The Effect of Corruption on Foreign Direct Investment, the Reverse Causality and the Role of Institutions in the Association of Southeast Asian Nations: An
Empirical Investigation
QuickTime™ and a TIFF (ongecomprimeerd) decompressor
are needed to see this picture.
Author: Marie de Graaf s1322397 Groningen, October 2006 University of Groningen Faculty of Economics
MSc. International Economics and Business Supervisors:
Table of contents:
Abstract page 4
1. Introduction page 5
2. Literature review page 8
- 2.1 Foreign Direct Investment(FDI) page 8 - 2.2 Ownership, Location and
Internalization(OLI) framework page 9
- 2.3 Institutions page 10
- 2.4 Corruption page 13
- 2.5 Effects of corruption on FDI page 16 - 2.6 Effects of FDI on corruption page 18 - 2.7 Other effects on corruption page 19 - 2.8 Country descriptions page 22
3. Hypotheses page 27
4. Methodology page 28
5. Data page 30
6. Empirical findings page 35
7. Limitations, future research and conclusion page 37
References page 40
If you want to buy a Sherman tank, a Red Cross blanket, or simply speed up the installation of a telephone, there is probably no easier place in the world in which to do just that than in Asia – if you are willing to part with some cash, that is. With pathetically few exceptions, the countries in this region are so riddled with corruption that the paying of ‘tea money’ has become almost a way of life. (Quah, 1999)1
There remains a significant degree of ambivalence among many policy makers about the real impact of corruption on the economy. This is due in part to the so-called East Asian puzzle: high rates of growth have been sustained over a long period despite high levels of corruption in a number of East Asian countries (Campos et al., 2001 )
This paper is motivated by the paradox in East Asia: countries have high rates of investment together with high levels of corruption.
Abstract
1 Introduction
Foreign Direct Investment (FDI) was viewed with some wariness by developing countries up until the 1980’s (Mukim, 2005). The developing countries were afraid that large Multinational Enterprises (MNEs) would control large resources that would consequently give them considerable influence over economic and political affairs. However, FDI is now considered as a substantial part of the development process of these countries and governments have relaxed their policies in order to attract FDI. FDI has also become an important source for economic development for countries in South East Asia. FDI became even more important as a result of the Asian financial crises in 1997. One of the causes of the Asian Financial Crisis was the disproportionately high short-term financing of current-account deficits (Mukim, 2005). Developing countries were persuaded to implement investment policies to attract FDI due to the drying-up of commercial bank lending caused by the crises (Mukim, 2005). The change in the attitudes of these governments towards MNEs has improved the policies and investment regulations in favour of MNEs.
Not only do governments try to enhance FDI inflows by improving policies at a national level, Asian countries also combine their efforts in the Association of Southeast Asian Nations (ASEAN)2. One objective of ASEAN is to accelerate economic growth by enhancing individual country and regional competitiveness through cooperation. One way they try to establish economic growth is by attracting FDI (Mirza and Giroud, 2004).
Despite the success of ASEAN to attract FDI, countries like Vietnam, Indonesia and Thailand have, according to Transparency International (TI)3, a high level of corruption. Corruption has been recognized as an urgent international problem and many studies have shown that it has harmful effects on economic growth and development (King, 2003). Investors might be discouraged to invest in a specific country due to its high level of corruption. High growth rates along with high levels of corruption is referred to as the East Asian puzzle (Campos et al, 2001).
2 More information about ASEAN is presented in section 2.8
3 Transparency international is a non-governmental organization with the aim of curbing corruption. It
The focus of this research paper will be on the effect of corruption on FDI and vice versa. Corruption is defined as the abuse of public office for private gain4. It arises when there is a lack of efficient market institutions which is caused by weak government regulations (Johnson, 2004). Lederman, Noayza and Soares (2005) claim that corruption is regarded as one of the most serious obstacles to development. One channel through which corruption hinders growth is its impact on FDI (Hellman, Jones and Kaufman, 2002). Until 1995 most studies only suggested that there was a negative effect. Hines (1995) was the first who reported empirical results showing a negative effect of corruption in the host country on the FDI levels of different countries. Later Wei (2000), Lambsdorff and Cornelius (2000), Wei and Smarzynska (2002) and Hellman et al. (2002) came with similar results after their empirical investigations. There are also opposing views that state that corruption is efficiency-enhancing; it speeds up the business process by overcoming burdensome regulations (Merton, 1957; Lui, 1985; Werner, 1989; Shleifer and Vishy ,1994).
Some studies have also focused on the effects of FDI on corruption. North (1990) was the first who suggested that organizations might affect institutions. Larrain and Tavares (2004) found that FDI inflows are significantly negatively related to corruption. MNEs are also expected to decrease the level of corruption in the host country in this paper. The quality of institutions also seems to affect the level of corruption (Dreher, Kotsogiammis, McCorrinstion, 2005). It is argued that high corruption levels seem to be the result of not having a good institutional framework. The focus will be on members of the Association of Southeast Asian Nations (ASEAN). Data from six ASEAN countries are used, namely Indonesia, Malaysia, the Philippines, Thailand, Vietnam and Singapore. ASEAN countries are attractive countries for FDI, even though it is widely known that some ASEAN counties still have high corruption levels. This makes the ASEAN group an interesting area to focus on. Another point to mention is that there is a lack of empirical evidence to assess the impact that corruption has on FDI, and vice versa, in the ASEAN countries.
4 Definition by Senturia, (1931). It is frequently used in scientific studies and used by the World Bank
2 Literature review
Foreign direct investment (FDI) and the factors that influence the level of FDI inflows, the role of institutions and corruption will be dealt with subsequently in this literature review. Then the effect corruption has on FDI will be discussed. The literature also points out that there is a reverse causality, although little empirical research has been done testing this relationship. Other factors influencing the level of corruption are discussed subsequently. The section concludes by describing the various country characteristics of the ASEAN countries.
2.1 Foreign direct investment (FDI)
FDI refers to the acquisition or building of production units in a foreign country into domestic structures, equipment and organizations (Gilman, 1981). The growth of MNE activity in the form of FDI has grown at a faster rate than most other international transactions like trade flows (Blonigen, 2005). FDI is attractive since it enhances capital formation, increases employment, it can promote manufacturing exports and it can bring special resources such as capital, managerial skills and knowledge flows which will result in a positive contribution to economic growth (Grossman and Helpman, 1991; Balasubramanyam, Salisu and Sapsford, 1996); Markusen and Venables, 1999). Empirical studies have shown the positive effect of FDI on economic growth (Baharumshah and Thanoon, (2006)). Most developing countries and former centrally-planned economies have recognized the positive effects of FDI on economic growth and are eager to attract FDI (Wei and Smarzynska, 2002). FDI has become very popular in these countries since they cannot finance their investment needs because for example domestic savings might not be sufficient or they might have insufficient foreign exchange to transform domestic into foreign resources (Drabek, 1999; Mukim, 2005). FDI has also played an important role in the rapid economic development of the newly industrializing and developing economies of the ASEAN countries (Baharumshah and Thanoon, 2006).
divided in market seeking and efficiency seeking. The firm will avoid costs associated with cross-border trade by supplying the market directly through an affiliate, which is called horizontal foreign direct investment (HFDI). The firm can also choose to engage in vertical foreign direct investment (VFDI) by exploiting international differences in factor prices by splitting up the production process (Navaretti and Venables, 2004). For the first type of FDI market size and high tariffs are important determinants. For VFDI production cost-minimizing is important, such as low labour costs or natural resources. These two theories can give some contradicting predictions. An MNE will invest horizontally in order to avoid high trade costs. However, an MNE that wants to invest vertically in order to take advantage of low production costs is unable to avoid the high trade costs and will therefore be less likely to invest when the trade costs are high. So the theory of HFDI predicts that FDI increases with trade costs whereas VFDI theory states that it decreases with trade costs. Another example is that HFDI substitutes trade, whereas VFDI creates trade. There are also factors that affect both types of FDI, for example the economic and political stability and the infrastructure.
2.2 OLI framework
2.3 Institutions
describes corruption as: ‘A symptom of deep institutional weaknesses and it leads to inefficient economic, social, and political outcomes.’
MNEs get into contact with host-country institutions when they start their business in a foreign economy and play a large role in the continuous operations of the MNEs (Johnson, 2004). When developed institutions are lacking there will be arbitrariness in enforcing the rules. Bureaucrats will interpret the regulations and laws to their advantage and will extract monetary gains. So when there is a weak institutional framework in the foreign country MNEs will also have to deal with the existence of corruption.
Institutions are present at local, provincial, national level and are becoming increasingly important at the international level. Governments combine their efforts to increase international trade and to combat corruption since a country’s government might be unable to do so. International institutions like the World Trade Organization (WTO), Association of Southeast Asian Nations (ASEAN) and Asia-Pacific Cooperation (APEC) initiate agreements to increase the investment relations between countries. One aspect of this is also to recognise the negative effect of corruption and take action.
The WTO describes itself as the only global international organization dealing with the rules of trade between nations.5 It replaced the General Agreement on Tariffs and Trade (GATT) after the Uruguay Round in 1995. WTO agreements and WTO-related rules are set up to make it easier for investors and to encourage them to do business abroad. Foreign investors are protected through the General Agreement on Trade and Services (GATS), the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) and by the Agreement on Trade-Related Investment Measures (TRIMS). Another advantage is that being a WTO member increases the ‘psychological’ costs of deviation form international agreements; countries feel more obliged to keep agreements (Furusawa, 2003). Furusawa (2003) concludes that safeguards policy and continuing WTO trade negotiation rounds facilitate international cooperation. WTO members might thus be less likely to engage in opportunistic behaviour when doing business with each other.
ASEAN aims to play a leading role in accelerating economic growth and promoting regional peace and stability in South East Asia6. The ASEAN Free Trade Area (AFTA) was initiated at the fourth ASEAN Summit in Singapore in 1992. AFTA’s purpose is to strengthen intra-ASEAN trade and investment relations. One way they try to accomplish this is by eliminating tariff and non-tariff barriers within the region to stimulate firms to export goods and services to other ASEAN countries. The ASEAN Investment Agreement (AIA) was implemented in 1998 as an extension of AFTA. The agreement provides transparent investment policies and investment barriers will be eliminated for ASEAN investors by 2010 and for all in 2020. Its basic idea is that the measures in the agreement will increase FDI inflows. Some countries which joined ASEAN are also a member of Asia-Pacific Cooperation (APEC) which are; Brunei, Indonesia, The Philippines, Malaysia, Singapore and Thailand7. Its goal is to promote trade and economic cooperation between the countries of East Asia and the Pacific. APEC leaders have also agreed that corruption deters FDI. The first meeting of APEC Anti-Corruption Experts was in Santiago on 25-26 September 20048.
Another international institution that has paid attention to the fight against corruption is the Organisation for Economic Co-operation and Development (OECD). The OECD has played this role since 1989. It also assists non-member countries in improving their governance and anti-bribery standards through a number of outreach activities9. In cooperation with the Asian Development Bank (ADB) it has launched
the Anti Corruption Initiative for Asia-Pacific at a conference held in Seoul, Korea in 2000. Its 27 members including Indonesia, Malaysia, Philippines, Singapore, Thailand and Vietnam are committed to cooperate and find ways to fight corruption. The initiative assists countries in developing and promoting anti-corruption policies at national and regional levels.10 They have adopted an anti-corruption action plan in 2001 in Tokyo. It defines the participating countries' objectives in building sustainable legal and institutional frameworks to fight corruption.
The World Bank (WB) is also working to address corruption and strengthen fundamental institutions of good governance. The World Bank Institute (WBI), in collaboration with many units in the World Bank Group does not only focus at the
6 http://www.aseansec.org/64.htm
7 http://www.apecsec.org.sg/content/apec/member_economies.html
8 http://www.apec.org/content/apec/apec_groups/som_special_task_groups/anti-corruption.html 9 http://www.oecd.org/about/0,2337,en_2649_37447_1_1_1_1_37447,00.html
global level to combat corruption but also at country level and by means of Bank-funded projects. It is helping countries to build transparent and accountable institutions and design and implement anti-corruption programs at country level. The projects are focussed on investigating allegations of fraud and corruption.11 It focuses on joint initiatives at the global level by expanding partnerships with multilateral and bilateral development institutions. The WBI anti-corruption activities started in 1994. Institutions contribute to the location advantages attracting FDI. Having a well functioning institutional framework will attract FDI. Corruption is seen as a determinant for a malfunctioning institutional framework and as a ‘location disadvantage’ for attracting FDI.
2.4 Corruption
Sullivan (1996) states that the role of the government is to regulate the creation and maintenance of the market system through establishment of laws that protect individuals and firms from corrupt practices. A weak government is unable or unwilling or assume its responsibilities. These government failures lead to broader failures in political, economic and civic institutions and this creates opportunities for crime and corruption (OECD, 2006). A high level of corruption as a result of not having a good institutional framework due to an irresponsible government goes hand in hand with corrupt practices in market institutions.
Bhagwati (1982), Krueger (1974) and Susan Rose-Ackerman (1978) were among the first who contributed to acquiring knowledge through theoretical research on the causes of corruption (Mauro, 1998). One of their main findings was that corruption could occur where rents exist as a result of government regulations and where public officials have discretion in allocating licences (Mauro, 1998). Klitgaard (1988) has developed a simple model to explain the level of corruption: C (Corruption) = M (Monopoly Power) + D (Discretion) – A (Accountability). According to Klitgaard (1988), the extent of corruption depends on the monopoly power, the discretional power that an official exercises and the degree of accountability. Monopoly power can be large in highly regulated economies and discretionary power can be high in developing countries and transition economies where administrate rules and
regulations are often poorly defined (Kindra, 1998). Kindra (1998) states that accountability may be weak when there are poorly defined ethical standards of public service, weak administrative and financial systems and ineffective watchdog agencies. Transactions within the government always imply some asymmetry of information between the parties involved. The possibility of rent extraction and the precise nature of informational problems depend largely on the institutional design (Lederman, 2005). An example of government regulations where rents are available and thus where corruption can occur that might affect MNEs are trade restrictions (Mauro, 1998). Examples of trade restrictions are tariffs and import quotas. When an import quota is imposed, government officials may give the associated licenses to the entrepreneurs who are willing to pay the highest bribes. Other examples of government regulations where rents are available are favouritism in industrial policies (subsidies and tax deductions), price controls, multiple exchange rate practices and foreign exchange allocation schemes (Mauro, 1998).
Corruption is seen as a primary impediment to economic growth (World Bank)12. It is regarded as one of the most serious obstacles to development (Lederman et al, 2005) and it has been blamed for the failure of certain countries to develop (Treisman, 2000). However, this does not mean that very corrupt countries cannot experience high FDI inflows. Vietnam has had relative high FDI inflows (1610,1 US$ million in 2004)13 even though it is seen as a corrupt country. More corrupt countries, however, tend to have lower investment and growth levels and larger informal sectors (Alesina and Weder 1999).
Mauro (1995) and Burki and Perry (1998), Kaufman and Kraai (2002) found that corruption reduces economic growth via reduced private investment. Corruption results in higher costs since entrepreneurs might be asked for bribes before entering the market or corrupt officials might request a share in the proceeds of their investments (Mauro, 1998). Not only does corruption impede economic growth by reducing private investment. Another way corruption may harm economic performance is by distorting the composition of government expenditure (Mauro, 1998); corrupt politicians might spend more public resources on those items on which it is easier to exact large bribes and where it is easy to keep it secret. Mauro (1998)
12http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTPUBLICSECTORANDGOVERNAN
CE/EXTANTICORRUPTION/0,,menuPK:384461~pagePK:149018~piPK:149093~theSitePK:384455, 00.html
argues that they will get the chance to do this in markets where the degree of competition is low and where the value of items is difficult to trace. As a result, officials will engage in rent-seeking rather than productive activities and will thus distort the composition of government expenditure
However, there are scientists who argue that corruption is beneficial and less time consuming than a business process would have been if it had been done in a legal way. Merton (1957) argued that corruption is the necessary grease to move the slow-moving bureaucracies in developing countries. Leff (1964) and Huntington (1968) also claimed that corruption can be efficiency-enhancing because it removes government imposed rigidities. Lui (1985) stated that bribing could speed up the service when bribery is allowed since it minimizes the average value of the time to do business. There are also scientists who argue that corruption might result in more productive government employees (Mauro, 1998). The employees will benefit from bribe-taking when they work harder to be able to procure more bribes. Corruption might be the only way to encourage investment by offering alternative ways to conduct business in developing countries where the government is ineffective (King, 2003). These “grease-money” theories were often used to explain the high growth rates in Southeast Asia even though these countries had high levels of corruption (Tanzi, 1998). However, these theories forget about the extra costs that have to be incurred to speed up the business process.
One point to be cautious about is the fact that corruption in one place may not be seen the same way in another due to the varying norms and morals accepted by society in different countries (King, 2003). Firms from different countries may face different costs due to their cultural background. Corruption in corrupt countries will not be seen as extra costs by its domestic firms but as part of the business process. Some countries also have strict anti-corruption laws and firms from these countries will be less likely to engage in corrupt behaviour.
Mauro (1995) was the first to provide an empirical treatment on the relationship between corruption and growth. Several corruption indexes became available. Business International (BI), which is a subsidiary of Economist Intelligence Unit14, has conducted a measurement of corruption during 1980-1983, which is based on surveys. It is an integer from zero (most corrupt) to ten (least corrupt) according to the
degree to which business transactions involve corruption or questionable payments. A second measure was composed by the International Country Risk Group (ICRG)15, having a scale between zero (most corrupt) and six (least corrupt). Transparency International (TI)16 also compiled a measure that is called the Corruption Perception Index (CPI). It scales from zero (most corrupt) to ten (least corrupt). Previous relevant literature often used these indexes of corruption and it is proven that the different subjective indexes are all highly correlated. This supports the argument that the measures are reliable and that the same results hold across the different measures. Treisman (2000) and Lederman et al. (2005) found that several corruption indexes were positively correlated. Wei (2000a) compared the corruption indexes of TI, BI and the ICRG and stated that the results were fairly similar. An advantage of the CPI is that is measures corruption annually and therefore this index will be used in this paper.17
2.5 Effect of corruption on FDI
One channel through which corruption hinders growth is its impact on FDI (Hellman et all. 2002). A lot of studies have focused on the negative effects corruption has on economic growth and development. The effect of corruption on FDI is a relatively new area of interest (Wei, 2002). MNEs might not want to invest in a country when they believe that there is a high level of corruption. Corruption is likely to result in higher transaction costs because bribes might be necessary before entering the market or when corrupt officials request a share in the proceeds of their investments (Mauro, 1998). An MNE will be uncertain about the costs when investing in a corrupt country and might decide not to invest. Some researchers have done empirical research on the impact of corruption of FDI using quantitative data (Wei, 2002). Hines (1995) was the first who found a negative effect of corruption on FDI. He found evidence that corruption affects the growth of US controlled FDI during 1977 and 1982, controlling for the growth of the host country GDP. Lambsdorff and Cornelius (2000) come to the same conclusion using a sample of African countries. Smarzynska and Wei (2000) also found that corruption in a host country reduces the probability of an investment using firm-level data for investments in Eastern Europe and the former Soviet Union.
15 For more information: http://www.prsgroup.com/icrg.icrg.htm
Later Wei and Smarzynska (2002) also concluded that corruption has a significant negative impact on the levels of FDI using firm-level data encompassing MNEs from the US and other countries. Hellman et al. (2002) concluded that corruption reduces FDI inflows and attracts lower-quality investment in terms of governance standards. Egger and Winner (2004) studied the impact of corruption on FDI using a data set of 21 home countries investing in 59 developed and less developed host countries between 1983 and 1999. They also found a negative relationship. Campos (2001) states that corruption has a less negative impact on investments when it is more predictable. When corruption is relatively well organized it becomes more predictable and MNEs will not have to deal with uncertain costs.
As already mentioned there are opposing views as well. Merton (1957), Lui (1985), Shleifer and Vishy (1994) state that corruption greases the wheels of commerce. They share the view that in the presence of inefficient institutions, corruption may facilitate entrepreneurship, especially when the associated payment and private benefit are predictable. Werner (1989) found that corruption encourages FDI by allowing a mechanism to overcome burdensome regulations. Bardhan (1997) reasoned that when a country suffers from a rigid bureaucracy, bribes can speed up decision making. The contract will go to the most efficient company since it will be able and willing to pay the largest bribe because of its high profit. However, he also states, as mentioned before, that corruption will make rent-seeking more attractive than productive investments and this will impede economic growth. Another effect of this rent-seeking behaviour is that it can cause low quality of the public infrastructure and services. An infrastructure of poor quality deters FDI inflows (Mauro, 1998). .
2.6 Effect of FDI on corruption
As shown above quite some empirical work is focused on the influence corruption has on FDI. The reverse effect of FDI on corruption has been studied considerably less. North (1990) suggested that organizations might affect institutions; MNEs may decrease or increase corruption. Johnson (2004) reasoned that one way how an MNE might decrease corruption is that it could mention the existence of corruption in the host country while negotiating the conditions for investment and that the host country might react by offering tax holidays or try to compensate for the costs of corruption. The MNE will have some bargaining power if the host country government thinks the investment is going to be beneficial for the country. Kimberley (1997) concluded that foreign standards of honesty have an important impact on local officials and their behaviour. The MNE could threaten to locate its production in another country if the government does not improve the formal business regulations to decrease corruption (Johnson, 2004). Since governments see FDI as an important tool for boosting the economy this might be a possible channel. However, when corruption is not kept in check by the government, foreign investors will exit the market due to the international capital mobility (Larraín, Tavares, 2004). This might be a reason for the government to decrease corruption in order to attract FDI.
MNEs can also affect corruption after they have entered a country. Johnson (2004) states that large MNEs might have a strong position in the host economy and might be able to start controlling the local firms and government officials in the long term. He reasons that MNEs might influence the government to adopt anti-corruption policies Another way in which an MNE might decrease corruption is by bringing its own business culture to the host country. The MNE’s anti-corruption attitude could possibly influence other firms to adopt a similar point of view due to the interaction between the different business cultures of the MNE and the local firms (Johnson, 2004). Larraín and Tavares (2004) and Johnson (2004) found that FDI is a robust determinant of corruption by using qualitative data. They found that larger FDI inflows decrease national corruption.
their expected earnings are higher than domestic firms, so foreign investors can often obtain some local market power in exchange for bribes (Johnson, 2004). It makes economic sense to bribe officials today and recover the cost later on by charging higher prices (Larrian and Tavares, 2004). Kaufman (2004) found evidence that MNEs headquartered in the OECD exhibit practices consistent with domestic firms within the non-OECD countries. Therefore even though OECD conventions are implemented, MNEs still adopt to the local bribery practices.
Larrain and Taveres (2004) found that not only FDI but also income per capita explains a substantial share of cross-country variation in corruption. Having discussed the potential effects of FDI on corruption other determinants influencing the level of corruption are discussed below.
2.7 Other effects on corruption
The differences in the corruption level between countries can be explained by various factors. An official will weigh the costs of a corrupt action against its benefits. An important cost is the risk of getting caught and being punished (Treisman, 2000). The effectiveness of a country’s legal system will determine the probability of getting caught (Treisman, 2000). La Porta et al (1998) claimed that countries having a common law system (like the United Kingdom) have lower levels of corruption because the system has strong property rights. They found that countries with a French civil law system are more likely to have corrupt government officials. Treisman (2000) also found evidence that countries that have been under British rule have lower levels of corruption.
found a significant negative relationship between corruption and wage levels. However, Treisman (1999) did not find clear evidence that higher government wages reduce corruption. He suggests that this is due to endogeneity; corrupt politicians may allocate themselves high wages.
Larrain and Taveres (2004) assessed the effect of openness on corruption, using FDI as a measure of openness and found FDI to be significantly negatively related with corruption. They also found GDP per capita to be negatively associated with corruption; countries with a high income level are expected to have efficient and transparent institutions because good institutions become more ‘affordable’ and may be seen as a commodity. Treisman (1999) also came with the result that there was a strong correlation between economic development and perceived corruption.
of exposing corrupt activities such as free press. Another way how corrupt practices can be exposed is by using the internet. An easy rider18 in Vietnam mentioned the fact that corruption is decreasing due to the use of Internet. Internet is becoming more available and this increases transparency.
This paper will be focussed on the ASEAN countries. ASEAN was established in 1967 and its members were Brunei, Indonesia, Malaysia, the Philippines, Singapore and Thailand. Vietnam joined ASEAN in 1995, Laos and Myanmar in 1997 and Cambodia in 1999. One of the objectives of the ASEAN group is to promote an investment region by enhancing individual country and regional competitiveness through cooperation (Mirza and Giroud, 2004). One way they cooperate to promote the region is by lowering tariffs and non-tariff barriers. As already mentioned in section 2.3, the ASEAN Free Trade Area (AFTA) was launched in 1992 and it aims to promote the region’s competitive advantage as a single production unit. The elimination of tariff and non-tariff barriers among the countries is expected to promote economic efficiency, productivity, and competitiveness19. The ASEAN Economic Community is also committed to building a stable, prosperous and highly competitive ASEAN economic region in which there are free flows of goods, services and investment in 2020 2. Its main goal is to establish ASEAN as a single market and production base. The free-trade area will make it more attractive for MNEs to enter into the region.
18 A person who earns his money taking tourists around Vietnam on a motorcycle. The person is unable
to get a ‘proper’ job because his parents or family have fought on the side of the Americans during the war.
2.8 Country descriptions
As already mentioned, ASEAN countries cooperate in promoting its region in order to attract FDI. FDI plays an important role in the rapid economic development of newly industrializing and developing economies in the ASEAN countries (Baharumshah and Thanoon, 2005). The countries differ in the origin of the FDI inflows due to their differences in economic structure and their historical ties (Thomson, 1999). Indonesia, Malaysia, the Philippines and Thailand all have export-oriented market-based economies, they have dualistic economies with more than 50% of their populations living in the agricultural rural sector (Tongzon, 1998, p.13). Countries from the ASEAN have continued with economic development and growth through international trade and investment, and in order to establish this they changed their policies considerably. They relaxed the rules and regulations of FDI in order to attract more investments. FDI represented 40% of the net resource flows to the ASEAN countries between 1993 and 1998. Malaysia, Myanmar and Vietnam received more than 50% of the total FDI inflows (Baharumshah and Thanoon. 2005). Most ASEAN countries have suffered from the financial crisis in 1997. However they have experienced different levels of economic recovery and FDI levels. Unlike Indonesia and the Philippines Thailand and Malaysia are successful in attracting FDI (Mirza and Giroud, 2004).
Table 1: Corruption levels of ASEAN countries
Country TI 2005 CPI country rank CPI level
Singapore 5 9.4 Malaysia 39 5.1 Thailand 59 3.8 Vietnam 107 2.6 Philippines 117 2.6 Indonesia 137 2.2
Source: Transparency International: Corruption Perception Index 2005. The CPI measure is scaled from zero (most corrupt) to ten (least corrupt). 158 countries were examined during the year 2005.
Indonesia
Indonesia is very corrupt according to Transparency International, which is indicated by its CPI level of 2.2. Corruption was already a problem during the Dutch colonial period. Corruption was caused by the low wages of the Dutch East India Company personnel (Quah, 1999). This reason for corruption remained during Sukarno’s and Suharto’s rule. The Indonesian government started to curb corruption after the 1955 election by arresting people involved in corrupt practices. The government assigned a Corruption Eradication Team in 1968 and in 1970 it assigned a Commission of Four elder statesman to fight corruption (Quah, 1999). However, attempts for implementing anti-corruption plans have remained unsuccessful since corrupt officials simply ignored new regulations. Quah (1999) states that Suharto and his family became more corrupt during the 1980s and 1990s; Suharto faced a $570 million corruption charge in the fall of 2000. Shari and Einhorn (1998) claim that Indonesia will remain corrupt as long as the civil servants earn miserably low salaries.
Malaysia
Malaysia has had a low level of corruption. Malaysia became an independent state in 1963, after British and Dutch colonization. In the 1980s, Dr. Mohamad Mahathir became prime minister. He instituted economic reforms that led to enormous economic growth. He turned the country into what was later called one of the Asian Tigers.20 In 2003 Badawi succeeded Mahathir and initialized anti-corruption policies and instituted reforms from his installation. Foreign trade and investment liberalization policies have increased the FDI inflows to GDP significantly during 1970-1998 (Marwah and Tavakoli, 2004).
The country is also an oil exporter like Indonesia. The country enjoys high income levels because of its rich endowment of natural resources (Marwah and Tavakoli, 2004). Its population is relatively small.
Philippines
Corruption became evident in the Philippines during its colonial period (Quah, 1999). Civil servants had low wages and many opportunities for corruption when the Spanish ruled (Corpuz, 1957). During the American colonial period there was less corruption due to higher wages and stricter rules. However, corruption increased again after the Second World War (Quah, 1999). Despite the attempts of several presidents corruption has remained a problem in the Philippines ever since. Presidents have tried to combat corruption since Quirino (1948-1953) by setting up several commissions. However, none of these were effective. Wages of civil servants have remained too low for the commissions to have the effect that was hoped for (Quah, 1999). The Philippines’ Peso was devalued in 1970 and 1984 in order to expand exports. However, it had a negative effect on the economy. The countries infrastructure has also not been sufficient to attract FDI. The Philippines’ economic performance has lagged far behind its neighbours (Marwah, and Tavakoli, 2004).
Thailand
Corruption in Thailand arose in the sixteenth century when civil servants held revenues collected for the king for themselves (Chai-Anan, 1977). The Prime Minister appointed a Counter Corruption Committee (CCC) in 1974. However, the commission
was ineffective. This was because there was a constant conflict between the cabinet and the bureaucrats (Dalpino, 1991) and because the CCC lacked direct authority to punish public officials (Quah, 1999). As a consequence, it was given more power, however, they still could not take action against politicians (Quah, 1999). Most Thai political leaders have not been committed in the fight against corruption and anti-corruption strategies have remained inefficient.
Thailand has been a British colony from 1824 until 1896 when the Anglo-French guaranteed the independence of Thailand. The country has had a stable economic growth, low inflation rate and a manageable level of external debt since the 1960s until the financial crisis of 1997 (Marwah and Tavakoli,2004).
Singapore
During its British colonial period and the Japanese occupation corruption was a way of life (Quah, 1999). Corruption became illegal in 1871 with the introduction of the Penal Code of the Straits Settlement. The first anti-corruption law was introduced in 1937 and in 1960 the Prevention of Corruption Act (POCA) was enacted and the Corrupt Practices Investigation Bureau (CPIB) became the anticorruption agency responsible for the POCA’s provision (Quah, 1999). From that time more amendments were made in order to strengthen the POCA further. The improvement of wages began in 1972. The low level of corruption nowadays can be explained by the reduced incentives for corruption; fines and improved salaries (Quah, 1999). Singapore is seen as the third most attractive destination for FDI in Asia, behind China and Hong Kong. Its ability to attract FDI is due to its highly supportive business environment. One key element in this business environment is the good infrastructure; it has the busiest container port in the world and one of the best connected airports in Asia1. In addition, its workforce is highly skilled with low incidence of labor disruption. Another important factor that stimulates the business environment is the stable political and social environment.21
Vietnam
According to Transparency International Vietnam is one of the most corrupt countries in Asia. In January 1995 the General Secretary Do Moi stressed the rising corruption and deteriorating morality of state officials. Combating corruption was a major theme of the 1998 and 1999 Party Plenums, producing a number of ordinances and degrees to solve the problem, and an official anti-corruption campaign was launched in 2000.22Largely due to a lack of transparency, accountability and media freedom, widespread official corruption and inefficient bureaucracy remain serious problems. Even the Communist Party and the Government of Vietnam admit they must address corruption squarely and soon (Political Risk yearbook 2005).
Vietnam has placed its priority on using international trade and foreign investment for its economic development objectives since it launched the open-door policy of ‘Doi Moi’ in 1986 (Freeman, 2001). It has gone through a major economic transition process. However, weaknesses in the formal and informal institutions remain major obstacles to business (Tenev et al, 2003). It has been quite successful in attracting FDI since it initiated the Foreign Investment Law in 1987 (Freeman, 2001). Vietnam became a member of the ASEAN in 1995. It joined ASEAN because it desired to share the wealth of the rapidly emerging region. Vietnam worked with the other members to encourage FDI to and within the region and was the third largest recipient of FDI flows in ASEAN in 2002 (Mirza and Giroud, 2004). Vietnam is attractive because of its large population, political stability and diversified industrial base
(Mirza and Giroud, 2004).
22http://unpan1.un.org/intradoc/groups/public/documents/APCITY/UNPAN005937.pdf#search=%22Vi
3. Hypotheses
This paper will first analyse the effects corruption in a country has on the business activities of MNEs in that country. As explained in section 2.5, investors might be discouraged to invest in a country when there is a high level of corruption. MNEs might be asked for bribes before entering the market or corrupt officials might request a share in the proceeds of their investments. Therefore, an MNE investing in a corrupt country can expect to pay higher costs. Countries like Vietnam, Thailand, Philippines and Indonesia still have a high level of corruption according to the Transparency International. Research has shown that FDI increases economic growth and that corruption has a negative effect on economic growth. This paper will examine how FDI and corruption are related. Therefore, the first hypothesis is as follows.
H1: Corruption has a negative effect on FDI
The opposite direction of causality will also be tested. An MNE might be in a bargaining position when the government assumes that the investment of the MNE is going to be beneficial for the country. As pointed out in section 2.6, MNEs could then influence the government to adopt anti-corruption policies. Therefore, FDI is likely to decrease the level of corruption.
H2: FDI has a negative influence on corruption
North (1990) stated that the main goal of institutions is to reduce transaction costs. Institutions put constraints on the behaviour of economic agents and reduce the risk of opportunistic behaviour. When there is a weak institutional framework corruption is likely to arise because bureaucrats will have the possibility to extract monetary gains by interpreting the regulations to their advantage. A high corruption level can be interpreted as a part of a weak institutional framework. It can therefore be expected that when the institutional quality is high, corruption will be lower. Therefore the third hypothesis is:
4. Methodology
Since we want to test for the effect of corruption on FDI and the reverse causality we will use a multiple regression model (MR). This methodology has also often been used in this line of research previously. The most commonly used statistical tool to be able to explain the changes in the dependent variables in response to the changes in the independent variable is the ordinary least squares (OLS) (Hill et al., 2001). We will use control variables besides the corruption variable in the first regression and FDI in the second regression, so using a multiple regression will be the best option since it can estimate the coefficients simultaneously. Another advantage of the multiple regression model is that it has the capacity to work with quantitative as well as qualitative data. Dummy variables will be used to assess the effect of qualitative country- specific properties.
Econometric Model
The empirical analysis includes three stages defined in the theoretical framework. * In the first empirical model, the effect of corruption on FDI will be studied. To be able to analyze the effects of corruption on FDI cross-country time series data or panel data will be used. Using panel data will make sure that there are enough observations to do reliable tests. Data of six ASEAN member countries of the year’s 1997- 2003 is used. Therefore we have 42 observations. Based on the literature review the assumption is made that corruption will have a negative influence on the size of FDI inflows. However, to make the regression model more accurate we will need to control for other variables that are important for the amount of FDI inflows. The OLI framework of Dunning (1977) stresses that location advantages are important determinants of the amount of FDI inflow. Therefore, the regression will include the basic FDI location determinants of the OLI framework. A multiple regression model will be used since the control factors besides corruption need to be incorporated.
FDI = 1 + 2CORR + 3Xi + 1 = the intercept term
2 = measures the effect of a change in corruption upon the expected value FDI, all other variables held constant.
CORR = annual level of corruption
Xi = control variables other than corruption = error term
* In the second empirical model the effect of FDI on corruption will be examined. As stated in the literature review there are other variables that need to be controlled when examining the effects of FDI on corruption. For this model a multiple regression model will be used as well.
CORR = 1 + 2 FDI + 3X + 1 = the intercept term
2 = measures the effect of a change in FDI upon the expected value of corruption, all other variables held constant.
CORR = annual level of corruption FDI = annual FDI inflow
Xi = control variables other than corruption = error term
5. Data 23
Dependent variables
The dependent variables FDI and corruption will be explained by the independent variables mentioned later on. However, when FDI is the dependent variable in the regression model, corruption will be added as an independent variable. When corruption is the dependent variable, FDI will be added as an independent variable. * FDI; this study will investigate whether the volume of FDI inflow is affected by the level of corruption in the country. Therefore no distinction will be made between source countries and so data covering bilateral investments will not be used. Instead, data covering countries total annual FDI inflows will be used.
* An exact measurement of corruption is difficult. Previous literature often used subjective indexes of corruption and it is proven that the different subjective indexes are all highly correlated which will support the argument that the measures are reliable and that the same results hold across the different measures. This study will use the Corruption Perception Index (CPI) from Transparency International (TI) as a measure for the ASEAN member countries’ corruption level. The advantage of this index is that a value is given for each country every year. The data used in this paper covers the years 1997-2003 since data for most ASEAN countries has been available since 1997. The CPI is a composite index drawing on corruption indexes of nine independent institutions, such as the World Bank, the Economist Intelligence Unit (EIU), The World Economic Forum (WEF), the Institute for Management Development (IMD) and the Political and Economic Risk Consultancy (PERC) (Egger and Winner, 2004). According to Transparency International, the CPI is an attempt to assess the level at which corruption is perceived by people working for multinational firms and institutions how it impacts commercial and social life (Quah, 1999). The CPI ranges from 0 to 10; 10 equals a perfectly ‘clean’ country while 0 indicates a country where business transactions are entirely dominated by corruption. However, like Johnson (2004) and Wei (2000), we will transform the CPI index so that high figures correspond to a high degree of corruption.
Independent variables explaining the level of FDI:
The choice of control variables is motivated by the empirical literature on the location decisions of MNEs.
* Market seeking: market size is a very important determinant of FDI location. Market demand plays a large role when the foreign investor is seeking to sell its product or service in a country through local production. E.g. Culem (1998) and Brenton et al. (1999) came to the conclusion that the size of GDP or GDP per capita has a significant effect on FDI inflows. GDP per capita, GDP, population and a country’s labour force are variables used in previous research as a proxy for a country’s market size.
* Openness: another factor that determines whether the MNE will invest in a country is the degree of openness. Brenton et al (1999), Helpman (1984) and Markusen and Venables (1998) came to the conclusion that flows of trade and FDI are complements. We will therefore measure the degree of openness as: exports or imports, both divided by GDP. Another measure for openness of a country is the level of trade taxes. Having high tax levels will discourage MNEs to invest in the country when it wants to invest vertically. It might be a reason for a multinational to invest horizontally to avoid the taxes on international trade. The variable used in this study is taxes on international trade (see appendix for definition).
* Government expenditure: a high level of government expenditure might be associated with a high level of state intervention. MNEs are averse to high degrees of state intervention and will favour a free market system.
Independent variables that influence the level of corruption.
The choice of the control variables is based on previous studies that examine factors influencing corruption.
* Developed country: countries that are more developed have lower levels of corruption (Ades and Di Tella (1999); Sandholtz and Koetzle (2000); Treisman, (2000)). An indicator for a developed country is GDP per capita. It is an important control variable since per capita income can be seen as a proxy for having efficient and transparent institutions. Higher income per capita leads to higher public demand for institutions: good institutions are more affordable in high-income countries (Larraín and Tavares, 2004). The GDP level can also be used as a country’s welfare indicator.
* Openness: corruption will be low when the country has a high exposure to competition from imports. Openness to trade reduces corruption by decreasing the pools of rents available. This will decrease the incentive of regulators to try to get their part by demanding bribes and kickbacks (Sandholtz and Koetzle, 2000; Treisman, 1999; Larrain and Taveres, 2004). Ades and DiTella (1999) have examined the effects of rents on corruption and concluded that corruption increases in countries where local companies are able to isolate themselves from foreign competition. They used the share of imports in GDP as an indicator of rents available in the economy and found that an increase in international trade exposure will decrease the level of corruption.
* Government expenditures: larger involvement of the public sector will result in increasing opportunities for corruption; this implies greater involvement of the government in economic activity (Larrain and Tavares, 2004).
Section 2.7 explained that the level of corruption could also be influenced by country specific properties. Having been a British colony, or having been a communist state is likely to influence the level of corruption.
captures the interaction effect of being a communistic state and FDI on the corruption level.
Institution index
As already mentioned in section 2.3 and 2.4 corruption is more likely to exist when there is a lack of efficient institutions, which is caused by weak government regulations. Governance24 indicators of the World Bank are used for the correlation test to examine whether corruption and institutional quality are negative correlated or not25.
* Governance indicators capture six key dimensions of institutional quality of governance. The index rates countries on a scale between 0 and 100. A higher ratio indicates higher institutional quality. The first dimension is voice and accountability: measuring political, civil and human rights. The second one is political instability and violence: measuring the likelihood of violent threats to, or changes in government. The third dimension is government effectiveness: measuring the competence of the bureaucracy and the quality of public service delivery. This measures to what extent the government acts in an effective manner. The fourth dimension is regulatory quality: this variable focuses on the policies; it also includes the perception of burdens that occur from excessive regulations on trade and business development, price controls and banking supervision. The fifth dimension is rule of law: it measures the quality of contract enforcement, the police and the courts, as well as the likelihood of crime and violence. The last dimension is control of corruption: measuring the exercise of public power for private gain. The indicators cover time series data for the years 1996, 1998, 2000, 2002, 2004 for 209 countries and are based on hundreds of individual variables measuring perceptions of governance, drawn from 37 separate data sources constructed by 31 different organizations.26
24 World Bank defines governance as the exercise of political authority and the use of institutional
resources to manage society's problems and affairs.
25 A similar test is done by King (2003). This paper however uses the governance indicators as the
building blocks for the institutions.
6. Empirical findings
Tables 6 and 7 in the appendix show the regression models for FDI and Corruption. The models are tested for autocorrelation and heteroskedasticity. The results show that no autocorrelation or heteroskedasticity is present in the models except in the first models in both tables. The models are controlled for heteroskedasticity in model 2 using a log-log model.
interaction variable is insignificant and no strong conclusions can be drawn. Taxes on international trade are added in model 7 and are expected to have a negative effect on FDI. However, we see that there is a positive relationship between FDI and taxes on international trade in model 7,8,9 and 10 meaning that most MNEs might invest horizontally to avoid tariffs. The other variables in model 7, 8, 9 and 10 all have the expected sign. Most variables are significant except population, the corruption level and the dummy variables. Adding the tax on international trade into the regression model improves the model significantly; more variables become significant. The R-square shows the proportion of variation in the dependent variable explained by all the explanatory variables in the linear model (Hill et al., 2001). The F-test tests the overall significance of the model and is significant for every model.
Model 1 in table 7 shows a single OLS regression model illustrating the effect of FDI on corruption. FDI is negatively correlated with corruption and is significantly different from zero. Model 2 controls for heteroskedasticity using a log-log model. Table 4 shows the correlation coefficient for the variables that are likely to influence the level of corruption. FDI, imports/GDP and GDP per capita are highly correlated. Since we do not want the independent variables to explain the same variance in the dependent variable, import/GDP and GDP per capita will be left out. Therefore the independent variables added in model 3 are; government expenditure/GDP, GDP and population. The variables; government expenditure/GDP, GDP and population all have the expected signs and are significant except government expenditure/GDP. FDI is added in model 4 and is significantly negatively correlated with corruption, as was expected.
an ex-communist country will decrease to a larger extent than a non-communist country having the same level of FDI. It might be possible that Vietnam has other factors that result in a stronger negative effect of FDI on corruption, such as the requirements to enter the WTO. However, since the interaction variable is insignificant in both models no strong conclusions can be drawn The dummy for having been a British colony is added in model 9 and has a negative influence on the level of corruption as expected, however, insignificant. The interaction variable that captures the interaction of having been a British colony and the level of FDI is added in model 10. The negative sign is what was expected and is significantly different from zero; having been a British colony decreases the level of corruption more than a country that has not been a British colony with the same level of FDI.
Table 8 shows a correlation matrix between the corruption level and the various institutional quality variables. Corruption correlates negatively with the institutional quality variables; a lower degree of corruption relates to higher levels of institutional quality. Corruption correlates strongly with the different institutional quality variables (under -0.8), indicating that multicollinearity is present. However, voice and accountability can be called significantly independent.
A problem that might arise when examining only pair wise correlation is that the collinear relationship may involve more than two of the explanatory variables (Hill et al, 2001, p.190). Therefore we run an auxiliary regression (table 9, appendix) and find that the value for R-square is 0.9322 when CPI is used as the dependent variable. This means that 93.22 percent of the variation in the corruption perception index about its mean is explained by the variation in the institutional variables. Therefore, we can use the CPI index also as a measure for the quality of institutions.
7. Limitations, future research and conclusion
As mentioned before, no research has focused on the relationship between corruption and FDI in the ASEAN member countries. This is possibly due to data limitations. As shown in the regression tables we could only use 42 observations and only 31 observations when tax on international trade was added in table 6. Obtaining an extended data set and a larger sample is a suggestion for future research since more data will become available in the future.
that can influence the level of FDI could not be controlled for due to lack of data. It can be expected that the low wage costs play an important part in attracting FDI to Asia. These factors can be controlled for in future research to examine the effects on FDI.
A suggestion for future research to examine the level of corruption is to take into account the government wage levels. As already mentioned in the literature review, Van Rijckeghem and Weder ( 2001) and Treisman (1999) have tested the relationship between wage level and corruption empirically and found mixed results. Another factor to take into account is whether a country is abundant in natural resources or not. As already discussed in the literature review, countries relying on the export of natural resources tend to be more corrupt (Treisman, 1999; Leite and Widerman, 1999; Gylfason, 2004). Another factor to take into account is the use of internet. Making more use of the internet will increase the transparency and might help in reducing corruption. The World Development Indicators (WDI) of the World Bank (WB) is already collecting data of internet usage as a percentage of the total population. However. more data needs to be collected in order to be able to test this effect empirically.
size of the countries (measured by the countries population and GDP) persuade firms to invest.
Another way to combat corruption is to make sure that the country has a well functioning institutional framework. The results of this study showed that a good institutional framework goes hand in hand with low levels of corruption. Therefore, a country that has a good institutional framework is not likely to have high corruption levels. Governments should make sure that their institutions are well organized. Problems of corruption have also become an international problem due to the increase in trade and FDI27. Therefore international institutions like to WTO, APEC and the OECD have become more important in combating corruption. The international institutions assist countries in developing anti-corruption policies and programmes. Countries will feel more obliged to comply to the anti-corruption policies because doing so will likely result in higher trade and FDI inflows.
The results in this study also showed that FDI is significantly negatively related to corruption. This result holds to the inclusion of additional control factors. FDI does reduce corruption. It might be interesting to analyze the different ways in which corruption can be reduced by FDI empirically.
In this paper we found that corruption does not have a significant negative influence on the level of FDI after having controlled for other influential factors such as country size and government expenditure. However, we did find a significant negative effect of FDI on corruption. We can therefore assume that FDI will decrease the level of corruption further in the future. Perhaps the joint effect of government regulations, an improved institutional framework and an increase of FDI will decrease corruption and will start rolling the ball.
References:
Ades, A., and Di Tella, R. (1999) Rents, Competition and Corruption.
American Economic Review 89(4): pp. 982-994
Akçay, S.(2006), Corruption and human development, Cato Journal, 26(1): pp. 29-45
Alesina, A., Weder, B. (1999), Do corrupt governments receive less foreign aid?, National bureau of economic research working paper no. 7108, Cambridge
Baharumshah, A.Z., Thanoon, M.A. (2006), Foreign capital flows and economic growth in East Asian countries, China Economic Review, 17(1): pp. 70-83
Balasubramanyam, V.N., Salisu, M., Sapsford, D. (1996), Foreign direct investment and Growth in EP and IS countries, Economic Journal, 106(434): pp.
92-105
Bardhan,P. (1997), Corruption and development, a review of issues, Journal
of Econoomic literature, 35(3) :pp. 1320-46
Bhagwati, J. (1982), “Directly Unproductive Profit Seeking Activities,
Journal of Political Economy, 90(5): pp. 988-1002
Brenton, P., Di Mauro, F., Lucke, M. (1999), Economic integration and FDI: an empirical analysis of foreign investment in the EU and in Central and Eastern Europe, Empirica, 26(2): pp. 95-121
Blonigen, B.A. (2005), A review of the empirical literature on FDI determinants, University of Oregon and NBER
Burki, S. J., Perry, G. E. (1998), Beyond the Washington consensus: institutions matter. Washington, D.C.: The World Bank, Pre-Publication Edition,
1998
Campos e., Lien D., and Pradhan S. (2001), Corruption and its implications
for investment , the boom and bust of East Asia, Quezon City, edited by Campos, J.
E , University Press
Chai-Anan, S. (1977), Problems of bureaucratic corruption in Thailand: A study of legal codes, administrative and institutional arrangements, Bureaucratic
behavior in Asia project, Pattaya, Thailand
Chen, J-R. (2003), The Role of International Institutions in Globalization: The Challenges of Reform, Edward Elgar Publishing Ltd
Corpuz, O.D. (1957), The bureaucracy in the Philippines, Institute of public
administration, university of the Phillipines
Culem, C.G. (1998), The locational determinants of direct investment among industrialized countries, European Economic Review, 32(4): pp 885-904
Dalpina, C.E. (1991), Thailand’s search for accountability, Journal of democracy, 2(4); pp. 61-71
Drabek, Z., Payne, W. (2002), The impact of transparency on foreign direct investment, Staff working paper, ERAD-99-02, WTO
Dreher, A., Kotsogiammis, C., McCorristion, S. (2005), How do institutions affect corruption and the shadow economy?, Public economics, nr. 0502012
EconWPA, revised 24 Feb 2005.
Dunning, J.H. (1977), Trade, Location of Economic Activity and the
Multinational Enterprise: A search of an Eclectic Approach, New York, Holmes &
Meier The international allocation of economic activity, Hesselborn & M. Wijkman, eds,
