The trade effects of financial flow liberalization: a focus on Africa.

Lewer, Joshua ; Terry, Neil

ABSTRACT

The purpose of this paper is to empirically test the impact of financial flow regulation on international trade. An extended gravity model for 74 countries is employed to test the impact of capital restrictions and investment barriers on trade for the years 1995-1999. The results indicate that a 10 percent reduction in capital flow regulation will enhance international trade by roughly 1.7 percent for the entire sample, and roughly 1.2 percent for the nine African country subsample. The empirical results indicate that national income, geographic location, and freedom of exchange in financial markets are the primary determinants of bilateral trade in Africa.

INTRODUCTION

During the last three decades, international investment has been growing at over twice the rate of international trade. Yet, as with trade, the general public does not perceive intertemporal investment and other forms of asset transactions with foreigners as being welfare enhancing. For example, a recent Associated Press poll revealed that three out of five Americans were in favor of restricting foreign capital flows, and over half of all respondents agreed that foreign investment in the U.S. was "dangerous." (Scheve & Slaughter, 2001). This negative sentiment toward international investment is pervasive worldwide as indicated by the many financial barriers that remain on foreign direct investment, foreign asset flows, and multinational bank lending.

Research on the relationship between international investment and macroeconomic factors like economic growth and international trade are still in the early stages. The main problem is that researchers are faced with a lack of historical data and evidence. It was not until the 1980's that many countries started to dismantle the barriers to international investment that were erected after World War II. Moreover, the deepening of international investment to include many more types of assets like foreign direct investment (FDI) and international equity (stocks) is a recent phenomenon; it was only twenty years ago when nearly all of the international financing was in the form of bonds or bank lending.

With the return of greater international investment, the risks of default and sudden reversal of investment flows have also grown. For example, in the early 1990's capital flows to developing countries rose to new heights, but defaults and sharp reversals in capital flows to Mexico in 1994, a number of East Asian countries in 1997, Brazil in 1998, and Russia in 1999, have caused concern about the volatility of unregulated international investment markets. It is not surprising that despite the potential welfare gains from international capital flows, there are frequent calls to manage the international investment sector.

Despite these challenges, economists have discovered several benefits from international investment. Several studies have documented the positive effects of international investment on technological progress (Romer, 1993; Moran, 1998; Aitken & Harrison, 1999), savings and investment allocation (Feldstein & Horioka, 1980), economic growth (De Long & Summers, 1991; King & Levine, 1993; Borensztein, De Gregorio & Lee, 1998; Temple, 1998), and asset diversification (French & Poterba, 1991; Obstfeld, 1994). But there is little empirical evidence on how international capital flows impact international trade. The purpose of this paper is to estimate the size and significance of this association. In order to test the relationship between financial freedom and international trade, the paper is organized as follows: First, gravitational underpinnings are used to develop hypotheses about the effects of asset regulation on bilateral trade flows. The next section presents the cross-sectional results for the 74 country global sample. Third, a subsample of African countries is analyzed. The final section concludes with a review of the findings and the resulting implications.

THE GRAVITY MODEL SPECIFICATION OF BILATERAL TRADE

To examine the extent in which financial policy influences trading patterns, one must hold constant all other natural economic determinants. The gravity model has been extensively applied (see, for example, Frankel, Stein & Wei, 1995; McCallum, 1995; Eichengreen & Irwin, 1995; Frankel & Romer, 1999; Freund, 2000; Frankel & Rose, 2002) and widely accepted as the preferred systematic framework for measuring "natural" trade patterns based on economic size (i.e. mass) and geographic distance between economies. In addition to explaining roughly 70 percent of the cross section variation in world trade, the basic gravity equation is theoretically interesting because it can be derived from a number of traditional trade models (Linnemann, 1966; Leamer & Stern, 1970; Anderson, 1972). The purpose of this study is to determine how much of world trade is determined by gravity factors, and how much is left over to be attributed to financial policy, if any at all. The standard gravity equation, borrowed from physics, specifies trade between a pair of countries to be a negative function of the distance between the countries and a positive function of their combined national products. The underlying "gravitational" relationship is:

(1) TRAD[E.sub.ij] = f(GD[P.sub.i]GD[P.sub.j] / DIS[T.sub.ij]),

where TRADE is the total value of bilateral trade between countries i and j, GDP is the respective Gross Domestic Product in millions of U.S. dollars, and DIST is straight-line distance (in kilometers) between the economic centers of country i and j. Taking the logs of both sides yields (lower case in logs):

(2) trad[e.sub.ij] = [a.sub.0] + [a.sub.1] (gd[p.sub.i]gd[p.sub.j]) + [a.sub.2]dis[t.sub.ij] + [u.sub.ij].

Most studies augment equation (2) with variables to account for geographic, ethnolinguistic, and economic conditions. We follow numerous other authors and specify the following gravity equation to control for the basic determinants of international trade:

(3) trad[e.sub.ij] = [a.sub.0] + [a.sub.1](gd[p.sub.i]gd[p.sub.j]) + [a.sub.2](po[p.sub.i]po[p.sub.j]) + [a.sub.3]dis[t.sub.ij] + [a.sub.4]CON[T.sub.ij] + [a.sub.5]LAN[G.sub.ij] + [a.sub.6]LIN[K.sub.ij] + [a.sub.7]FT[A.sub.ij] + [u.sub.ij],

where po[p.sub.i]po[p.sub.j] is the log of the product of the populations in country i and j, CONT, LANG, LINK, and FTA are dummy variables which take the value 1 for pairs of countries which have a contiguous border, common language, common colonial linkage, and common free trade area agreement, respectively.

To estimate the impact of financial freedom on trade, an additional variable which captures freedom of exchange in financial markets (FINFREE) is added. FINFREE comes from Gwartney, Lawson, and Samida (2001), and has a value range from 1 to 10, where the value 1 is given to countries with severe restrictions on the freedom of its citizens, banks, and other financial institutions to engage in capital transactions with foreigners. An index value of 10 is reserved for those countries with no capital controls. The variable captures the degree in which markets are used to allocate foreign investment. When the FINFREE variable is used in the gravity equation, it is the average of country i and j's index. The model now becomes: (4) trad[e.sub.ij] = [a.sub.0] + [a.sub.1](gd[p.sub.i]gd[p.sub.j]) + [a.sub.2](po[p.sub.i]po[p.sub.j]) + [a.sub.3]dis[t.sub.ij] + [a.sub.4]CON[T.sub.ij] + [a.sub.5]LAN[G.sub.ij] + [a.sub.6]LIN[K.sub.ij] + [a.sub.7]FT[A.sub.ij] + [a.sub.8FINFRE[E.sub.ij] + [u.sub.ij].

It is anticipated that the coefficient on FINFREE will be positive and statistically significant.

ESTIMATION AND FINDINGS

Most studies estimate equations (3) and (4) by using double logarithmic form. Ordinary least squares estimation permits the coefficients to be interpreted as elasticities. However, one problem with this technique is that country pairs whose bilateral trade is zero are omitted. Roughly thirty percent of the observations on trad[e.sub.ij] are zero for that data employed in this study. These omitted observations contain information about why some countries do not trade at all.

One solution is to specify the dependent variable in levels and use Tobit estimation. Interpretation of Tobit output is complicated by the fact that coefficients and standard errors are normalized during estimation, and the constant elasticity relationship is lost. The approach employed in this study is the scaled OLS (SOLS) technique of Eichengreen and Irwin (1995), which yields results similar to Tobit estimation while maintaining the double log form. Here the dependent variable is expressed as log(1 + TRAD[E.sub.ij]). For small values of trad[e.sub.ij] the logarithm is close to zero, and for large values of trad[e.sub.ij] the logarithm of the transformed variable is close to the logarithm of TRAD[E.sub.ij]; therefore approximating a "semi-log Tobit relationship." When an equation is estimated with SOLS, the least squares estimates are multiplied by the reciprocal of the proportion of the observations in which trad[e.sub.ij] does not equal zero. William Green (2000) states, "A striking empirical regularity is that the maximum likelihood estimates can often be approximated by dividing the OLS estimates by the proportion of nonlimit observations in the sample" (Greene, 2000; pp. 912). The empirical results from both SOLS and Tobit estimation techniques are reassuringly similar. We therefore focus on SOLS estimates, which have a much easier interpretation.

Scaled OLS estimates for equation (3) are summarized in Table 1. All of the arguments of the augmented gravity model have the correct sign and almost all are significantly different than zero. Geographic distance and economic size matter for bilateral trade across the 74 country sample. For example, the coefficient on distance is -0.697 in 1999, suggesting that for every 10 percent increase in distance; bilateral trade is reduced by 6.97 percent. It is important to note that common membership in a regional free trade area (FTA) enhances trade among member countries. The statistically significant coefficient on FTA is 0.199 for the year 1999, suggesting that, ceteris paribus, countries with common membership trade roughly 22 percent ([e.sup.0.199]-1 = 0.220) more than they do when there is no common regional trading agreement. This result reinforces Frankel, Stein, and Wei (1995) who argue that free trade areas have contributed to the growth of regionalism, and that the regionalization of world trade may reduce world economic welfare relative to a most favored nation norm.

Table 2 reports the results of equation (4) for the entire sample of 74 countries. The augmented gravity model results in Table 1 do not change significantly with the addition of FINFREE, indicating the importance of geographic and institutional variables on trade. The

FINFREE coefficient is significantly positive for all years in which data are available. Notice that the size of the coefficient increases over time, going from 0.149 in 1995 to 0.169 in 1999. This indicates that a country's financial environment became increasingly important to international trade flows on average during the late 1990's. A 10 percent increase in financial freedom increased trade by 1.69 percent in 1999.

RESULTS FOR THE SUBSAMPLE OF AFRICAN COUNTRIES

In this section the extended gravity model is used to distinguish the impact of foreign capital and investment freedoms on nine African countries. The African cohort provides an opportunity to test the model and methodology for consistency across a subsample that is generally acknowledged as the greatest growth and development challenge in the world. The sample is representative of the diversity of the continent with three North Africa countries (Algeria, Egypt, and Tunisia), five Sub-Saharan Africa countries (Malawi, Nigeria, Tanzania, Zambia, and Zimbabwe), and South Africa. Country selection was tempered by the observation that complete economic data is not available for several African countries.

The North African countries in the research cohort have purchasing power parity income per capita in the year 2000 ranging from $3,600 in Egypt to $6,500 in Tunisia. The economies of North African tend to be large exporters of hydrocarbon and the majority of the population follows the Muslim religion. Regional policy reforms instituted during the mid-1990s include monetary polices aimed at reducing inflation, broader privatization of markets, liberalization of the investment code to increase foreign investment, and trade barrier reductions. A continuation of these broad reforms is the primary economic challenge for the future.

Sub-Saharan Africa's economic growth has been the weakest among developing regions. Between 1965 and 1985, Sub-Sahara Africa's GDP per capita increased by less than one percent a year on average. In fact, economic performance actually began to deteriorate in the mid-1970s. Real incomes were lower in 1985 than in the mid-1970s throughout much of the region. To reverse the economic decline, many Sub-Saharan African countries participated in structural adjustment programs with the assistance of the World Bank. The cornerstone of the reform programs instituted during the last fifteen years are similar to North Africa and include inflation control, balance-of-payment management, domestic deregulation, trade reform, and privatization of public enterprises. Reform has been particularly difficult because of the powerful vested interests that have been created through government intervention. To date, there has been little progress in reforming public enterprises and the financial sector. The Sub-Saharan African countries in the research cohort have purchasing power parity income per capita in the year 2000 ranging from $710 in Tanzania to $2,500 in Zimbabwe. Political instability, fully developing a market economy, and the rapidly growing problem of HIV/AIDS are only a few of the challenges facing many of the nations of Sub-Sahara Africa. South Africa is a middle-income country with an abundant supply of resources, well-developed financial, legal, communications, energy, and transport sectors. The purchasing power parity income per capita for South Africa in the year 2000 was $8,500. The country has a modern infrastructure supporting an efficient distribution of goods to major urban centers throughout the country. However, economic growth has not been strong enough to significantly reduce an excessive rate of unemployment, and many people suffer from the lack of economic empowerment as a vestige of the apartheid era. High crime rate, corruption, and HIV/AIDS are a few other significant social and economic problems. Recent reform efforts include a relaxing of restrictive labor laws, increased privatization, reduction in governmental spending, and an increase in foreign investment.

Table 3 reports the empirical results for the nine African country subsample. Although the empirical results are not as strong as the 74 country global sample, four of the eight independent variables are statistically significant with the anticipated sign. As expected, national income and the geographic location variables DIST and CONT are revealed as significant determinants of bilateral trade in Africa. The result highlights a problem facing many African nations. The most viable trading partners are neighbor countries that do not have high per capita incomes or large import demand. Hence, a policy focusing on the improvement of transport and infrastructure throughout Africa in order to increase the flow of goods and services beyond the region could facilitate trade and economic growth, especially for landlocked nations surrounded by poor neighbors throughout Sub-Sahara Africa. It is somewhat surprising that variable FTA has a negative coefficient for the African subsample, although the variable is not significant in four of the five research years. Political instability throughout many countries in the research cohort during the sample years is a potential reason for the unanticipated result. The FINFREE coefficient is consistently positive and significant for every year of the study, accentuating the important impact financial freedom has on trade. It is encouraging to note that many African leaders have implicitly committed to the principles of financial freedom in recent years. Following this commitment will undoubtedly play an integral part in the level future of trade flows and economic growth throughout the continent. Finally, country population, common language, and colonial linkages do not appear to have a significant impact on trade patterns for the African subsample. The sign on the corresponding coefficients all have the anticipated sign but none are statistically significant.

CONCLUSION

Despite the well-known uncertainties of the global financial market, there are many reasons why economists support international investment liberalization. Recent research on international investment points to several channels through which international asset flows can lead to an improvement in economic welfare. They include facilitating international transfers of technology, improvements in the allocation of savings to investments, enhancing economic growth, and diversifying asset portfolios.

The purpose of this article is to test for an additional association, namely if there is a significant relationship between international investment liberalization and bilateral trade volume. Using an extended gravity model that controls for geographical factors, support for a significantly positive relationship between asset flow openness and international trade is found. The findings suggest that a ten percent decrease in barriers toward foreign assets transactions will increase international trade by almost two percent for the entire sample of 74 countries. The empirical results for the Africa subsample reveals that bilateral trade is strongly influenced by national income, distance between economic centers of the trading nations, border nations, and financial freedom. Improving infrastructure and expanding financial freedom are important components to trade expansion and economic growth in Africa. The results of this study are of a preliminary nature and should be applied with caution, especially for the African subsample. Further research is needed on the possible differences in how asset flow liberalization impacts developed and developing countries.

Data Appendix

* International trade data are taken from the International Monetary Fund's Direction of Trade Statistics Yearbook, 2000.

* Data for Gross Domestic Product in millions of U.S. dollars, population, common members of regional trade blocks [Andean Group, Asia Pacific Economic Cooperation (APEC), Association of Southeast Asian Nations (ASEAN), European Union (EU), Latin American Integration Association (LAIA), Southern Cone Common Market (MERCOSUR), North American Free Trade Area (NAFTA), and Southern Africa Development Community (SADC)] come from the World Bank's 2001 World Development Indicators.

* The distance is the number of kilometers between capital cities, and comes from the U.S. Geological Survey at ftp://kai.er.usgs.gov/pub/.

* Data on common border, common language, and common colonial link come from the CIA World Factbook 2000.

REFERENCES

Aitken, B.J. & A.E. Harrison. (1999). Do domestic firms benefit from direct foreign investment? Evidence from Venezuela. American Economic Review, 89(3), 605-618.

Anderson, J.E. (1979). A theoretical foundation for the gravity equation. American Economic Review, 69(1), 106-116.

Borensztein, E., J. De Gregorio & J. Lee. (1998). How does foreign direct investment affect economic growth? Journal of International Economics, 45(1), 115-135.

De Long, B. & L. Summers. (1991). Equipment investment and economic growth. Quarterly Journal of Economics, 106(2), 445-502.

Eichengreen, B. & D. Irwin. (1995). Trade blocs, currency blocs and reorientation of world trade in the 1930s. Journal of International Economics, 38(1), 1-24.

Feldstein, M. & C. Horioka. (1980). Domestic savings and international capital flows. Economic Journal, 90(358), 314-329.

Frankel, J. & D. Romer. (1999). Does trade cause growth? American Economic Review, 89(3), 379-399.

Frankel, J. & A. Rose. (2002). An estimate of the effect of common currencies on trade and income. Quarterly Journal of Economics, 117(2), 437-466.

Frankel, J., E. Stein & S. Wei. (1995). Trading blocs and the Americas. Journal of Development Economics, 47(1), 61-95.

French, K. & J. Poterba. (1991). Investor diversification and international equity markets. American Economic Review, 81(2), 170-191.

Freund, C. (2000). Different paths to free trade: The gains from regionalism. Quarterly Journal of Economics, 115(4), 1317-1341.

Greene, W. (2000). Econometric Analysis (4th Ed.). Englewood Cliffs, NJ: Prentice Hall.

Gwartney, J., R. Lawson & D. Samida. (2001). Economic Freedom of the World 2001 Annual Report. Vancouver: Fraser Institute.

International Monetary Fund. (2000). Direction of Trade Statistics Yearbook. Washington D.C.: IMF.

King, R. & R. Levine. (1993). Finance and growth: Schumpeter might be right. Quarterly Journal of Economics, 108(3), 718-737.

Leamer, E. & R. Stern. (1970). Quantitative International Economics. Boston: Allyn and Bacon Publishers.

Linnemann, H. (1966). An Econometric Study of International Trade Flows. Amsterdam: North-Holland.

McCallum, J. (1995). National borders matter: Canada-US regional trade patterns. American Economic Review, 85(3), 615-623.

Moran, T. (1998). Foreign Direct Investment and Development. Washington D.C.: Institute of International Economics.

Obstfeld, M. (1994). Risk-taking, global diversification, and growth. American Economic Review, 84(5), 1310-1329.

Romer, P. (1993). Idea gaps and object gaps in economic development. Journal of Monetary Economics, 32(3), 543-573.

Scheve, K. & M. Slaughter. (2001). Globalization and the Perceptions of American Workers. Washington D.C.: Institute for International Economics.

Temple, J. (1998). Equipment investment and the Solow model. Oxford Economic Papers, 50(1), 39-62.

UNCTAD (2001). World Investment Report. New York: UNCTAD.

World Bank (2001). 2001 World Development Indicators. Washington D.C.: The World Bank.

Joshua Lewer, West Texas A&M University

Neil Terry, West Texas A&M University

Table 1: Augmented Gravity Model of Trade

 1995 1996 1997

Constant
 -6.637 -6.908 -6.984
 (-22.68) * (-23.58) * (-23.48) *

gd[p.sub.i] 0.732 0.750 0.761
gd[p.sub.j] (76.77) * (77.70) * (76.24) *

po[p.sub.i] -0.104 -0.122 -0.118
po[p.sub.j] (-8.22) * (-9.68) * (-9.18) *

dis[t.sub.ij] -0.703 -0.711 -0.727
 (-27.25) * (-27.90) * (-28.33) *

CONT 0.730 0.678 0.656
 (6.17) * (5.79) * (5.58) *

LANG 0.277 0.395 0.326
 (3.09) * (4.48) * (3.67) *

LINK 0.357 0.331 0.219
 (4.12) * (3.87) * (2.55) *

FTA 0.192 0.195 0.216
 (3.39) * (3.48) * (3.83) *

Adjusted 0.816 0.820 0.818
[R.sup.2]

 1998 1999

Constant
 -6.899 -6.099
 (-23.31) * (-22.66) *

gd[p.sub.i] 0.731 0.705
gd[p.sub.j] (75.78) * (81.88) *

po[p.sub.i] -0.085 -0.068
po[p.sub.j] (-6.93) * (-6.03) *

dis[t.sub.ij] -0.672 -0.697
 (-26.51) * (-29.45) *

CONT 0.718 0.676
 (6.19) * (6.24) *

LANG 0.298 0.375
 (3.40) * (4.58) *

LINK 0.209 0.208
 (2.46) * (2.62) *

FTA 0.289 0.199
 (5.23) * (3.86) *

Adjusted 0.817 0.836
[R.sup.2]

Notes: Figures in parenthesis are heteroskedasticity-consistent
t-statistics. * indicates significant at the 95% level. There are
74 countries in the study, 2701 data points (=74*(73/2)).

Table 2: Financial Freedom and Trade

 1995 1996 1997

Constant -5.824 -6.172 -6.382
 (-19.13) * (-20.50) * (-21.13) *

gd[p.sub.i] 0.651 0.667 0.679
gd[p.sub.j] (48.27) * (49.53) * (49.44) *

po[p.sub.i] -0.002 -0.014 -0.008
po[p.sub.j] (-0.09) (-0.79) (-0.44)

dis[t.sub.ij] -0.767 -0.773 -0.781
 (-28.87) * (-29.59) * (-29.92) *

CONT 0.705 0.652 0.635
 (6.03) * (5.65) * (5.47) *

LANG 0.166 0.277 0.212
 (1.86) (3.14) * (2.39) *

LINK 0.277 0.258 0.160
 (3.22) * (3.04) * (1.88)

FTA 0.159 0.164 0.190
 (2.85) * (2.97) * (3.42) *

FINFREE 0.149 0.158 0.159
 (8.45) * (8.71) * (8.52) *

Adjusted R2 0.821 0.825 0.823

 1998 1999

Constant -6.337 -5.902
 (-21.43) * (-22.06) *

gd[p.sub.i] 0.642 0.648
gd[p.sub.j] (49.70) * (57.37) *

po[p.sub.i] 0.038 0.016
po[p.sub.j] (1.23) (1.05)

dis[t.sub.ij] -0.735 -0.733
 (-28.65) * (-30.70) *

CONT 0.687 0.656
 (6.03) * (6.11) *

LANG 0.168 0.279
 (1.93) (2.26) *

LINK 0.149 0.177
 (1.79) (2.26) *

FTA 0.258 0.189
 (4.75) * (3.69) *

FINFREE 0.185 0.169
 (10.15) * (9.73) *

Adjusted R2 0.839 0.839

Notes: Figures in parentheses are heteroskedasticity-consistent
t-statistics. * indicates significant at the 95% level.
There are 74 countries in the study, 2701 data points (=74*(73/2)).

Table 3: Financial Freedom and Trade for African Countries

 1995 1996 1997

Constant -3.315 -3.611 -3.757
 (-4.48) * (-4.76) * (-4.79) *

gd[p.sub.i] 0.496 0.522 0.553
gd[p.sub.j] (19.56) * (19.93) * (20.06) *

po[p.sub.i] -0.031 -0.042 -0.055
po[p.sub.j] (-0.91) (-1.20) (-0.26)

dis[t.sub.ij] -0.658 -0.689 -0.725
 (-9.91) * (-10.21) * (-10.43) *

CONT 1.288 1.433 0.987
 (3.38) * (3.67) * (2.44) *

LANG 0.182 0.208 0.179
 (1.09) (1.23) (1.01)

LINK 0.242 0.208 0.164
 (1.70) (1.41) (1.08)

FTA -0.372 -0.439 -0.324
 (-1.81) (-2.09) * (-1.48)

FINFREE 0.104 0.113 0.098
 (3.77) * (4.04) * (3.39) *

Adjusted 0.726 0.728 0.719
[R.sub.2]

 1998 1999

Constant -4.189 -3.807
 (-5.42) * (-5.28) *

gd[p.sub.i] 0.539 0.541
gd[p.sub.j] (20.92) * (22.81) *

po[p.sub.i] -0.026 -0.010
po[p.sub.j] (-0.74) (-0.03)

dis[t.sub.ij] -0.653 -0.659
 (-9.45) * (-9.99) *

CONT 1.324 1.359
 (3.28) * (3.48) *

LANG 0.195 0.158
 (1.11) (0.93)

LINK 0.166 0.157
 (1.11) (1.08)

FTA -0.268 -0.051
 (-1.24) (-0.24)

FINFREE 0.119 0.124
 (3.08) * (2.55) *

Adjusted 0.719 0.737
[R.sub.2]

Notes: Figures in parentheses are heteroskedasticity-consistent
t-statistics. * indicates significant at the 95% level.
There are 621 data points. Countries include: Algeria, Egypt, Malawi,
Nigeria, South Africa, Tanzania, Tunisia, Zambia, Zimbabwe.

Country Appendix

Algeria Estonia Malaysia South Africa
Argentina Finland Mexico Spain
Australia France Nepal Sri Lanka
Austria Germany Netherlands Sweden
Bangladesh Greece New Zealand Switzerland
Belgium Guatemala Nicaragua Tanzania
Bolivia Honduras Nigeria Thailand
Brazil Hong Kong Norway Trinidad & Tobago
Canada Hungary Panama Tunisia
Chile India Papua New Guinea Turkey
China Indonesia Paraguay Ukraine
Colombia Ireland Peru United Kingdom
Costa Rica Italy Philippines United States
Czech Republic Jamaica Poland Uruguay
Denmark Japan Portugal Venezuela
Dominican Rep. Korea Romania Zambia
Ecuador Latvia Russian Federation Zimbabwe
Egypt Lithuania Singapore
El Salvador Malawi Slovak Republic