERs in attracting FDI, controlling for a set of factors that are traditionally regarded as influencing this last macroeconomic variable (for example, human capital, market size, political stability, openness of the economy) (see Faeth, 2009; Chanegriha et al., 2016). In order to pursue such endeavour, we resort to econometric panel data techniques involving 125 countries over the period 1995-2012. The set of countries considered is rather heterogeneous in terms of NRERs: around one quarter of the countries do not possess ‘proven reserves’ (e.g., Finland, Portugal or Sweden), or have no/negligible share of NRERs in their total exports (e.g., Botswana, Central African Republic, Haiti), whereas 18% of the countries hold vast quantities of ‘proven reserves’ (e.g., China, USA, India or Australia), and about 10% are large NRERs exporters (e.g., Algeria, Nigeria, Yemen, Rep., Kuwait, Saudi Arabia). The paper is organised as follows. Section 2 gives a brief overview of the literature on endowments of NRERs and FDI. The methodology is described in Section 3, with details on the econometric model, the proxy variables and relevant data sources. The empirical results of the model are presented and discussed in Section 4. The last section sets out the main contributions, policy implications, limitations of the study, and put forward future lines of research.

3

2. FDI and natural resource endowments: literature review A myriad of factors are consider relevant to attract FDI. Among these stand (see Faeth, 2009; Chanegriha et al., 2016): endowment related factors, most notably, non-renewal natural energy resources (NRERs) and human capital (Cleeve, 2008; Asiedu, 2006); economic and political factors, which include market size (Mohamed and Sidiropoulos, 2010; Vijayakumar et al., 2010), market growth, openness of the economy (Asiedu, 2006; Botrić and Škuflić, 2006), economic (in)stability, production costs, financial and tax incentives, and infrastructure facilities (Biswas, 2002; Asiedu, 2006); institutional related factors, namely corruption, political (in)stability (Asiedu, 2006; Mohamed and Sidiropoulos, 2010), and institutional quality. The importance of each of these factors for attracting FDI is closely related to the motives for investors to carry out this type of investment. Dunning and Lundan (2008) summarize the four main motives for companies producing abroad and conduct FDI: ‘market-seeking’, ‘resource-seeking’, ‘efficiency-seeking’, and ‘asset-seeking’ FDI. The main purpose of market-seeking FDI is to supply the recipient country or nearby countries. Resourceseeking FDI aims at achieving specific resources of better quality and at lower cost than the investor’s country of origin. Efficiency-seeking FDI aims at rationalize the structure of the existing investment and that was motivated by the search for markets or resources. Finally, the main objective of strategic asset-seeking FDI is to keep or improving investor’s global competitiveness, thus promoting its ‘long-term strategic objectives’ (Dunning and Lundan, 2008). In this way, as stated by Dunning (1998, p. 9), “[d]ifferent kinds of investment incentives are needed to attract inbound MNE activity of a natural-resource-seeking, c.f. that of a market - or efficiency-seeking, kind”. Table 1 synthetizes the main motives for FDI, their objectives, as well as some host country attractiveness factors.

4

Regarding the resource-seeking FDI, which is particularly relevant to our work in that the focus is on the role of natural resources, Sanfilippo (2010) report that recently several African countries have been targeted as the main destination of Chinese FDI given the growing needs of the country in natural resources, thus advocating a positive relationship between natural resource endowment and FDI. Indeed, the NRERs are currently at the centre of the discussion about energy security (Moran, 2010; Wolf, 2014; Capellán-Pérez et al., 2015). According to some authors (for example, Velthuijsen and Worrel, 1999; Salim et al., 2014), natural energy resources, especially NRERs, such as oil, coal and natural gas, have been playing a key role in economic development. Empirically, existing studies on the relationship between the natural resources endowments and FDI show mixed results (cf. Table 2). Most studies (Deichmann et al.., 2003; Asiedu, 2006; Cheung and Qian, 2009; Ledyaeva, 2009; Mohamed and Sidiropoulos, 2010; Sanfilippo, 2010; Aziz and Mishra, 2016) obtained a positive relation between natural resources endowments and FDI. For instance, Ledyaeva (2009) looked at the nations from the exUSSR in the period 1995-2005 and noted that the regions richer in natural resources, measured by the oil and natural gas production index, attract higher amounts of FDI. Controlling for a huge set of factors that may influence the inflow of FDI to Eurasia countries in the period 1989-1998 (for example, reform measures, importance of private sector in the economy, GDP and per capita GNP, inflation rate, number of years the economy is (was) under central planning, rule of law, investment climate; human and social capital), Deichmann et al. (2003) mention that these countries, rich in oil and natural gas, would not be attractive were it not for these resources. Ezeoha and Cattaneo (2011) and Ajide and Raheem (2016) obtained, in some estimated models, a positive relation between natural resources endowments and FDI whereas in other models results were inconclusive. Mhlanga et al. (2010) also obtained inconclusive results for the impact of natural resources endowments on FDI in the SADC countries.

5

Finally, there are studies (Asiedu and Lien, 2011; Jadhav, 2012; Asiedu, 2013) which found a negative relation between FDI and natural resources endowments. Asiedu and Lien (2011), studying a group of 112 developing countries for a period of 25 years (1982-2007), and analysing the interaction between democracy, natural resources export intensity and FDI, concluded that natural resources have a negative direct effect on FDI and “significantly alter the relationship between FDI by reducing the positive effect of democracy on FDI” (Asiedu and Lien, 2011, p.106). Later, and focusing on BRICS economies, Jadhav (2012) also evidenced that natural resource availability had significant negative effect on total inward FDI, suggesting that FDI in these countries were not resource-seeking. In a more recent paper, using data from 99 developing countries in the period 1984-2011, Asiedu (2013) also obtained a negative effect of natural resources on FDI.3 The author established that the presence of good institutions attenuates the negative impact of natural resources on FDI but does not completely neutralize it.

All the empirical studies mentioned above use econometric models to gauge the relevance of natural resources in attracting FDI in various countries. It can be seen that, even though the studies that examine the relevance of natural resources to attract FDI are unanimous as to the importance of this determinant, most of them do not look specifically at NRERs. Those that do (for example, Deichmann et al., 2003; Ledyaeva, 2009; Mohamed and Sidiropoulos, 2010) focus on very specific regions of the world (Central and Eastern 3

Asiedu and Lien (2011) explain the existence of a negative relationship based on the following arguments. First, the abundance

of natural resources tends to induce an appreciation of the respective local currency which reduces competitiveness of the country’s exports, discouraging FDI in other sectors. Second, natural resources, especially oil, “are characterized by booms and busts”, contributing to a high fluctuation of the exchange rate; furthermore, a high share of exports of fuels and minerals in total exports indicates low export diversification and consequent increase in the country's exposure to external shocks. These two factors “generate macroeconomic instability and therefore reduce FDI” (Asiedu and Lien, 2011, p.104). Finally, in countries rich in natural resources FDI tend to converge to the natural resource sector which although calling for a large initial capital investment requires small amounts for operations that follow. “Thus, after the initial phase, FDI may be staggered” (Asiedu and Lien, 2011, p.104).

6

Europe, Central Asia, and the MENA countries). Thus, our study is intended to add empirical evidence on the special relevance of NRERs and their (possible) correlation with FDI. We use panel data techniques and look at a large group of countries, including countries having NRERs endowments, with the aim of establishing a relation between a country’s endowment of such resources and its performance in terms of FDI attraction.

3. Methodology 3.1. Data and variables The present study examines the determinants of FDI inflows in 125 countries over the period from 1995 to 2012 (see Table A1 in the Appendix for the list of countries). Table A2, in the appendix, provides information about the proxies for the relevant variables, namely the sources of data and their definitions. Following the literature (see Aziz and Mishra, 2016), we categorize the determinants of FDI inflows in three main categories: endowment (non-renewal energy resources, human capital), economic and policy (market size, market growth, trade openness, economic instability, production costs, tax rate, infrastructure), and institutional quality (corruption control, political stability, rule of law).

Foreign direct investment (FDI) As is standard in the literature (see Biswas, 2002; Asiedu, 2006; Mohamed and Sidiropoulos, 2010; Asiedu and Lien, 2011), the dependent variable is the net inflows of FDI in percentage of the GDP. It encompasses new investment inflows less disinvestment regarding the acquisition of a lasting management interest (10 percent or more of voting stock) in an enterprise operating in an economy other than that of the investor.

7

Endowments Non-renewal natural resources Several studies (e.g., Deichmann et al., 2003; Mhlanga et al., 2010; Ezeoha and Cattaneo, 2011; Ajide and Raheem, 2016) use dummy variables as proxy for natural resources endowments. Ezeoha and Cattaneo (2011) and Ajide and Raheem (2016) chose to use a dummy variable which takes the value one if the country is classified as abundant in natural resources and zero in the opposite situation, whereas Mhlanga et al. (2010) used a dummy variable related to investments in mining industry. In turn, Deichmann et al. (2003) ranked a country’s natural resources endowment at three levels (poor, moderate and high endowment). Other authors (e.g., Ledyaeva, 2009; Sanfilippo, 2010; Aziz and Mishra, 2016) use production indicators rather than dummy variables. For instance, the study by Aziz and Mishra (2016) on the location determinants of foreign direct investment to 16 Arab economies employs the total oil supply as proxy for natural resources endowments (which includes the production of crude oil, natural gas plant liquids, and other liquids, and refinery processing gain). The share of mineral fuel (oil, coal, gas) exports in total exports is the most used proxy (see Asiedu, 2006; Asiedu and Lien, 2011; Asiedu, 2013). Mohamed and Sidiropoulos (2010), studying the MENA countries, also used mineral fuels but in levels. Analysing FDI from the investor’s point of view, Cheung and Qian (2009) use a wider proxy (including, beside mineral fuels, ores and metals) which represents the demand for sundry raw materials in the various countries. In this context, and following the scarce literature available, one of the proxies used in the present study to measure NRERs endowments is the share of fuel minerals (i.e., coal, oil and natural gas) in total exports. It is important to note that most non-renewable resources are not wholly available for extraction, since only a small fraction of minerals overcomes the mineralogical barrier, with 'proven reserves' being the economically extractable part of a resource (Grafton et al.., 2004). Although it is technically possible to extract resources from beyond the mineralogical barrier, the cost is excessively high and so extraction takes

8

place only up to the barrier. ‘Proven reserves’ are thus confined to a small part of existing resources, which affects the scarcity of resources and stresses the importance of such reserves to economic development (Cleveland and Stern, 1999). A number of countries, such as Cameroon, Chad or Ecuador, although possessing a negligible amount of ‘proven reserves’ of NRERs nonetheless register a high ratio of fuel exports in their total exports. It is therefore pertinent to use a proxy based on ‘proven reserves’ in addition to the more traditional proxy, the share of fuel minerals in total exports. To the best of our knowledge no empirical study has yet been published that examines the effect of holding oil, coal and natural gas ‘proven reserves’ on FDI attraction. Given that the proven reserves of each of the three resources are expressed in different units (oil in barrels, coal in tonnes, and natural gas in m3), for comparability purposes they have been converted to TOE (tonnes of oil equivalent), taking barrels of oil to be American barrels (42 US gallons being approximately 158.9873 litres). Resorting to The International Energy Agency definition of tonne of oil equivalent (TOE), we considered 1 BOE (barrel of oil equivalent) = 0.14 TOE; 1 TCE (tonne of coal equivalent) = 0.7 TOE; and 103 m3 natural gas = 0.82 TOE (see also Heitor et al., 2000).

Human capital Several seminal studies (e.g., Lucas, 1990; Dunning, 1998; Noorbakhsh et al., 2001) underline that that lack of human capital discourages foreign investment in both less-developed and developing countries. Zhang and Markusen (1999) and Teixeira and Tavares-Lehmann (2014) evidence that the availability of skilled labour in the host country is a direct requirement of multinationals and affects the volume of FDI inflows. In line with more recent literature in the area, human capital is measured by the average number of years of schooling of the working-age population (Teixeira, 2005; Barro and Lee, 2013).

Economic and policy 9

Market size Market size is seen by the empirical literature as being crucial to attracting FDI (Schneider and Frey, 1985; Mhlanga et al.., 2010), such that countries with a larger domestic market will be more attractive to investors because of the greater number of potential consumers. Some authors (Botrić and Škuflić, 2006; Mohamed and Sidiropoulos, 2010) use the number of inhabitants as a proxy of this determinant. But it is felt that this indicator does not give a true picture of the attractiveness of the market, especially in a broad sample of countries which includes underdeveloped, developing and developed nations, as a large population need not translated into a large number of consumers if they lack purchasing power (Ietto-Gillies, 2005). Based on the empirical literature, therefore, per capita GDP was deemed a more suitable indicator to measure the influence of market size.

Market growth When it comes to potential for market growth, the literature (for example, Mhlanga et al.., 2010; Mohamed and Sidiropoulos, 2010) generally suggests using the GDP or GNP growth rate as a proxy for this determinant. A high growth rate for the market should attract more FDI since a growing economy offers more opportunities for higher profits. So the real GDP growth rate is used as a proxy as it is corrected for the effect of price variation, thereby giving more credible information on GDP growth. Trade openness Openness of the economy is seen in the literature as one of the key determinants of FDI (Vijayakumar et al., 2010; Chanegriha et al., 2016). A country can increase its attractiveness by adopting a policy that favours foreign trade, encouraging domestic producers to export, increasing their profitability and attracting foreign investors (Mohamed and Sidiropoulos, 2010). Based on the empirical literature (Botrić and Škuflić, 2006; Cleeve, 2008), we decided to use the weight of foreign trade in GDP to measure the degree of openness,

10

such that the greater the ratio the more open the country and the more FDI it would attract (Cleeve, 2008). A positive relation of this determinant with FDI is thus expected.

Economic instability Since high or volatile rates of inflation are a clear sign of economic instability (Botrić and Škuflić, 2006), the rate of inflation was chosen as a proxy for measuring each country’s economic instability. High inflation rates distort economic activity and reduce investment in productive industries, leading to lower economic growth (Mohamed and Sidiropoulos, 2010). So we expect that high inflation discourages FDI.

Production costs Issues of cost reduction and increasing competitiveness often tempt firms to relocate their production facilities in places where such costs are lower (Dunning and Lundan, 2008), specifically labour costs, with worker’s wage being the proxy most often referenced in the literature (Schneider and Frey, 1985; Biswas, 2002). We therefore expect that low production costs attract larger FDI inflows. In our study, the large size of the sample, on the one hand, and the inclusion of countries with scanty statistical information on the other mean that this indicator could not be used. Two other indicators were chosen instead. The first is the unemployment rate, which aims at measuring labour market rigidity (see Cockx and Ghirelli, 2016) and thus an important part of labour related costs. The higher the unemployment rate the more rigid the labour market and the less attractive it will be for investors. The second indicator relates to the cost of imports (measured in USD by container). This includes all import costs - administrative charges, the cost of keeping customs facilities, transport, customs clearance, and other expenses - that can be a determinant in the choice of location (see Husan, 1996), since this can be a significant cost in raw materials or machinery that has to be imported.

11

Tax rate Even though the empirical literature suggests temporary tax exemptions, tax concessions and ease of repatriation of profits as indicators of financial and tax incentives (Cleeve, 2008), the large size of the sample meant that none of these data could be obtained for all the countries. So the total tax rate (as percentage of commercial profits) was chosen instead, since this indicator expresses all the taxes payable by a firm. According to the literature (see Tavares-Lehmann et al., 2012) it is expected that countries with lower tax rates will tend to attract greater inward flows of FDI.

Infrastructure With respect to infrastructure, two proxies were used to measure their quality: the number of phone lines per 100 inhabitants and the losses of electric power transmission and distribution (in percentage of the output). As the sample includes a range of countries with widely divergent degrees of development – developed and developing countries – from all over the world, the first proxy should fit the development level of the developing countries better and the second should identify the different degrees of development among the developed countries. Bearing in mind the relevant literature, we expect that good infrastructure (expressed by the high number of phone lines and/or a smaller loss of electric power transmission and distribution) would be attractive to foreign investors (Biswas, 2002; Asiedu, 2006).

Institutional quality Corruption control In the spirit of extant empirical literature (for example, Asiedu, 2006; Cleeve, 2008; Aziz and Mishra, 2016), we chose the World Bank’s Worldwide Governance Indicator, ‘control of corruption’, as a proxy for a country’s level of corruption. The higher the ‘control of corruption’ (a standardized measure with a maximum of, approximately, 2.5, and a minimum of -2.5), the lower the perceptions of citizens that public

12

power is exercised for private gain and/or the "capture" of the state by elites and private interests is substantial. High ‘control of corruption’ figures are thus linked to higher foreign investment.

Political stability Some authors, such as Mhlanga et al. (2010), use the risk rating of a country to measure political stability. But given the difficulty in obtaining this indicator, an alternative was chosen: the political stability and absence of violence and terrorism measure. This proxy measures the perceived improbability of political instability and/or politically-motivated violence, including terrorism. It can take values from -2.5 to 2.5 and the higher the score the greater the stability. It is expected that high levels of political stability, which reflect low political risk, tend to attract more FDI.

Rule of law With respect to institutional quality, we followed Asiedu (2006) and took the degree of effectiveness of the rule of law as a proxy. This indicator measures the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence. The closer to 2.5 (maximum) the greater the confidence. It is thus expected that a high degree of effectiveness of the rule of law should attract investors, since it offers them greater security.

3.2. Econometric estimation Panel data estimations have a few advantages over the traditional cross-country models, permitting to study the dynamics of adjustment and to estimate their effects over a long period of time. On the one hand, panel data can analyse a set of variables for a great number of countries, which provides more information. Panel data estimations also assume that countries are heterogeneous, with specific and unobservable

13

characteristics. ‘Cross-section’ and ‘times series’ estimations do not control for this heterogeneity, meaning the results may be biased (Greene, 2011). Therefore, to estimate the effects of relevant variables, namely non-renewal energy resources (NRERs), on FDI inflows we opted for the estimation of panel data, such as in more recent studies which seek to assess FDI determinants (e.g., Helmy, 2013; Okafor et al., 2015). According to the literature review, the econometric model specification to estimate is (all variables are expressed in logs):

FDIit = 1 + 2 NRERit +3 HCit + 4 MSit + 5 MGit + 6 TOit+7 EIit+8 PCit +9 Taxit +

10 INFRit

+11 CCit +12 PSit +13 RLit +ui+εit

where i represents the countries’ index and t represents time. FDI

FDI inflows as percentage of the GDP

NRER non-renewal energy resources (exports of NRER in total exports; proven reserves) HC

human capital stock (average years of schooling of the adult population)

MS

market size (GDP per capita)

MG

market growth (real average GDP annual growth)

TO

trade openness (sum of exports and imports in total GDP)

EI

economic instability (inflation rate)

PC

production costs (unemployment rate; cost of imports)

Tax

tax rate (total tax rate in percentage of commercial profits)

INFR infrastructure (fixed telephone subscription per capita and losses, in percentage of output, of electric power transmission and distribution) CC

corruption control

PS

political stability

14

RL

rule of law

ui

the unobserved time-invariant fixed effect

εit

the unobserved random coefficient

4. Results In order to investigate the possibility of non-stationarity in the dataset, it is first necessary to determine the existence of unit roots in the data series. For this study we have chosen Levin–Lin–Chu (LLC) (2002) and Harris–Tzavalis (HT) (1999) tests.4 The combined results of the panel unit root tests indicate that all variables are I(0), showing that the null hypothesis of a panel unit root (non-stationarity) in the level of the series can be rejected (see Table 3).

The three proxies for institutional quality are (expectedly) highly correlated (see Table A3 in Appendix). Moreover, one of the proxies for infrastructure (fixed telephone subscriptions) is also highly correlated with several variables (human capital, control of corruption, rule of law, GDP per capita). Thus, to avoid multicollinearity, we considered 6 distinct specifications (see Table 4): three considering each institutional quality proxy in isolation (Models 1: control of corruption; Model 2: Political stability; Models 3: Rule of law), combined with the two infrastructure proxies (Models 1-3: fixed telephone subscriptions; Models 4-6:

4

The Levin–Lin–Chu (LLC) (2002) test has as the null hypothesis that all the panels contain a unit root. Because the LLC test

requires that the ratio of the number of panels to time periods tend to zero asymptotically, it is not well suited to datasets with a large number of panels and relatively few time periods. Thus, it is adequate to complement the analysis with the Harris and Tzavalis (HT) (1999) test. This also has a null of unit root versus an alternative with a single stationary value, but it is designed to be applied to data sets which are relatively short in T.

15

electric power transmission and distribution losses). Then, we estimate the six specifications including, separately, each natural resource proxies (Models A – NRER exports; Models B – Proven NRER).5 The fixed effects model was favoured over the random effects model by the Hausman test. The models estimated present a reasonable fit, with F-stat indicating that all models are overall significant and R2 with values in line with previous studies using similar estimation methods (e.g., Asiedu, 2006; Mohamed and Sidiropoulos, 2010).

Our estimates unambiguously evidence that even when controlling for economic, policy and institutional quality factors, countries whose exports are highly concentrated in oil, coal and gas tend to attract FDI. All other factors being held constant, if the share of non-renewal energy resources (NRER) exports increase by one percent, we would expect that, on average, the ratio of FDI in GDP of the countries included in our sample increases by 0.02-0.03 percent. This result is in line with the studies by Asiedu (2006) or Cheung and Qian (2009), but contrast with those by Asiedu and Lien (2011), Jadhav (2012), and Asiedu (2013) (all these studies use the same proxy as we for measuring countries’ natural resources endowments). Thus, as the evidence gathered for Sub-Saharan Africa (Asiedu, 2006) and China’s outward FDI (Cheung and Qian, 2009), we conclude that FDI is largely driven by natural resources. In short, our results are consistent with the resource-seeking strategy (Dunning and Lundan, 2008). Interestingly, however, when we use ‘proven reserves’ (i.e., recoverable and economically profitable reserves) as proxy for natural resources endowments we find that NRERs fail to emerge as a significant determinant of FDI.

5

Human capital and GDP pc are also highly correlated. We estimated additional combinations of models considering each of these

variables separately. However, estimations results did not differ from those presented in Table 4. Therefore to keep the presentation of results simple, we opted to not present the whole combined estimations.

16

Up to the present date, no study used ‘proven reserves’ as proxy for NRER endowments. The closest proxy might be the production/supply of oil and gas used by Ledyaeva (2009), Sanfilippo (2010), and Aziz and Mishra (2016). Differently from our results, Ledyaeva (2009) suggests that oil and gas resources are key determinants of FDI inflows into Russia, whereas Sanfilippo (2010) demonstrates that Chinese investors in Africa are motivated by the search for long-term access to natural resources, and Aziz and Mishra (2016) find that the total oil supply positively affects FDI inflows to Arab economies. Our results seem to entail that being highly endowed in ‘proven’ NRER does not necessarily imply that a country will attract more FDI. In other words, the impact of natural resource abundance on FDI is highly contingent on countries’ exports being little diversified and dependent on mineral fuel related resources (petroleum, coal, or natural gas). Regarding the remaining FDI determinants, our results follow quite closely extant literature in the area. Human capital emerges significantly, albeit less strongly as other studies suggest (see Chanegriha et al., 2016). Such result advises the need for schooling/ skills to be promoted/ developed among adult active population if a country wishes to attract FDI. Thus, as Aziz and Mishra (2016, p. 343) refer a “well-educated labour force can be a key element in attracting FDI”. In our sample, over the period of analysis (1995-2012), both economic and policy, and institutions emerge as critical factors for attracting FDI. The most significant determinants are ‘rule of law’ (related to the quality of contract enforcement, property rights), ‘trade openness’, and ‘market growth’. In concrete, a one percent improvement in the perceptions of quality of contract enforcement or in trade openness leads, on average, to 0.18 percent increase in FDI inflows. The corresponding figure for ‘market growth (measured by real annual GDP growth) is about 0.13 percent. This suggests, in line with other previous studies (e.g., Felisoni de Angelo et al., 2010; Helmy, 2013; Okafor et al., 2015), that efforts made to improve the quality of institutions, fostering of policies that liberalise trade regimes or stimulate internal markets demand do have an impact on FDI.

17

We further find that other policy and institutional factors, especially the factors that make the investment environment costly, such as labour market rigidity, tax rate and corruption, have significant influence on FDI inflows. Specifically, our estimations evidence that a one percent decrease in the unemployment rate or in total tax rate (in percentage of commercial profits) leads to 0.04 and 0.08 percent increase in FDI, respectively. Moreover, a one percent improvement in the perceptions of corruption control results in a 0.10 percent increase in FDI inflows. Market size, economic stability, and infrastructure facility do not seem to help build up the confidence of foreign investors and increase FDI inflows, which contrast with several extant studies (e.g., Aseidu, 2006; Xaypanya et al., 2015).

5. Conclusions The present study analysed the impact of a country’s Non-Renewable Energy Resources (NRERs) endowments on FDI inflows in a wide sample of countries over almost two decades (1995-2012). Controlling for a number of factors traditionally believed to influence FDI (e.g., human capital, trade openness, tax rates, institutional quality), and resorting to fixed effect panel estimations the study contributes to the scientific literature in the area in three main ways. First, it contributes to the very scarce literature that specifically addressed the role of NRERs on FDI inflows. Among the few (13) studies that includes NRERs as a determinant of FDI, only four – Asiedu (2006; 2013), Ezeoha and Cattaneo (2011), and Asiedu and Lien (2011) – explicitly examine the interaction between natural resources and FDI. Second, the present study includes a more updated and wider heterogeneity of countries (125 countries: developed, developing, less developed, and emergent; located in all continents; rich/poorly endowed in natural resources), over a longer period (18 years: 1995-2012), than existing studies which explicitly or implicitly includes NRERs. The vast majority of the latter focuses particular/small set of countries: Arab

18

economies (Aziz and Mishra, 2016); Economic Community of West African States (Ajide and Raheem, 2016); Eurasia (Deichmann et al., 2003); Sub-Saharan Africa countries (Asiedu, 2006; Ezeoha and Cattaneo, 2011); and BRICS (Brazil, Russia, India, China and South Africa) (Jadhav, 2012). The works by Asiedu and Lien (2011) and Asiedu (2013) encompass a large number of countries (respectively, 112 and 99) but only consider developing countries. Regarding the time span, although four studies – Mohamed and Sidiropoulos, 2010; Asiedu and Lien, 2011; Asiedu, 2013; Aziz and Mishra, 2016 - cover a longer period than ours, most of the existing studies focus their analysis on a period of study inferior to fifteen years, with the latest year analysed being 2008 or earlier. Third, it is to the best of our knowledge the first empirical study to assess the impact of ‘proven’ (the economically extractable part of) NRERs on FDI inflows. We plainly demonstrate that the impact of NRERs endowments on FDI attraction is contingent on how those endowments are measured. Highly resource endowed countries in terms of ‘proven reserves’ (of oil, coal and gas) do not necessarily attract FDI, whereas those countries whose mineral fuel exports constitutes a large part of the corresponding total exports tend to be privileged recipients of FDI. This evidence seems to support Mina’s (2007) content that rich ‘proven’ NRER countries have abundance of financial resources that makes them less reliant on overseas finance or that resource seeking FDI targets mainly economically shaky countries (Ajide and Raheem, 2016). Our results, nevertheless, stand in sharp contrast with the recent contribution by Asiedu (2013), who found that, even after controlling for the quality of institutions and other relevant determinants of FDI, the share of natural resources in total exports harmfully affects FDI. As referred above, when using the share of mineral fuel exports in total exports we found a highly significant and positive effect of NRER endowments on FDI. Moreover, our estimates confidently indicate that when controlling for NRERs endowments, FDI attraction is fostered once countries make convincing efforts to open up their economies to international trade and devote resources to the enhancement of their human capital, control corruption, and have more beneficial tax rates.

19

Such results highlight the prominence of governmental actions over Nature’s idiosyncrasies, in particular the importance of investing in human capital, improving the quality of institutions (by controlling corruption and enforcing contracts and property rights), and promoting policies to open/ liberalise the economy (by adopting export-oriented policies and eliminating/lowering taxes on corporate profits). Despite the contributions of the present study, several limitations should be itemized, which nevertheless constitute promising avenues for future research. First, we did not consider the different final uses of the three types of resource – oil, coal, and natural gas – which may lead to interesting conclusions on the targeting location of FDI based on the type of fuel. Future research could explore this issue. Second, although our sample included a wider diversity of countries, such diversity was somehow overlooked. It would also be interesting to divide the sample in developed and developing countries and assess the extent to which results would hold for these two groups. Third, further improvements could be accounted for at the methodological level by using the latest dynamic panel data models which allow for unobserved countryspecific effects, measurement errors, estimator bias due to persistency in time series, and even endogeneity problems (Bond et al. 2001; Vedia-Jerez and Chasco, 2016). References Ajide, K.; Raheem, I. (2016), “Institutions-FDI Nexus in ECOWAS Countries”, Journal of African Business, 17 (3), 319-341. Asiedu, E. (2006), “Foreign direct investment in Africa: The role of natural resources, market size, government policy, institutions and political instability”, The World Economy, 29 (1), 63-77. Asiedu, E. (2013), “Foreign direct investment, natural resources and institutions”, International Growth Centre (IGC) Working Paper Series, March. Asiedu, E.; Lien, D. (2011), “Democracy, foreign direct investment and natural resources”, Journal of International Economics, 84 (1), 99-111.

20

Aziz, O.G.; Mishra, A.V. (2016), “Determinants of FDI inflows to Arab economies”. Journal of International Trade and Economic Development, 25 (3), 325-356. Barro, R.; Lee, J.-W. (2013), "A new data set of educational attainment in the world, 1950-2010", Journal of Development Economics, 104: 184-198. Betz, M.R.; Partridge, M.D.; Farren, M.; Lobao, L. (2015). “Coal mining, economic development, and the natural resources curse”, Energy Economics, 50, 105-116. Biswas, R. (2002), “Determinants of foreign direct investment”, Review of Development Economics, 6 (3), 492-504. Bond, S.R.; Hoeffler, A.; Temple, J. (2001), “GMM estimation of empirical growth models”, Discussion Paper 3048, Centre for Economic Policy Research (CEPR). Botrić, V.; Škuflić, L. (2006), “Main determinants of foreign direct investment in the southeast European countries”, Transition Studies Review, 13 (2), 359-377. Brooks, D.H.; Hasan, R.; Lee, J.-W.; Son, H.H.; Zhuang, J. (2010) “Closing development gaps: challenges and policy options”, ADB Economics Working Paper Series 209, Manila: Asian Development Bank. Capellán-Pérez, I., Mediavilla, M., de Castro, C., Carpintero, Ó., Miguel, L.J. (2015), “More growth? An unfeasible option to overcome critical energy constraints and climate change”, Sustainability Science, 10 (3), 397-411. Cheung, Y-W.; Qian, X. (2009), “Empirics of China’s outward direct investment”, Pacific Economic Review, 14 (3), 312-341. Cleeve, E. (2008), “How effective are fiscal incentives to attract FDI to Sub-Saharan Africa?”, The Journal of Developing Areas, 42 (1), 135-153. Cleveland, C.J. e Stern, D.I. (1999), “Indicators of natural resource scarcity: a review and synthesis”, in Jeroen C.J.M van den Bergh (org.), Handbook of Environmental and Resource Economics, Massachusetts: Edward Elgar Publishing, pp. 89-107.

21

Cockx, B.; Ghirelli, C. (2016), “Scars of recessions in a rigid labor market”, Labour Economics, 41, 162176. Crawford, V.P.; Sobel J.; Takahashi I. (1984), “Bargain, strategic reserves, and international trade in exhaustible resources”, American Journal of Agricultural Economics, 66 (4), 472-480. Deichmann, J.I.; Eshghi, A.; Haughton, D.M.; Sayek, S.; Teebagy, N.C. (2003), “Foreign direct investment in the Eurasian transition states”, Eastern European Economics, 41(1), 5-34. Dunning, J.H. (1998), "Location and the Multinational Enterprise: A Neglected Factor?", Journal of International Business Studies, 29 (1), 45-66. Dunning, J.H.; Lundan, S.M (2008), Multinational Enterprises and the Global Economy, Cheltenham: Edward Elgar Publishing Limited. Ezeoha, A. E.; Cattaneo, N. (2011), “FDI flows to Sub-Saharan Africa: The impact of finance, institution and natural resource endowment”, Paper Submission for the CSAE Conference 2011: Economic Development in Africa, in http://www.csae.ox.ac.uk/ conferences/2011-EdiA/papers/294-Ezeoha.pdf, last accessed on 27th October 2016. Faeth, I. (2009), “Determinants of foreign direct investment – a tale of nine theoretical models”, Journal of Economic Surveys, 23 (1), 165-196. Felisoni de Angelo, C.; Eunni, R.V.; Manoel Martins Dias Fouto, N. (2010), “Determinants of FDI in emerging markets: Evidence from Brazil”, International Journal of Commerce and Management, 20 (3), 203-216. Gohou, G., Soumaré, I. (2012). Does Foreign Direct Investment Reduce Poverty in Africa and are There Regional Differences?, World Development, 40 (1), 75-95. Grafton, R.Q.; Adamowicz, W.; Dupont, D.; Nelson, H.; Hill, R.J.; Renzetti, S. (2004), The Economics of the Environment and Natural Resources, Malden: Blackwell Publishing. Greene, W.H. (2011), Econometric Analysis, (7th Edition), Pearson Education, New York University.

22

Harris, R.D.F.; Tzavalis, E. (1999), “Inference for unit roots in dynamic panels where the time dimension is fixed”, Journal of Econometrics 91: 201–226. Heitor, M.; Ferrão, P; Diogo, A. (2000), “Perspectivas sobre a utilização racional de energia no sector da cristalaria: a utilização do gás natural”, Instituto Superior Técnico: Centro de estudos em inovação, tecnologia e políticas de desenvolvimento, in http://in3.dem.ist.utl.pt/laboratories/PDF/Energia.pdf, accessed on June 6, 2011. Helmy, H.E. (2013), “The impact of corruption on FDI: Is MENA an exception?”, International Review of Applied Economics, 27 (4), 491-514. Henson, S.; Yap, F. (Eds.) (2016). The Power of the Chinese Dragon. Implications for African Development and Economic Growth. Series: Palgrave Readers in Economics, UK: Palgrave Macmillan. Husan, R. (1996), “The limitations of low labour costs as an inducement to foreign direct investment: An example from the motor industry”, European Business Review, 96 (4), 26-29. Ietto-Gillies, G. (2005), Transnational Corporations and International Production, Cheltenham: Edward Elgar Publishing. Jadhav, P. (2012), “Determinants of foreign direct investment in BRICS economies: Analysis of economic, institutional and political factor”, Procedia - Social and Behavioral Sciences, 37, 5 – 14. Kumar, N.; Chadha, A. (2009), “India’s outward foreign direct investments in steel industry in a Chinese comparative perspective”, Industrial and Corporate Change, 18 (2), 249-267. Ledyaeva, S. (2009), “Spatial econometric analysis of foreign direct investment determinants in Russian regions”, The World Economy, 32 (4), 643-666. Levin, A.; Lin, C.-F.; Chu, C.-S.J. (2002), “Unit root tests in panel data: Asymptotic and finite-sample properties”, Journal of Econometrics 108: 1–24.

23

Mhlanga, N.; Blalock, G.; Christy, R. (2010), “Understanding foreign direct investment in the southern African development community: an analysis based on project-level data”, Agricultural Economics, 41 (3-4), 337-347. Mina, W. (2007), “The location determinants of FDI in the GCC countries”, Journal of Multinational Financial Management, 17 (4), 336-348. Mitchell, T. (2009), “Carbon democracy”, Economy and Society, 38 (3), 299-432. Mohamed, S.E; Sidiropoulos, M.G. (2010), “Another look at the determinants of foreign direct investment in MENA countries: an empirical investigation”, Journal of Economic Development, 35 (2), 75-95. Moran, T.H. (2010), “Is China trying to “lock up” natural resources around the world?”, in http://voxeu.org/index.php?q=node/4684, accessed on May 4, 2011. Noorbakhsh, F.; Paloni, A.; Youssef, A. (2001), “Human capital and FDI inflows to developing countries: New empirical evidence”, World Development, 29(9), 1593–1610. Okafor, G.; Piesse, J.; Webster, A. (2015), “The motives for inward FDI into Sub-Saharan African countries”, Journal of Policy Modeling, 37 (5), 875-890. Salim, R.A.; Hassan, K.; Shafiei, S. (2014), “Renewable and non-renewable energy consumption and economic activities: Further evidence from OECD countries”, Energy Economics, 44, 350-360. Sanfilippo, M. (2010), “Chinese FDI to Africa: What Is the Nexus with Foreign Economic Cooperation?”, African Development Review, 22 (S1), 599–614. Schneider, F.; Frey, B.S. (1985), “Economic and political determinants of foreign direct investment”, World Development, 13 (2), 161-175. Tavares-Lehmann, A.T.; Coelho, Â.; Lehmann, F. (2012), “Taxes and foreign direct investment attraction: A literature review”, Progress in International Business Research, 7, 89-117.

24

Teixeira, A.A.C. (2005), “Measuring aggregate human capital in Portugal: 1960-2001”, Portuguese Journal of Social Science, 4 (2), 101-120. Teixeira, A.A.C.; Tavares-Lehmann, A.T. (2014), “Human capital intensity in technology-based firms located in Portugal: Does foreign ownership matter?” Research Policy, 43(4), 737-748. UNCTAD (2007), Transnational corporations, extractive industries and development, World Investment Report 2007, United Nations, New York and Geneva. Vedia-Jerez, D.H.; Chasco, C. (2016), “Long-run determinants of economic growth in south America”, Journal of Applied Economics, 19 (1), 169-192. Velthuijsen, J.W.; Worrel, E. (1999), “The economics of energy”, in Jeroen C.J.M van den Bergh (org.), Handbook of Environmental and Resource Economics, Massachusetts: Edward Elgar Publishing, pp. 177-194. Vijayakumar, N.; Sridharan, P.; Rao, K.C.S. (2010), “Determinants of FDI in BRICS countries: A panel analysis”, International Journal of Business Science and Applied Management, 5 (3), 1-13. Wang, M. (2009), “Manufacturing FDI and economic growth: evidence from Asian economies”, Applied Economics, 41 (8), 991-1002. Wolf, M. (2014), “Limits to growth in the 21st century”, in Rob Van Tulder , Alain Verbeke , Roger Strange (ed.) International Business and Sustainable Development (Progress in International Business Research, Volume 8) Emerald Group Publishing Limited, pp.23 – 43. Xaypanya, P.; Rangkakulnuwat, P.; Paweenawat, S.W. (2015), “The determinants of foreign direct investment in ASEAN. The first differencing panel data analysis”, International Journal of Social Economics, 42 (3), 239-250. Table 1: Summary of the main motives for FDI Motives for FDI Market-seeking

Objectives

Host country attractiveness factors

Maintain or protect markets previously supplied through exports or exploiting or promoting new markets.

25

Market size and growth; Psychic distance; Easy access to adjacent markets.

Resource-seeking

Minimizing costs and ensuring security of supply sources (e.g. demand for resources such as mineral fuels, agricultural products, among others); Reducing the costs of labour (e.g. demand for abundant supply of cheap unskilled or semi-skilled labour).

Possession of natural resources such as mineral fuels; Abundant endowment of unskilled and cheap labour; Access to skilled labour and its cost.

Efficiency-seeking

Re-organize international production (concentrating production in a few locations from the which supplies several markets) in order to take advantage of several factors such as different factor endowments, institutional frameworks, economic policies, among others.

Costs of resources such as land, raw materials or labour “adjusted for productivity of labour inputs”; Reduced transport and communication costs from and within the country; To be a member of a regional integration agreement.

Asset-seeking

Increase the investor’s overall portfolio in terms of physical assets and human skills, which it considers to contribute to maintain or strengthen its specific advantages or weaken the ownership specific advantages of its competitors.

Competition policy, particularly with regard to mergers and acquisitions; Several created assets (e.g. technological, managerial, among others); Availability of ports, roads and other physical infrastructure.

Source: Adapted from Dunning and Lundan (2008, p. 325)

Table 2: Factor endowments in natural resources and FDI – summary of empirical studies FDI targeta (period of analysis)

Proxy

Method for data analysis

Effect

Author(s) (year)

Eurasia (1993-1998)

Dummy variable (=0 if the country is poorly endowed in natural resources; =1 if moderate; =2 if high)

OLS regression

+

Deichmann et al. (2003)

Dummy variable (=1 if the country is classified as resource rich and zero otherwise)

GLS regression

+/0

Ezeoha and Cattaneo (2011)

Dynamic panel data

+/0

Ajide and Raheem (2016)

Pooled OLS and Panel data

+/0

Mhlanga et al. (2010)

30 SSA countries (1995-2008)

15 ECOWAS (2000-2013) 1,320 projects across 14 SADC countries (1994–2005)

Dummy variable (=1 if the country is classified as resource rich and zero otherwise) Investments in the extractive industry and utility provision (water, gas, electricity) (dummy)

Ex-USSR (1996-2005)

Production index oil and gas

Spatial autoregressive model

+

Ledyaeva (2009)

41 African countries (1998–2007)

Production of crude oil

Panel data

+

Sanfilippo (2010)

Total oil supply (production of crude oil, natural gas plant liquids, and other liquids, and refinery processing gain)

Dynamic panel data

+

Aziz and Mishra (2016)

12 MENA + 24 Developing countries (1975-2006)

Exports of mineral fuels (in log)

Panel data

+

Mohamed and Sidiropoulos (2010)

50 largest host countries (1991-2005)

Share of raw material exports (including fuels, ores and metals) to its total merchandise exports

Panel data

+

Cheung and Qian (2009)

22 SSA countries

Share of mineral fuels in total

Panel data

+

Asiedu (2006)

16 Arab economies (1984 to 2012)

26

(1984-2000)

exports

112 developing countries (1982-2007) BRICS (Brazil, Russia, India, China and South Africa) (2000-2009) 99 developing countries (1984 to 2011)

Dynamic panel data

-

Asiedu and Lien (2011)

Panel unit-root test, and multiple regressions

-

Jadhav (2012)

Dynamic panel data

-

Asiedu (2013)

Legend: + positive effect & statistically significant; - negative effect & statistically significant; 0 effect not statistically significant. SSA - Sub-Saharan Africa; MENA - Middle East and North Africa; SADC - Southern African Development Community; ECOWAS - Economic Community of West African States. Notes: a Country is the analysis unit for all studies cited. Source: Compiled by the authors.

Table 3: Descriptive statistics and unit roots tests of the variables Group of determinant

Variable

Proxy

Mean

St. deviation

Min

Max

FDI

Net FDI inflows (in % of the GDP)

4.1

6.065

-16.1

87.4

NRER exports (% of merchandise exports)

16.3

25.648

0.0

99.7

Proven NRER

35458.0

177226.1

0.0

2812635.0

Human capital

Mean years of schooling

6.5

1.762

0.6

13.1

Market size

GDP per capita, PPP (constant 2011 international $)

14550.1

15749.6

369.8

95751.3

Market growth

GDP per capita growth (annual %)

2.7

4.802

-18.9

92.4

Trade openness

Trade (% of GDP)

80.3

49.370

15.6

450.0

16.5

138.697

-27.0

8.6

6.049

1417.7

Natural factors Endowments

Economic instability

Economic and policy

Production costs

Tax rate

Infrastructure

Institutions

Institutional quality

LLC -5.399*** (0.000)

Unit root tests HT Time trend -26.737*** Not included (0.000)

3.869 (0.999)

-15.704*** (0.000)

Not included in HT

-63.883*** (0.000) -20.213*** (0.000)

-1.256 (0.104) -0.073 (0.4711)

Not included in LLC

-21.481*** (0.000)

5.754 (1.000)

-6.469*** (0.000) 7.3e+14 (1.000)

-38.693*** (0.000) -4.374*** (0.000)

5399.5

-8.509*** (0.000)

-35.442*** (0.000)

Not included

0.6

39.3

-1.987** (0.024)

-6.032*** (0.000)

Not included

966.026

64.4

8525.0

-0.451 (0.326)

-10.904*** (0.000)

Included

50.1

35.354

7.4

339.1

2.9e+13 (1.000)

-2.076** (0.019)

Not included in HT(1)

19.0

19.261

0.0

74.8

-15.919*** (0.000)

9.789 (1.000)

Not included in LLC

15.1

11.757

0.0

98.4

-11.550*** (0.000)

-14.386*** (0.000)

Not included in HT

Control of corruption

0.0

1.039

-2.1

2.6

-2.794*** (0.003)

-5.979*** (0.000)

Not included

Political stability and absence of violence/terrorism

-0.2

0.941

-3.0

1.7

-17.034*** (0.000)

-5.764*** (0.000)

Not included in HT

Inflation, GDP deflator (annual %) Unemployment, total (% of total labour force) Cost to import (US$ per container) Total tax rate (% of commercial profits) Fixed telephone subscriptions (per 100 people) Electric power transmission and distribution losses (% of output)

Not included in LLC Included

Not included Not included in HT

-2.316*** -0.717 Not included in LLC (0.010) (0.237) Note: LLC is Levin–Lin–Chu (adjusted t*), HT is Harris-Tzavalis (z). For LLC and HT, the null hypothesis: panels contain unit root, while the alternative Ha: panels are stationary. P-values in brackets. *** (**) [*] denotes significance at the 1% (5%) [10%] levels for p-values; (1) no constant and subtracting cross-sectional Rule of Law

-0.1

0.995

-2.2

27

2.0

means in the case of HT test.

Table 4: Determinants of FDI attraction: fixed effects panel data estimations (dependent variable –FDI/GDP ratio in logs) Determinan t

Factor Endowment endowments s

Economic and policy

1 NRER exports (% of merchandise exports) Proven NRER

0.023

2 **

*

0.023

3 **

*

0.026

Models B

4 **

*

0.023

5 **

*

0.023

6 **

*

0.025

1

2

3

4

5

6

**

*

(0.006) (0.006) (0.002) (0.005) (0.005) (0.002) 0.002 0.002 0.001 0.003 0.002 0.002 (0.632) (0.733) (0.888) (0.554) (0.644) (0.752) 0.075* 0.067 0.059 0.075* 0.066 0.064 0.088** 0.079* 0.072* 0.086** 0.076* 0.075* (0.089) (0.130) (0.180) (0.078) (0.122) (0.129) (0.047) (0.075) (0.100) (0.045) (0.077) (0.078)

Human capital

Mean years of schooling

Market size

GDP per capita, PPP (constant 2011 international $)

Market growth

GDP per capita growth (annual %)

Trade openness

Trade (% of GDP)

Economic instability

Inflation, GDP deflator (annual %)

0.012 0.010 0.016 0.011 0.010 0.016 0.011 0.009 0.015 0.011 0.009 0.014 (0.348) (0.435) (0.199) (0.358) (0.447) (0.212) (0.380) (0.484) (0.243) (0.385) (0.491) (0.252)

Unemployment, total (% of total labour force)

0.040** 0.041** 0.044** 0.040** 0.040** 0.044** 0.038** 0.038** 0.041** 0.038** 0.038** 0.041** (0.021) (0.021) (0.011) (0.022) 0.021) (0.012) (0.031) (0.030) (0.020) (0.031) (0.030) (0.020)

Cost to import (US$ per container)

0.005 0.003 0.012 0.003 0.001 0.009 0.008 0.006 0.014 0.006 0.004 0.012 (0.719) (0.840) (0.423) (0.830) (0.955) (0.542) (0.602) (0.715) (0.347) (0.687) (0.807) (0.435)

Production costs

Tax rate

Infrastructur e

Institutional quality

-0.011 -0.006 -0.026 0.001 0.006 -0.009 -0.002 0.004 -0.014 0.007 0.012 -0.001 (0.690) (0.834) (0.342) (0.973) (0.815) (0.722) (0.952) (0.884) (0.613) (0.796) (0.631) (0.969) 0.131** 0.137** 0.137** 0.127** *

*

*

0.133** 0.133** *

*

0.125** 0.131** *

*

0.131** *

0.122** 0.127** *

*

0.127** *

(0.001) (0.000) (0.000) (0.001) (0.000) (0.000) (0.001) (0.001) (0.000) (0.001) (0.001) (0.001) 0.176** 0.176** 0.174** 0.176** 0.175** 0.174** 0.188** 0.188** 0.186** 0.188** 0.187** 0.186** *

*

*

*

*

*

*

*

(0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000)

0.084** 0.089** 0.075** 0.083**

*

0.000)

*

*

*

(0.000) (0.000) (0.000)

0.084** 0.090** 0.076** 0.084** 0.089** 0.075** ** 0.073 * * * * * * * * * * * (0.012) (0.004) (0.002) (0.010) (0.004) (0.002) (0.004) (0.002) (0.009) (0.004) (0.002) (0.010)

Fixed telephone subscriptions (per 100 people)

0.010 0.009 0.018 (0.603) (0.639) (0.342)

Electric power transmission and distribution losses (% of output) 0.106** (0.018)

Political stability and absence of violence/terroris m

F-test (p-value) R square

0.088**

0.004 0.004 0.012 (0.811) (0.840) (0.517)

0.017 0.019 0.019 (0.218) (0.175) (0.166) 0.102** (0.023) 0.008 (0.632)

Rule of Law

Goodness of fit

*

Total tax rate (% of commercial profits)

Control of corruption

Institutions

Models A

Proxy

0.016 0.017 0.017 (0.256) (0.209) (0.209) 0.106** (0.018)

0.007 (0.682)

0.103** (0.021) 0.006 (0.719)

0.005 (0.766)

0.186**

0.180**

0.173**

*

*

*

0.169** *

(0.000)

(0.000)

(0.000)

(0.000)

15.39 14.86 16.33 15.51 15.02 16.43 14.66 14.12 15.39 14.78 14.27 15.51 (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) 0.408

0.442

0.446

0.444

0.442

0.446

0.444

0.440

0.443

0.442

0.440

0.443

Hausman test (p- 29.96 30.31 33.22 29.33 30.86 35.56 29.14 25.53 31.91 28.42 26.05 30.55 value) (0.002) (0.001) (0.001) (0.001) (0.001) (0.000) (0.001) (0.008) (0.014) (0.003) (0.006) (0.001) Note: *** (**)[*] estimates are statistically significant at 1% (5%) [10%]. Grey cells highlight the statistical significant estimates. P-values are in brackets. All variables are in logs. All models include a constant term. Number of observations: 2250 (125 countries*18 years)

Table A1: List of countries in the sample

28

Albania

Dominican Republic

Latvia

Russian Federation

Algeria

Ecuador

Lebanon

Rwanda

Angola

Egypt, Arab Rep.

Lesotho

Saudi Arabia

Argentina

El Salvador

Lithuania

Senegal

Armenia

Ethiopia

Macedonia, FYR

Serbia

Australia

Finland

Madagascar

Sierra Leone

Austria

France

Malawi

Singapore

Azerbaijan

Georgia

Malaysia

Slovak Republic

Bangladesh

Germany

Mali

Slovenia

Belarus

Ghana

Mexico

South Africa

Belgium

Greece

Moldova

Spain

Benin

Guatemala

Mongolia

Sri Lanka

Bolivia

Guinea

Morocco

Sweden

Bosnia and Herzegovina

Haiti

Mozambique

Switzerland

Botswana

Honduras

Nepal

Syrian Arab Republic

Brazil

Hong Kong SAR, China

Netherlands

Tanzania

Bulgaria

Hungary

New Zealand

Thailand

Burkina Faso

India

Nicaragua

Togo

Cameroon

Indonesia

Niger

Tunisia

Canada

Iran, Islamic Rep.

Nigeria

Turkey

Central African Republic Ireland

Norway

Uganda

Chad

Israel

Oman

Ukraine

Chile

Italy

Pakistan

United Kingdom

China

Japan

Panama

United States

Colombia

Jordan

Papua New Guinea

Uruguay

Congo, Dem. Rep.

Kazakhstan

Paraguay

Uzbekistan

Congo, Rep.

Kenya

Peru

Venezuela, RB

Costa Rica

Korea, Rep.

Philippines

Vietnam

Cote d'Ivoire

Kuwait

Poland

Yemen, Rep.

Croatia

Kyrgyz Republic

Portugal

Zambia

Czech Republic

Lao PDR

Romania

Zimbabwe

Denmark Table A2: Model to be estimated – summary of variables and their proxies Determinant

Dependent variable

Natural Endowments factors

Variable

Description

Source

Foreign direct investments (FDI) net inflows in percentage of the GDP

Foreign direct investment are the net inflows of investment (new investment inflows less disinvestment) to acquire a lasting management interest (10 percent or more of voting stock) in an enterprise operating in an economy other than that of the investor.

World Bank, World Development Indicators [International Monetary Fund, International Financial Statistics and Balance of Payments databases, World Bank, International Debt Statistics, and World Bank and OECD GDP estimates.]

Fuel minerals (coal, oil and natural gas) exports (% of merchandise exports)

Fuels comprise SITC section 3 (mineral fuels, lubricants and related materials).

International Trade Centre

Proven reserves

Non-Renewable Energy Resources (Crude Oil +Natural Gas+ Coal) proved reserves in tonne of oil equivalent. Proved reserves of crude oil/Gas/Coal are the estimated quantities

Expected effect on FDI

Positive

29

International Energy Statistics, in http://www.eia.gov/beta/

Human capital

Market size

Mean years of schooling

GDP per capita, PPP (constant 2011 international $)

which geological and engineering data demonstrate with reasonable certainty to be recoverable in future years from reservoirs under existing economic and operating conditions. Average number of years of education received by people ages 25 and older, converted from educational attainment levels using official durations of each level. GDP per capita based on purchasing power parity (PPP). PPP GDP is gross domestic product converted to international dollars using purchasing power parity rates. Data are in constant 2011 international dollars.

Barro and Lee (2013), UNESCO Institute for Statistics

Positive

World Bank, World Development Indicators [World Bank, International Comparison Program database]

Positive

World Bank, World Development Indicators [World Bank national accounts data, and OECD National Accounts data files] World Bank, World Development Indicators [World Bank national accounts data, and OECD National Accounts data files]

Market growth

GDP per capita growth (annual %)

Annual percentage growth rate of GDP per capita based on constant local currency.

Openness of economy

Trade (% of GDP)

Trade is the sum of exports and imports of goods and services measured as a share of gross domestic product.

Inflation, GDP deflator (annual %)

Inflation as measured by the annual growth rate of the GDP implicit deflator shows the rate of price change in the economy as a whole. The GDP implicit deflator is the ratio of GDP in current local currency to GDP in constant local currency.

World Bank, World Development Indicators [World Bank national accounts data, and OECD National Accounts data files.]

Negative

Unemployment, total (% of total labour force)

Unemployment refers to the share of the labour force that is without work but available for and seeking employment.

World Bank, World Development Indicators [International Labour Organization, Key Indicators of the Labour Market database]

Negative

World Bank, World Development Indicators [World Bank, Doing Business project]

Negative

World Bank, World Development Indicators [World Bank, Doing Business project]

Negative

World Bank, World Development Indicators [International Telecommunication Union, World Telecommunication/ICT Development Report and database.]

Positive

Economic instability

Economic and policy Production costs

Cost to import (US$ per container)

Tax rate

Total tax rate (% of commercial profits)

Fixed telephone subscriptions (per 100 people) Infrastructure

Cost measures the fees levied on a 20-foot container in U.S. dollars. All the fees associated with completing the procedures to export or import the goods are included. Total tax rate measures the amount of taxes and mandatory contributions payable by businesses after accounting for allowable deductions and exemptions as a share of commercial profits. Fixed telephone subscriptions refers to the sum of active number of analogue fixed telephone lines, voice-over-IP (VoIP) subscriptions, fixed wireless local loop (WLL) subscriptions, ISDN voice-channel equivalents and fixed public payphones.

Positive

Positive

Electric power transmission and distribution losses include losses in transmission between World Bank, World sources of supply and points of distribution Development Indicators [IEA Negative and in the distribution to consumers, including Statistics © OECD/IEA 2014] pilferage. Control of Corruption captures perceptions of the extent to which public power is exercised Control of World Bank, Worldwide for private gain, including both petty and Positive corruption* Governance Indicators grand forms of corruption, as well as "capture" of the state by elites and private interests. Political Stability and Absence of Political stability Violence/Terrorism measures perceptions of World Bank, Worldwide Institutional and absence of the likelihood of political instability and/or Positive Institutions Governance Indicators * quality violence/terrorism politically-motivated violence, including terrorism. Rule of Law captures perceptions of the extent to which agents have confidence in and abide by the rules of society, and in particular the World Bank, Worldwide Rule of Law* Positive quality of contract enforcement, property Governance Indicators rights, the police, and the courts, as well as the likelihood of crime and violence. Notes: * Estimate gives the country's score on the aggregate indicator, in units of a standard normal distribution, i.e. ranging from approximately -2.5 to 2.5. Source: Compiled by the authors. Electric power transmission and distribution losses (% of output)

30

Table A3: Correlation matrix

FDI

Group of determin ant

Proxy

Net FDI inflows (in FDI % of the GDP) Fixed telephone subscription s (per 100 people) Infrastru Electric cture power transmissio n and distribution losses (% of output) Mean years Human of capital schooling Inflation, Economi GDP c deflator instabilit (annual %) y Unemploy ment, total (% of total labour Producti force) on costs Cost to import (US$ per container) Control of corruption Political Institutio stability and absence of nal violence/ter quality rorism Rule of Law Total tax Financial rate (% of and tax incentive commercial profits) s GDP per capita, PPP (constant Market 2011 size internationa l $) GDP per capita Market growth growth (annual %) Openness Trade (% of of GDP) economy

Net FDI inflo ws (in % of the GDP )

Infrastructure

Fixed telephon e subscrip tions (per 100 people)

Electric power transmis sion and distribut ion losses (% of output)

Hum Econo an mic Production costs capit instabi al lity

Mean years of school ing

Inflatio n, GDP deflator (annual %)

Unemploy ment, total (% of total labor force)

Cost to import (US$ per contai ner)

Institutional quality

Contro l of corrup tion

Political stability and absence of violence/ter rorism

Rule of Law

Finan cial and tax incent ives

Mar Marke ket t size grow th

Open ness of econo my

Factor endowm ents

Total tax rate (% of comme rcial profits)

GDP per capita, PPP (constan t 2011 internati onal $)

Trade (% of GDP)

NRER exports (% of merchand ise exports)

GDP per capita growt h (annu al %)

1.00 0

0.14 0

1.000

0.04 0

-0.464

1.000

0.19 2

0.806

-0.379

1.000

0.03 1

-0.178

0.103

0.082

1.000

0.04 8

0.191

-0.009

0.256

0.051

1.000

0.05 0

-0.367

0.206

0.303

0.091

0.000

1.000

0.13 5

0.699

-0.518

0.501

-0.317

-0.014

0.400

1.000

0.15 5

0.511

-0.260

0.414

-0.287

0.001

0.242

0.649

1.000

0.14 0

0.711

-0.512

0.549

-0.338

0.013

0.431

0.928

0.698

1.00 0

0.13 0

-0.124

0.057

0.184

0.206

-0.015

0.162

0.223

-0.270

0.29 1

0.13 4

0.897

-0.490

0.760

-0.212

0.089

0.429

0.738

0.514

0.74 -0.226 3

1.000

0.15 1

0.023

0.037

0.057

-0.053

0.032

0.028

0.013

0.071

0.00 -0.018 3

-0.019

1.00 0

0.42 2

0.215

-0.075

0.270

-0.030

0.045

0.183

0.166

0.261

0.18 -0.256 4

0.239

0.08 4

31

1.000

1.000

Factor endowme nts

NRER exports (% of merchandis e exports)

0.02 0

0.102

0.042

0.09 0.335 -0.189 6 Note: Grey cells highlight very high correlations. Proven NRER

0.117

0.113

0.098

0.044

0.137

-0.135

0.15 6

0.086

0.213

0.00 9

0.060

1.000

0.327

0.045

0.099

0.111

0.133

0.058

0.17 -0.046 9

0.326

0.03 5

-0.193

0.250

32