International Openness and Social Development as Endogenous Determinants of Growth and Convergence of the Countries in the MENA Region

FEM33-11 | December 2009

Title

« International Openness and Social Development as Endogenous Determinants of Growth and Convergence of the Countries in the MENA Region »

By

Jan J. Michalek, University of Warsaw, Warsaw, Poland

Contributeurs

Michal Brzozowski, University of Warsaw, Poland; Andrzej Cieslik, University of Warsaw, Poland; Lukasz Goczek, University of Warsaw, Poland; Dagmara Mycielska, University of Warsaw, Poland; Jerzy Mycielski, University of Warsaw, Poland; Agnieszka Pugacewicz, University of Warsaw, Poland; Victoria Roshal, Hebrew University, Israel; Alfred Tovias, Hebrew University, Israel

Note :

This document has been produced with the financial assistance of the European Union within the context of the FEMISE program. The contents of this document are the sole responsibility of the authors and can under no circumstances be regarded as reflecting the position of the European Union.

Summary :

The main goal of the report is to evaluate the direct and indirect influence of various social indicators (human capital measures, education, inequality etc.) and aspects of openness (FDI and trade in goods and services) on growth and convergence in the MENA region, and offer policy recommendations in these areas. The report is motivated by the fact, that social development is an important factor affecting the quality of life and social cohesion. However, it is less obvious whether or not the level of social development exerts an impact on economic performance.

The report begins with the analysis of relationship between social development and international trade, then we move to the relationship between social development and foreign direct investment, and finally we discuss the relationship between social development and economic growth. We find that while social indicators strongly affect trade flows, they affect FDI inflows in a limited way and they have no direct impact on economic growth. However, we have identified a number of indirect effects. In particular, the higher level of openness to international trade leads to increased FDI inflows. Furthermore, a higher level of openness to both international trade and FDI is positively related to a higher rate of economic growth. Therefore, social development affects growth at least through the channel of external openness.

We first analyze the results obtained for general measures of social development such as the Human Development Index (HDI) – the most general and widely available measure of social development, the Gini index – the most widely used index of income inequality, the Gender-related Discrimination Index (GDI) – the composite index measuring inequalities between men and women; the Gender Empowerment Measure (GEM) – another measure of gender inequality, and the Human Poverty Index (HPI) that measures deprivations of population in access to resources. Then, we analyze the results obtained for specific measures such as the level of education and infant mortality rate.

Due to data constraints, we are able to use only two of the aggregate measures of social development (HDI and Gini) in the panel regressions for the period 1980-2004. Our panel results were complemented with cross-section results for 2006 for both the whole sample and the sample of Mediterranean countries only. The cross-section approach allows for a much broader analysis of various aspects of social development captured by both aggregate and specific indices that became available only in the recent years. In our project, we analyze the trade data of exporting countries (all EU, OECD, and MENA countries) and of partner countries (all countries of the world above 200 thousand inhabitants). For convenience, we use the terms of “exporters” in the case of exporting countries and “importers” in the case of partner countries.

Social development and international trade

Our empirical results for the relationship between social development and international trade show that in the case of the HD index there exists a positive and statistically significant relationship between social development and the level of trade. This relationship exists both for the whole sample and in the case of cross-section regressions for 2006. There the HD index is positively related to exports and imports, while in the case of panel regressions we found the positive relationship to exist only in the case of exports. However, in the case of the limited sample of Mediterranean countries only we were unable to confirm this relationship which may be due to a low variability of the HD index among these countries.

The empirical results for the Gini index show that in the case of cross-section regressions for 2006 for the whole sample there is a negative relationship between the level of the Gini index and the level of trade. This result suggests that a more equal distribution of income stimulates both exports and imports. However, this result has not been confirmed by the panel data analysis where the Gini coefficient was not statistically significant either in the case of exports nor in the case of imports. In the case of the sample consisting of only Mediterranean countries we were also unable to verify this relationship due to the small number of observations.

The empirical results for the GDI are statistically significant for both exporting and importing countries in the cross-section. These results confirm the positive relationship between the GDI and the level of trade. In other words, the higher value of the GDI index is associated with lower gender inequality. However, for a limited sample including only Mediterranean countries the GDI index appeared to be statistically significant only for the exporting country. This means that the lower level of gender inequality in the Mediterranean countries is positively related to their exports. The empirical results for the GEM reveal that it is positive and statistically significant only for the exporting country in the case of the cross-section. The lower gender inequality in economic participation and decision-making power is associated with a higher level of exports. The same result holds for the limited sample of Mediterranean countries. Finally, the empirical results for the HPI display a negative and statistically significant relationship for both exporting and importing. This means that the lower level of poverty is associated with the higher level of trade flows. This result is partially confirmed for the sample of Mediterranean countries in the case of reporters (exporters). This implies that the lower level of poverty in Mediterranean countries at least is positively related to their exports.

The disaggregation of the HDI into its components reveals that only some of them might affect international trade flows. In particular, we find that the lower infant mortality rate, which reflects the better quality of the healthcare system, is positively associated with exports both in the case of the cross-section regression for 2006 and in the case of the panel regression for the whole sample. The same result is obtained for the sample of the Mediterranean countries. At the same time, we did not find such a relationship for imports.

The results obtained for the level of education show that the literacy rate is positive and statistically significant for the importing country in the cross section for 2006 while the average level of schooling in the case of the panel regression is positive and statistically significant for the exporters. We do not confirm this result for the Mediterranean countries only.

The disaggregation of the GEM into its components in the cross-section regression for 2006 leads to mixed results. For example, the variable that measures political participation of women (i.e. the percentage of seats held by women in national parliaments) displays a positive sign. It means that the higher participation of women in parliament is positively related to the level of trade.

Finally, we studied the relationship between civil liberties and the level of trade. It turned out that in the case of the cross-country regression for 2006 the civil rights were statistically significant only for the exporting countries while in the case of the panel data for both the exporting and importing countries. This means that the higher level of civil liberties is positively associated with the level of trade. The latter results were confirmed for the sample of Mediterranean countries.

Social development and foreign direct investment

Our empirical results for the relationship between social development and FDI inflows show that in the case of the HDI there exists a positive and statistically significant relationship between social development and the level of FDI. However, this relationship exists only for the whole sample in the panel regression and it is not confirmed in the case of cross-section regressions for 2006. The HDI has been found to shape foreign investors’ decisions over a long time horizon, but not in any given year. Therefore, our empirical results only to some extent support the notion that higher level of human development attracts foreign investors.

The empirical analysis of specific components of the HDI based on the aggregate FDI inflows in the panel specification revealed that there is a positive relationship between life expectancy, the level of education and FDI inflows, while the relationship between the infant mortality rate and FDI inflows is negative. Therefore, a long and healthy life and the educational attainment have been found to be important FDI boosters. The role of education is more pronounced when coupled with positive income growth. On the other hand, decent standards of living are not critical to investors contemplating to locate their capital abroad unless income distribution is uneven.

However, not all of the aforementioned results were confirmed by the cross-section regressions. In particular, life expectancy in the importing countries hosting FDI turned out not to be statistically significant. In addition, the cross-section approach allowed us to identify the negative relationship between the HPI and FDI inflows, which means that a low level of poverty stimulates FDI inflows.

Moreover, the study of inward FDI determinants has revealed that there are crucial determinants of FDI inflows, which are robust to the selection of the estimation method and control variables, namely the degrees of openness to international trade and of civil liberties, which were statistically significant in all regressions. This can be interpreted as follows. First, the share of imports and exports in GDP plays pivotal role in attracting foreign investors, pointing to the fact that trade and FDI are complementary. Openness to trade should be a central part of development strategies because a more intense FDI inflow is its byproduct. Second, the prospect of social stability is conducive to FDI inflows. The fear of social and political turmoil is one of the most important barriers to foreign investment. In addition, a higher level of civil liberties, that is associated with the higher level of contract enforcement, promotes FDI and seems to be more important than gender equality policies.

It stems from our analysis that the greater commitment to gender equality is not an important facet of an FDI enhancing policy. Investment in education and health improvements are best suited to create an environment that encourages foreign investment. The greatest effort should however be made to maintain and to widen the degrees openness and of civil liberties. It should be noted that all these steps may turn out to be insufficient because the sample including Mediterranean countries attracts about 50 percent FDI flows less than countries sharing similar characteristics, but located in different parts of world.

Social development and economic growth

Our empirical results for the relationship between social development and economic growth demonstrate that there exists a positive and statistically significant relationship between the HD index and economic growth. However, this relationship should be treated with caution as the impact of this variable on growth was not theoretically obvious. Given the fact that none of the HDI social components were significantly related to GDP growth we hypothesize that that the significance of the HDI variable is driven mainly by the initial level of per capita GDP.

The component measuring life expectancy turned out not to be statistically significant as well as the impact of infant mortality rate on the rate of growth. Therefore, our empirical results suggest that proponents of efforts to improve health in developing countries, which have not already lowered infant mortality rates nor increased life expectancy, should rely on humanitarian rather than economic arguments. The third component of the HDI – education – was also not statistically significant. However, it should be remembered that there is a long lag between the actual investment in schooling and its ultimate impact on economic productivity. Nevertheless, improvements in educational activity do not translate into an upgrade in workforce quality for some 10–15 years, at least until the people who are now being educated finally enter the labour force. Therefore, our empirical analysis suggests that specific measures taken in favour of social development are not directly related to economic growth which would therefore continue to be explained by the set of standard explanatory variables used by economists: the initial level of GDP per capita, the investment rate, the inflation rate, the government’s size, exchange rate variability, and openness to international trade and investment flows.

From the perspective of this report, we are particularly interested in the relationship between external openness and economic growth. The empirical results reveal that external openness to both international trade and FDI are positively related to the rate of economic growth.  Even so, our empirical results demonstrated that the social development measures are important in determining both trade and FDI flows. Moreover, the positive relationship between social development and trade is stronger than the relationship between social development and FDI. At the same time FDI inflows are positively related to the level of trade openness. Although we were unable to confirm the existence of a direct relationship between social development and economic growth we found that social development affects growth in an indirect way through external openness.

Policy recommendations

Our empirical results demonstrated that the social development measures are important in determining the level of trade, foreign direct investment. According to these results, social development indicators directly affect trade and FDI patterns, while they have no direct impact on the rate of economic growth. However, they exert an indirect impact on growth as openness to trade and FDI are significant growth determinants. Thus, in our view, openness to trade should be a central part of development strategies as more openness to trade induces more intense FDI inflow and stimulates economic growth.

The majority of social development variables are positively related to both trade and FDI. However, the estimated impact of specific variables is highly differentiated. Among the aggregate variables, the most important role is played by the HDI and HPI.

The significance of the HDI variable is especially important for inducing exports and encouraging FDI inflows as it stems from our panel analysis of both trade and FDI flows. However, in the case of cross-sections our results are less clear-cut. The positive correlations between the HDI variable and trade and FDI flows are driven mainly by the quality of the healthcare system measured in our analysis by the infant mortality rate and various measures of the level of education.

This suggests that the investments in the healthcare and educational systems would result in the increased stock of human capital and consequently in a higher level of exports of Mediterranean countries.  Moreover, investments in education and health improvements are best suited to create an environment that encourages foreign capital inflows. Although, we were unable to confirm these results for Mediterranean countries, due to the small number of observations, we believe that these general conclusions are valid also in the case of these countries.

The significance of the GEM and GDI variables is demonstrated only when relying on trade regressions. In that vein, the policy oriented towards more equal gender treatment in Mediterranean countries could be conductive to more trade. At the same time, our empirical analysis shows that greater commitment to gender equality is not directly affecting FDI inflows. However, the impact of the higher participation of women in political and economic life may also indirectly affect both FDI inflows and growth through increased trade openness.

Moreover, we identified a negative relationship between the HPI, trade flows and FDI inflows. This means that the lower level of poverty stimulates both trade and FDI inflows. Thus, the measures of combating poverty are important not only from the social point of view – the better the prospects of social stability – but can have a positive impact on the level of trade and FDI inflows.

Finally, the higher civil liberties are positively related to trade and FDI. Higher standards in keeping the rule of law are encouraging international trade and the enforcement of contracts may encourage the inflow of foreign direct investment. In spite of the fact that the conclusions and policy guidance based on regressions should be treated with some caution, the obtained results provide an additional rationale for intensifying reform efforts to address the problem of relatively low levels of external openness of Mediterranean countries. These results suggest that the investment in the social development would result in more FDI and trade, and only then higher income growth. This is similar to arguing, that the indirect effect of social development on growth via FDI and trade is the more important transmission channel of social development on growth, than the direct effect of social development. The estimated effects were robust and insensitive to the specification tests. Social development enhances growth only through its impact on FDI and trade; it could be that it encourages entrepreneurship and productive activity. However, through its impact on trade and FDI, it could still increase the productive capacity of an economy, drive job creation, bring innovation and new technologies, and boost income growth. In sum, we can be quite sure that although more openness brings greater efficiency, it does not automatically lead to higher growth rates. We must therefore presume that while some permanent effect probably exists, freer trade and FDI are only two of many factors accounting for improved growth performance.

Although the conclusions and policy guidance based on regressions should be treated with caution, the obtained results provide an additional rationale for intensifying the reform efforts to address the problem of low openness of the Mediterranean countries. It could be argued that an economy tends to grow better and is less prone to shocks, if it follows policies that foster economic efficiency. Sound macroeconomic environment, liberalizing trade, and FDI to build an economy based on a nation’s true comparative advantage – work in these areas is important, pursuing such policies is not likely to generate a growth miracle, but it is possible to improve Mediterranean countries’ growth prospects over time.