As to what concerns the efficiency of FDI in bringing economic development, Kokko in his survey "globalisation and FDI incentives" finds out that in order for it to be effective, local firms must be able to get used to the managerial and technical know-how. Moreover, Borenstein and Lee (1998) have shown that capital inflows have significant effect on the host country e.g. a one percent point rise in the ratio of FDI and GDP increase the rate of per capita income growth of the LDCs by 0.3 percent to 0.8 percent.
Link between tax incentives & Investment
Empirical literature demonstrated that indeed, other factors are significant in determining investment, but tax rates also have their importance. In order to assess the significance of tax rates on investment, research has led to the development of the Marginal Effective Tax Rates (METRs). According to Van Parys and James "The calculation of METRs allows assessing the impact of several tax parameters on the cost of capital in a straightforward way". However since there is no or very little evidence on how cost of capital affects investment, one cannot tell with certainty what is the real outcome on investment.
Nonetheless, according to the UNDP (2000), government should be very careful in granting fiscal incentives even if the latter are effective in attracting FDI, empirical has demonstrated that it is in fact a blend of tax and non-tax measures that is the best way to get the largest amount of FDI. This point is supported by Oman (2000) who concluded from his interviews that before investing, foreigners consider two things, firstly, the "fundamentals", which include political and macroeconomic stability, the quality of institutions and labour pool (its skills and literacy rate), accessibility to the market (domestic and international) and whether there is sound infrastructure. Secondly, they consider incentives. While Mc Kinsley‟s research revealed that incentives are counterproductive, Bora (2002) in a study of 71 developing countries found that fiscal incentives are the most popular form of incentives accounting for approximately 66% most frequently used incentives.
To what extent does foreign investment respond to changes in different tax incentives?
By reviewing the empirical evidence on FDI and tax, De Mooij and Ederveen (2003) found out that the average elasticity of tax to foreign capital turns around -3.3. But it should be noted that there are various divergence when it comes to the definitions that the different authors used, whether it concerns FDI or tax measure. The results however principally concern developed countries. Desai, Foley and Hines Jr (2004) found out that investment in Europe by American firms is quite sensitive to taxes than do investments in general. "The results reported that for countries outside of Europe, 10% higher tax rates are associated with 2.3% reduced investment, while for European countries, 10% higher tax rates are associated with 7.7% reduced investment"
Altshuler, Grubert, and Newlon (2001) estimate the impact of tax rates on U.S. manufacturing investment between 1984 and 1992. They found that tax rates have a major impact on multinational funds, and this connection becomes stronger over time (elasticity of 1.5 in 1984 and 3 in 1992).
Jensen (2005), in his paper, "Do low corporate tax rates attract multinational companies?" tries to provide empirical evidence on the relationship between corporate tax rates and FDI by means of statutory tax rates, effective marginal tax rates, and effective average tax rates. The data comprised of 15 countries studied over 1980 to 1993. The main conclusion of the paper was that no significant link between corporate tax rates and FDI exists.
Bloom, Griffith and Van Reenen (2002) found out that tax credits in developed countries helped to bring investment in R&D. Devereux, Griffith and Simpson (2007) show that an investor will be pleased by the incentive in the case that he can build his business in a region where he will find more existing businesses in his field. This is what the authors referred to as „agglomeration externality‟ and fiscal incentives are ineffective in its absence.
Wells and Allen (2002) made use of real investment data in order to measure how the removal of tax holidays in 1984 in Indonesia could have affected the economy. The difference between the number of projects approved and the inflow of foreign capital is not significant.
Reduced CIT, Investment Allowance & their effects on foreign investment
Klemm and Van Parys (2009) studied the correlation that might exist between the different types of tax incentives and foreign investment. They took a sample of 47 developing countries and the data range from1985 to 2004. They found out that on the one hand, reduced CIT does impact on the level of FDI, but on the other hand investment allowances do not affect FDI, while tax holidays do impact heavily on the level of FDI. Subsequently they found out that many countries in fact compete over tax holidays and not over investment allowances. As to what concerns private investment, they found out following their analysis that reduced CIT does not have a significant impact. They are of the view that the reason why reduced CIT attracts FDI and not private investment is because of the difference in the definitions of the two concepts. The paper also points out that many countries refuse to propose alternatives to tax holidays in spite of warnings that economists have emitted against it because the latter is much more profitable than investment allowances.
FDI and DTTs
Blonigen and Davies (2000) used data on inbound and outbound U.S FDI from 1966 to 1992. FDI activity was measured in three ways: FDI stock, FDI flows, and affiliate sales. To capture the effect of treaties, several approaches were used. In the first set of regressions, a simple dummy variable equal to one was used if there was a bilateral treaty in effect for a country pair. Here, only significant effects were found when using the FDI specification developed by Carr, Markusen, and Maskus (2001). In that case, the coefficient was positive in accordance with expectation. For the second set of regressions, a treaty age variable was used measuring how long a treaty had been in effect. There, positive and significant effects of treaty age on both inbound and outbound FDI for a variety of empirical specifications was found. Thus, initial reaction was that treaties have a positive impact on FDI and that these effects increase over time.
One problem with this approach is that it combines the effects of old treaties that were in place long before the data begins with more recent ones. Since the old treaty partners (Australia, Canada, Europe, Japan and New Zealand) are also the largest homes and hosts for U.S FDI, the treaty variables may have been capturing unobserved differences between these countries and other nations. Therefore, in more recent version of the paper Blonigen and Davies (2003), two separate dummy variables are used, one for old treaties and one for new treaties. A negative and significant coefficient on the pooled dummy was actually found. When separate effects for old and new treaties were estimated, similar to the earlier results, the old treaty dummy is positive and significant. The new treaty dummy, however, is almost always negative and frequently significant even after including country-specific fixed effects. When country-specific treaty coefficients were estimated, positive and significant effects for Russia, the Czech Republic, and Mexico, especially for U.S inbound FDI was found. However, many negative and significant effects of treaties were found. Thus the more recent results indicate that the treaty formation does not have the unambiguously positive effect theory suggests.
Similar results are found in Blonigen and Davies (2009), which estimates treaty effects for OECD countries. In this paper, data covers inbound and outbound FDI stock and flow data for OECD countries from 1982 to 1992. As in the 2000 paper, when old and new treaties are pooled, a positive treaty coefficient was estimated. However, when using separate old and new treaties dummies, the old treaty variable is positive and significant whereas the new treaty variable is negative and insignificant. When controlling for fixed effects, coefficients on the new treaty variable including significant results when using FDI stocks are negative.
Louie and Rousslang (2002) found little support for FDI promotion. In their paper, they use 1990s income tax return data for U.S. MNEs to calculate the rate of return for foreign subsidiaries. They then test whether this rate of return varies with treaty enforcement. They do not separate the effects of old and new treaties and instead only consider treaties in force prior to 1987. In their regressions, they only find one significant negative coefficient on their treaty dummy. They attribute this finding to omitted variable bias because after taking into considerations the proxies for corruption and political instability, the significance of the treaty dummy dies away entirely. This led them to conclude that good governance attracts both FDI and tax treaties but that treaties have no effect on FDI.
Hartman (1984) considered data for the period of 1965-1979 to carry out an empirical inference on the effect of Taxation on the time series of FDI in the U.S. He estimated 3 variables to affect FDI; the after tax rate of return realised by foreign investors in the U.S (because it is more appropriate for companies going for expansion of their business activities), the general after tax rate of return on capital in U.S (applicable to the acquirement of extraordinary assets not expected to earn extraordinary returns) and the tax rate on U.S capital owned by foreigners relative to the tax rate on U.S capital owned by U.S investors (to confine possibility of tax changes applying to U.S only)
The regression results were:
Positive relationship of after-tax rate of return variables with the ratio to U.S-GDP of FDI financed by retained earnings and
Negative correlation of the FDI-GDP ratio with the relative tax rate on foreigners compared to domestic residents.
His conclusion was that the effect of FDI is quite strong and that low taxation leads to increase in FDI and hence GDP.
The thesis of Ronald B. Davies (2003) evaluates the character and frequency of tax treaties and then he examines the way in which the treaties influence FDI. The consequences of treaty re-concession on FDI are also talked about. An empirical inspection of these effects is made using data on U.S FDI. The author finds that neither the tax treaties nor the renegotiation have a strong positive impact on FDI.
In contrast to the above, the work of Zahir Shah (2003) has something else to articulate. The author examines the foreign FDI system in Pakistan with particular concentration to the result of tax concessions in sinking the cost of capital. Pakistan has a broad range of fiscal incentives differing by vicinity and these are carefully examined and contrasted. The piece of work argues that fiscal incentives are more suitable in attracting FDI because they have no direct repercussion for public resources.
Conclusion
After having considered the various theories about tax incentives and double taxation treaties, and after having empirically reviewed how theory translates in practice, we are now more familiar with the subject and can carry on with the dissertation
