Data on Occupations Wanted

Mon, 7 Apr 1997 10:16:37 +0200 (MET DST)

Arno Tausch

The capitalist world economy and social development.

contribution to the volume:

Andreas Müller/Arno Tausch/Paul Zulehner (eds.)

Liberation Theology and the Social Sciences under review at

Macmillan, Basingstoke, 1997

1) Introduction

Economics and social sciences in general developed out of the discipline of moral science. For the development researcher in a 'developed western democracy', divinity studies - especially Latin American liberation theology - became again an interesting subject in the 1970s and 1980s, because many of the central issues which are at the basis of the development theory debate, were discussed by colleagues from the divinity departments. But it would be wrong to assume, that 'critical' development theory after the 1960s was just restricted to the debates in conjunction with dependency theory in the countries of the Third World. Development theory in the 'First World' took up these issues of 'critical political economy' and began to develop them in various directions, most importantly in the direction of the idea of capitalism as a world system, first proposed by Karl Polanyi 1944 and later on developed by Samir Amin, Andre Gunder Frank and Immanuel Wallerstein. A sub-field of these!
studies began to test - via established quantitative methods - the basic ideas of dependency theory, although a serious review of the literature will show that already. This quantitative sub-field is the main theme of this essay. Although largely overlooked by the world-wide debate that goes on about liberation theology, Marxism and dependency theory ever since the Vatican critique of liberation theology in the 1980s, this debate about the empirical essence of dependency and world systems theory has some important implications for both sides of the old moral sciences divide - the social scientific profession and divinity studies. It would be wrong to blame 'mainstream' quantitative social science for a neglect of the issues of dependency and capitalist instability. By contrast, the leading scholary journals of the world and the leading publishing houses carry contribution after contribution about 'disarticulation', 'income inequality', 'long cycles' or even about the Long 20t!
h Century.
Thus, this article uses advanced techniques of development and peace research to show the relevance of the underlying concern of liberation theology. Liberation theology foresaw these dimensions quite correctly, and it is up to divinity studies and the social sciences to further reflect these issues. A serious survey of the main contemporary issues of peace and development research will show, that the themes, presented by liberation theology, are very much on the agenda of the post 1989 world order.
Our empirical analysis focuses in particular on the long-term structural developmental blocks that result from a situation of dependency, understood as a high penetration of a country's capital stock by direct foreign investment. Thus, this essay leads back to the basic concerns of development theory in the 1970s.

2) The theoretical framework

Transnational capitalism has expanded tremendously since the early 1980s (see the contribution by Bob Ross to this volume). Far from history ending, the world - and with it the transnational capitalist system - continues to be beset by a never-ending series of contradictions, which are too well-known to be repeated here: the growing relative polarization of world incomes since the 1960s between the rich and the poor, the mounting problems of the sustainability of the development patterns of the industrialized world, the destruction of the world forests and - with it - of the indigenous people and cultures, and finally, the ever-growing question of security: social, environmental, health, freedom from repression, decay and crime.
Among the theories of social science, claiming compatibility with the empirical findings of cross-national development research, are those dealing with international competitiveness, liberal theories of institutional reform and transformation, and finally approaches in the tradition of social safety and social dialogue, which claim that the factor of societal 'embeddedness' is the most important determinant of world development. Although liberal approaches of institutional reform were rather dominant in the debate after 1989, the other theories and policy approaches equally claim quantitative evidence on a world scale in their favor.

Dependency and development

The almost unlimited number of empirical studies on peripheral capitalism and development on a world level in the B-phase of the Kondratieff cycle from 1965 onwards go back, in a way, to the classic essay, published by Johan Galtung in the spring issue of the Journal of Peace Research a quarter of a century ago (Galtung, 1971). Skepticism about the expanding world order, dominated by the transnationals, as a tool for achieving stable market-oriented development persisted, the neo-liberal transformation literature in the wake of the earthquake of the revolution of the year 1989 in Eastern Europe and the breakdown of communism notwithstanding (Aslund, 1992; Sachs, 1993). The second 'Great Transformation' of our century, that from socialism back to capitalism, can be seen in the perspective of the specific global capitalism aspects also inherent in the writings of Karl Polanyi: a Polanyian world view would hold, that authoritarian socialism in Eastern Europe crumbled like the old!
Hapsburg empire back in 1918 under pressure of the world market, that democracy had a chance in the region, like it had from 1918 onwards, but that this new chance, like the one in 1918 threatens to be contradicted by the 'obsolete market mentality'. For followers of the global capitalism school, Polanyi's anthropology of capitalism in a way foresaw the destabilization, the nationalist warfare, the unemployment, the international conflict, the social decay, and the more than 800000 excess mortality cases since the transformation in the whole of Eastern Europe and the former USSR, as Cornia's UNICEF study so aptly put it in 1994. These phenomena also, the argument goes on, imply why in the East the lessons of 'critical' development theory should remain on the agenda. Political turmoil continues in some regions of Eastern Europe and the former USSR 7 years after the transformation and transnational crime from the East and the South has become a very serious negative factor in i!
nternational relations. The sheer economic and political power of trasnationally organized crime is enormous. Turnover from drug trafficking is estimated around 500 billion $; 85 billion $ in drug profits are laundered through financial markets each year (UNDP, 1994: 37). Indeed, it is even plausible that the old big-power rivalries will continue under the new banner of culture and nationalism. Only four of the seventeen countries that adopted democracy during the Kondratieff cycle B-phase between 1915 and 1931 could save democracy; and today, freedom in the world is again on the retreat while violence and repression is on the increase.
Women in particular have to suffer from the tendencies of a world-wide new division of labor, as their jobs are being exported away to the still much-lower paid labor power of the periphery and the semi-periphery (Stiftung Entwicklung und Frieden, 1993). Neo-dependency schools would fear, that the most recent tendencies of world capitalism will strongly work against high female employment and create female unemployment, and they would expect two hypotheses to hold (i) transnational capital marginalizes female labor power (ii) the dynamics of growth turn away from those countries, where women still have a strong position on the labor market. The measurement scale, compatible with such hypotheses, would be the share of women in total employment and it's trade-off with growth rates. The new indicator series, developed by the UNDP for the UNO-Women-Conference in Beijing, provide further testing ground for the different feminist social theories of world development.
In international feminist literature, it has been maintained for the logic of the preceding Kondratieff cycle, that the world system - especially penetration by transnational corporations - creates a sexist pattern of development, which leads to a growing marginalization of women, especially in the periphery and semi-periphery. Feminist approaches in social policy would consider, that the share of women in the national legislature and the share of women in the total labor force reflect the positive role of women in society. More recent indicator series are available from the UNDP. Kondratieff cycles play an important role in contemporary political economy, mainly because they show the long-run instability of the system and the interrelationship between economic fluctuations and world political conflict. Latest evidence points to the fact that there are strong very short, and strong 20 year fluctuations, superseded by 50 year fluctuations of the world economy, that are closely !
linked with societal and technological change. These ideas go back to Schumpeter, and were fully developed today by the research of Bornschier, Goldstein, Kleinknecht, Modelski, Thompson and several others. Moving averages for 5 or 9 or 10 years then can 'filter' the more short-term cycles. Quantitative social scientists, like Joshua Goldstein, have also quantitatively established the relationship between these long cycles and world politics.
The heritage of the world depression in the 1980s, as was already explained in greater detail in Tausch/de Boer, 1996, is said to be unequal exchange, another important concept of dependency and world system theories (see also Raffer's contribution to this volume).
The main operationalizations of the concept of dependence in this study will be the share of inward foreign direct investments per total host country GDP by around 1985 (UNCTAD, 1996), and more traditional indicators of dependency, like terms of trade or trade dependency. A great number of economic nationalists and neo-nationalists cherish such an argument. Their position in part would be at least understandable, if MNC penetration indeed were to contribute to underdevelopment, short-term spurts of growth notwithstanding.
The share of foreign trade per total GDP is for some social scientific theories the symbol of the status of a nation as a small state in world society, conducive to social compromise and world economic adjustment. Opportunity of access to bigger markets will be deeply connected with this variable. Some dependency schools would regard however the share of trade per total product as an indicator of dependence (Hoell, 1983; Rothschild, 1944-1985). The main theoretical expectation of the refined global capitalism model in the tradition of Cardoso can still be summarized as follows:

(2.1a) structural imbalance of the development process (like high polarization of income distribution, insufficient human development, political or human rights violations) = constant + b1 * MNC penetration index + b2 * trade dependency index - b3 * terms of trade index

while 'classic' small state theories and early macroquantitative studies of development in addition would expect:

(2.1b) economic growth = constant - b1 * MNC penetration index - b2 * trade dependency index + b3 * terms of trade index

Over recent years, the liberal doctrine challenged the near dominance of global capitalism-paradigms in international social science that was evident in the late 1970s and the early 1980s. During the 1980s it became the dominant paradigm of politics in the USA, the UK and many other places. For the transformation countries of Eastern Europe it had a paramount importance. There are certain myths and fallacies involved in the interpretation of these theories.
First of all, - and this might be perhaps surprising to many - there is a certain structural similarity between many of the arguments from dependency theory and liberal development theory, once you 'translate' one theoretical language into the other and vice versa. For a true liberal economist, dependency is a special situation of the typical constraints, caused by a policy of import substitution and export discrimination. Policies, that create double deficits (huge current account balance deficit + large state sector budget deficits) and discriminate against internal savings, will lead to a high propensity to import foreign capital, often still aided by de-iure or de-facto policies that prevent enterprise creation and savings mobilization. Neo-classical economists would have to agree with 'dependentistas' in their critique of import substitution strategies, currency overvaluations, and - hence - the discrimination against exports in benefit of the urban sector and to the detr!
iment of rural society. The basic argument of such a sophisticated version of the neo-liberal school further runs as follows: in world politically stable countries with long recognized international borders, narrow distribution coalitions emerge in the wake of too big a state sector influence, and they will thwart growth perspectives and bring about stagnation and unemployment. High real appreciation of the national currency, declining private savings rates (Poland: 12.4% growth rate of private consumption), inadequate investment in the tradables sector, low export growth, not sustainable current account balance deficits per GDP (estimated by some economists to be not sustainable when twice the numerical value of the export growth rate or more), a growing reliance on short-term capital flows and large currency exposures are well known to neo-classical economics (Dadush and Brahmbhatt, 1995).
Neo-liberal cross-national studies of growth and development tended to conceptualize 'systems age' or 'age of democracy' by the number of years, that a polity enjoyed without changes to the externally recognized borders and in the framework of the establishment of free and competitive elections to the legislative chamber(s) (Weede, 1985-1992). Our operationalization of the concept of system's age is here the concept of the strength of international distribution coalitions: 'years of United Nations membership of a state', to account for the 'internationally recognized' position of a country in the international community. Thus, the power position as an distribution coalition broker in the international system is being measured:

(2.2a) stagnation = constant + b1 * age of the international post-war system participation + b2 * national state sector influence (like state sector expenditures per GDP)

or, on a world-scale:

(2.2b) stagnation (in the countries of the world system) = constant + b1 * chance for distribution coalitions to arise (years of UN membership) + b2 * state sector influence (like state sector expenditures, government consumption per GDP or share of public investment per total investment) - (b3 * proper internal and external conditions for the defense of democracy (a firmly entrenched democratic system and a stable integration into the Western security zone - b4 * cultural preconditions (like the Huntington Index)

To this, some other control variables from socio-liberal theories have to be added (Tausch/Prager, 1993).

3) The hypotheses and methods of analysis

As we already outlined, international social science since the mid-1960s studied patterns of international development in a cross-national perspective. This movement towards retrievability of research results, based on statistical analysis with internationally available and recognized data, which was initiated, amongst others, by the late Karl Wolfgang Deutsch from Harvard University, had important implications for international social policy. It allowed for the rigorous testing of hypothesis, contested in the political arena in an often passionate fashion.
Our attempt to estimate the determinants of world economic and social development from 1980 onwards tries to be based in this tradition. The UNDP Human Development Reports, our main new data source, emerged over the years as one of the leading socially scientific relevant data collections for cross-national research; the wealth of data contained in them shows concern for the global environment and for social decay and by far exceeds in quality other comparable products on the market today. The choice of the time period corresponds to the Kondratieff-type long cycle theories. Our data collection goes on to use some materials which are relevant for the description of the long-run position of a society in terms of ownership of the means of production (public investment, transnational investment), the social security program experience, and ethno-linguistic fractionalization from the Bornschier/Heintz data collection. In combining the new UNDP data with these older materials from !
the Bornschier/Heintz and the World Handbooks of Political and Social Indicators, I-III, pertaining to the earlier Kondratieff cycle, we fully integrate the new knowledge about cycles into our hypotheses.
Our data sources relied at least in part also on Fischer Weltalmanach; Nohlen; Seager and Olson and Stiftung Entwicklung und Frieden, which are excellent data handbooks for the study of international relations. Some data were also cross-checked with Tausch, 1993, 1994; UNECE; UNICEF (Cornia, 1993 and 1994); and the World Bank WDR and other sources. Our main sample of 123 nations comprised all the countries for which UNDP reports economic growth rates and life expectancies at two different periods. The countries of the ex-USSR are not being included for the reason of data limitations, while other 'real socialist' or ex-'real socialist' nations, like China and Hungary, at any rate integral parts of the conceptualizations of the capitalist world economy today, do form part of our main 123 countries' investigation.
Our leading, but not exclusive indicator of the process of dependence and globalisation is the legacy of MNC (multinational corporation) penetration in the 1980s (UNCTAD, 1996; as to earlier studies Tausch/de Boer, 1996).

Throughout this work, the following statements hold:

Regression coefficients at the level of error probability < or = 5% are printed in bold type. The following further conditions do hold:

Concepts: growth always refers to per capita income growth in real terms, if not specified otherwise

Time period: 1980s and beyond (if not specified otherwise)

Missing values: mean substitution, if possible, by known values for the economic or geographic region (like: countries with low human development, excluding India et cetera)

It should be explained here, what is meant under the term 'structural adjustment': the empirical measurement (and not normative concept) of adjustment compares the GNP per capita growth rates in two subsequent periods with a regression-based residual analysis. In other words: we try to answer the empirical, and not normative question, which countries accelerated their economic growth compared to the earlier cycle, and which countries adapted badly to the new conditions. Our measurement concept compares growth rates predicted for the period of the new Kondratieff cycle (post 1980/82)(^Yi) upon knowledge of the performance during the earlier Kondratieff B-phase (1965-80) with the actual growth rates Yi during the new Kondratieff cycle from 1980/82 onwards (Tausch/Prager, 1993)

(3.8) adjustment i = Yitn-^Yitn

Y = economic growth

Ytn = a + b1 * Ytn-1

Changes in the underlying logic of ascent and decline in the world economy will be especially observable at a time of comparison between the logic of a waning Kondratieff cycle and the emerging laws of a new cycle. Thus, adjustment will be a theoretically especially relevant phenomenon.

The trade-off between development level and social performance has been dealt with in great detail in other publications, so the reference to Tausch/Prager 1993 might be sufficient here.

4) The perspective of global capitalism

The global capitalism model of economic and social stagnation in the world economy, which was first developed by Bornschier, is clearly vindicated again with the new data. Our results reiterate the well-known dependency results in the tradition of Bornschier and Chase-Dunn, even for the new time period and the new indicators of world development, published by the UNDP.
Three variables measure dependency: MNC penetration, trade dependency, and terms of trade; the processes, relevant for liberal theories are government expenditures, UN member years, civil rights violations, and public investments. The reform theories are being measured by social security, women in parliaments, women as a percentage of the labor force, and the fertility rate. In addition, arms conversion and peace sub-theories in the reformist camp will be measured by military personnel ratios.
Among the factors, leading to deforestation, the export-oriented economy, the use of tropical wood in the world paper and furniture industry, and the burning of wood for cooking and heating purposes are the three most commonly mentioned factors. A great number of scholars, among them Leggett et al., 1991, already tried to bring deforestation rates systematically into a causal relationship with the kind of dependent capitalist development, analyzed amongst others by Bornschier and Chase Dunn, 1985.
Three measures were used to further test the relationship between globalisation and gender-related human development. One is maternal mortality, the second is the new UNDP gender-related development index, the third is the gender empowerment measure. The first and the third index are more distribution-oriented than the second indicator. Each year, 290 women per 100000 live births lose their lives in the moment of giving birth. What is the ultimate moment of happiness in a life for woman and man, to experience in togetherness the advent of a newly-born life, becomes the ultimate pain for millions of mothers around the world. They lose their lives due to the structural violence existing in the world system, they lose their lives in their ultimate moment of loneliness while giving birth, desolated and marginalized by a social order on the global level that produces more and more commodities, services and pollution but that forgets about the poor backyards, shanty towns and desola!
te clinics in the world poverty belts. In the industrialized countries of the OECD, maternal mortality is 11 per 100000. That is to say, at the global level there is an 'excess mortality' of 279 women per 100000 live births, considering the progress in medicine reached at the level of the western democracies. In Eastern Europe, maternal mortality already reaches 66 per 100000 live births, and in the developing countries, 420. All three indicators of the female situation de la vie are being significantly blocked by MNC penetration.
Our results indicate that dependency is by far the most important determinant of maternal mortality, and that the two dependency-related indicators: terms of trade and trade dependency co-determine the process of maternal mortality in the world system. In addition to the above-discussed measures of the social, political, and environmental well-being of nations, we have also included mean years of education as a measure of human capital formation, and the new UNDP capability poverty measure (CPM), which directly tries to estimate the percentage of people, marginalized from development in today's LDCs. Unfortunately, the UNDP did not calculate the CPM-poverty measure for the transition economies of Eastern Europe, where poverty and mortality often have increased during transformation. Again, MNC penetration significantly blocks development in the fullest sense of the word, negatively affecting the whole canon of development measures.
There are 19 variables and processes of development measured. 14 variables are explaining different aspects of development. MNC penetration significantly and negatively affects 15 of the 19 dimensions reported, the rest - life expectancy increases, the greenhouse index, ethno-warfare, and the existence of war and political destabilization in a country, are still affected in a fashion, as predicted by our theory, but not significantly.
One dimension, not significantly affected by MNC penetration, the greenhouse index, is shown to be a clear function of militarization; however, smaller nations in world society (high trade dependency index) still have a more favorable balance here. The new research design shows again, that women are particularly marginalized by the process of globalisation, as is shown in our results for gender development, gender empowerment, and maternal mortality. Female power and equity is incompatible, at the other hand, with the process of growth and adjustment in the era of globalisation.
Our findings suggest, that MNC penetration in the present Kondratieff cycle period significantly blocks adjustment, growth, the political and human rights record, human development, gender development, gender empowerment, life expectancy, a reduction of maternal mortality, and the protection of the world forests.
In addition, income inequality (the share of top 20% of income earners in the 81 countries according to Moaddel's data base, enlarged by WDR World Bank data, 1994) is well explained by the penetration of multinational corporations in the host countries, and is - in contrast to Weede's earlier findings - significantly enhanced and not lessened by militarization. MNC penetration increases significantly poverty, as measured by the new UNDP 1996 CPM poverty measure. International system participation age is an important control variable in the whole process of the explanation of post-1982 growth and development/stagnation. It significantly enhances the increase of life expectancy over time, gender empowerment, and the educational and employment record of a given country (the strong points of belonging long enough to the de-facto distribution coalition of long-standing UN members), while it significantly fails to block the deforestation process, especially due to the divergence bet!
ween professed ideals and dire realities in the long-standing UN member-nations in Latin America and in Eastern Europe. Terms of trade have a significant effect in the expected direction on the process of maternal mortality, and life expectancy.
However, our results need a series of qualifications: (i) militarization emerges - contrary to widely published research results by Weede about the world economy of the 1960s, the 1970s and the early 1980s - as one of the main development blocs in the period after 1980, with 5% significant results regarding adjustment, growth, employment, income redistribution, the greenhouse effect, and further notable effects on the two deforestation indicators. The main theoretical thrust of our new results with MNC penetration during the 1980s points, however, in the following further directions: (ii) fertility is negatively related to redistribution, employment and human capital formation, but there is perhaps somewhat surprising ceteris paribus positive effect on gender empowerment, mainly due to the relatively good gender empowerment performance of countries with a relatively higher historical fertility rate like Barbados, Bahamas, China., Ireland etc., whose gender empowerment is highe!
r than that of nations with a historically low birth rate like France (iii) The small and open economies in the world society also tend towards gender political power sharing and towards a better employment situation. A world economically open society with a high proportion of foreign trade per total product is also more likely to have - ceteris paribus - a lower greenhouse-index. To further support the predictions about the positive effects of trade dependency on development, developed by Katzenstein, one should emphasize also the positive effects on human capital formation and employment. But trade dependency also works as a classic transmitter of mechanisms of dependency by the effects it wields on life expectancy, maternal mortality, and the coverage of a nation with forests. (iv) Government activity is not significantly related to adjustment and growth; but a government-controlled economy increases also in this research design in a negative way the performance of a countr!
y regarding the political and human rights record. At the other hand, strong, and not weak government, enhances the human development index, and the gender development index, the human capital formation record of a nation, and government significantly reduces the amount of absolute poverty measured by the new UNDP CPM-measure to be found in developing countries. The effects of government or - more specifically - the social welfare state (social security index) on the indicators of the situation of the world woodland's are however not giving reason for optimism. The relative strength of the social welfare state has - like so many explanatory variables - positive and negative effects on the process of world development: it has a negative effect on employment, and it has a negative effect on human capital formation (mean years of education), while welfare states - ceteris paribus - still tend to behave in a protective fashion regarding their ecological systems (% forest coverage)!
.. (v) Established feminism (representation of women in parliaments and in the work force) emerges as the main loser of the world economic changes that have taken place since the 1980s. States with a high feminist power base at the one hand were - as in the above research design - successful in still increasing the feminist power base (gender empowerment index); and were performing relatively well on the forest protection front; but even the gender development index could not be affected in an upward direction by established feminism; and ceteris paribus, states with a well-developed feminist power base were performing very badly regarding adjustment, growth, life expectancy increases, political and civil rights performance, the human development index.

5) Conclusions and prospects

Thus, there is nothing where the main thrust of Christian social doctrine were to be contradicted by ongoing quantitative development research. Suffice to add, that the present author agrees with the general perspective, provided by students of the present international political economy, like Joshua Goldstein: the underlying logic of the capitalist world economy is a logic of conflict, intrinsically linked to the mechanisms of unequal exchange between the power centers of the world economy, and if the next 25 years are not being used for the creation of durable peace on earth, we could be threatened by a repetition of the same long-term cataclysms, that already led to the Thirty Years War, the Napoleonic Wars and the World Wars of this century.
Even the integration of the 'Near European East' - that is to say, the Czech Republic, Poland, Slovakia, Hungary and Slovenia, will be a painful process. For a long time to come, the official balance of trade and the balance on current account will be negative, a powerful force in the motivation of economic and political nationalism against the West.
A high MNC penetration during the past and present Kondratieff cycle works - amongst others - against a better civil rights performance, a better gender-related development and a better gender empowerment. The capitalist world economy is thus shown to have precisely the effects, that dependency theory already back in the 1970s and 1980s predicted to have on host-country development. Our results are fairly stable vis-à-vis the dependency indicator used: aid per capita, export processing zones and migration dependency neatly replicate the results, achieved for MNC penetration, during the latest phase of world capitalist development. A promising area for future research, however, will be the issue of the interaction between indicators of democracy and MNC penetration. Some, but not all of the dependency-oriented effects disappear, when we introduce political rights violations and or civil rights violations into the equations. Then the negative effect of MNC penetration is not dir!
ect, but via the repressive structures, that MNC penetration brings about in the long run. Again, some important questions regarding model specification, and, beyond that, degrees of freedom for political action to stabilize democracy in the periphery, arise.
Capitalism continues to be a system, full of contradictions, which are there to stay, also after the year 1989. Thus we are back at square number 1 - the great issues of moral science of the 18th century, that were taken up by the liberation theology debate in our age. Thus, the dialogue between 'critical political economy' and divinity studies is there to continue.

Samples and data

List of variables
The general model

Indicators of globalisation:

terms of trade index 1987-90 (UNDP, 1994, Weltalmanach, 1994, based on UN)
MNC penetration index: penetration by transnational capital, weighted by population and capital stock, mid-1970s (Bornschier/Heintz, 1979, based on OECD)
share of outward FDI stock in gross domestic product in 1985 (UNCTAD, 1996. For a very small number of countries, regional averages had to be taken to substitute for the few, missing values)
trade dependency index: exports plus imports as % of GDP 1990 (UNDP, 1993/94)
vulnerability of a nation in terms of the expansion of the new international division of labor, 1990 (share of women in the national labor force) (UNDP, 1993/94)
aid dependency (UNDP, 1994) net per capita aid in $. Donor countries are listed according to their donations per capita and are recorded as (-)
migration dependency (UNDP, 1993) net worker remittances per GNP/GDP at the beginning of the 1990s
export processing zones (Bailey et al., 1993). The variable codes the number of export processing zones per country at the beginning of the 1990s

Indicators for the institutional environment:

violation of political rights index, 1991 (Stiftung, 1993/94, based on Freedom House, combining free elections, role of the elected parliament in political decision making, party competition, protection of minorities)
violation of civil rights, 1991 (Stiftung, 1993/94, based on Freedom House, combining freedom of religion, the press, freedom of assembly and association, freedom of trade unions, the right to property and equality before the law)
government consumption per GDP, 1990 (UNDP, 1993/94)
government expenditures per GNP, 1991 (UNDP, 1993/94; UNICEF, Regional Monitoring Report, 1, 11, 1993, see Cornia, 1993; World Resources Research Institute)
public investment in the preceding Kondratieff cycle (Bornschier/Heintz)
years of membership in the United Nations (coded from Weltalmanach, 1995)
world political threats to a country, to be measured by the percentage of armed forces per population. Some neo-liberals maintain that world political threats increase the growth potential of a nation. Due to the skewness of the indicator, the natural logarithm ln (MPR+1) has to be taken (calculated from UNDP, 1993/94; see also: Weede, 1985)
ethno-linguistic fractionalization index, mid-1960s (Bornschier/Heintz, 1979, based on Taylor/Hudson). Ethnic discrimination is thought to be the purest form of a 'distribution coalition'
total area in thousands of square kilometers (World Bank, WDR, 1994. This variable measures the role, that territory and resources could play for attracting foreign capital)

Indicators for the social policy approach:

life expectancy at birth as an indicator for the quality of past social policy, 1990 (UNDP, 1993/94 and World Bank, WDR, 1994)
human development index (UNDP, 1994) as an indicator for the quality of past social policy
violation of political rights index, 1991 (Stiftung, 1993/94, based on Freedom House)
violation of civil rights, 1991 (Stiftung, 1993/94, based on Freedom House)
total number of inhabitants, divided by the surface area of a country (population density). Due to the skewness of the indicator, the squared root (population density^.50) had to be taken (calculated from UNDP, 1993/94)
increase/decrease of fertility rates 1960-90 (UNDP, 1993/94)
share of women in the membership of national legislature (lower house) (UNDP, 1993/94)
total fertility rate (most recent estimate) (UNDP, 1993/94)
social security benefits expenditure as % of GDP in the era of the evolving contemporary Kondratieff cycle, 1985-90 (UNDP, 1994). Social security benefits expenditures include here the compensations for the loss of income for the sick and the temporarily disabled; payments to the elderly, the permanently disabled and the unemployed; they also include family, maternity and child allowances and the cost of welfare services. The UNDP data collection is based on ILO sources

The dependent variables of the general model

GNP per capita growth (1965-1980 and 1980-90/1980-92, 1980-93) (UNDP, 1993/94/95/96; Fischer Weltalmanach) - dimension growth
capability poverty measure (CPM-value) (UNDP, 1996. The measure weights unattended births, underweight children, and female illiteracy. It is regarded by the UNDP as a direct measure of absolute poverty)
adjustment 1965/80-90/93(calculated from UNDP, 1993/94/96) - dimension growth
increase in life expectancy 1960-90 (calculated from UNDP, 1993/94 via a regression procedure, predicting 1990 life expectancy on 1960 life expectancy, and then taking the residuals as growth rates) - dimension redistribution and human development
violation of political rights index, 1991 (Stiftung, 1993/94, based on Freedom House) - dimension democracy
violation of civil rights, 1991 (Stiftung, 1993/94, based on Freedom House) - dimension democracy
human development index (HDI) (UNDP, 1994)
gender-related development index (GDI) (UNDP, 1995) - dimension female life chances. This index was developed by the UNDP especially for the 1995 women's conference in Beijing. The index weights the share of earned income for females and males, the gender-specific life expectancies, the gender-specific adult literacy rates, and the gross primary, secondary and tertiary enrollment ratios. It ranges theoretically from 0.0 to 0.999, with Sweden (0.919) at the top of the international scale, and Afghanistan (0.169) at the bottom. Since there unfortunately no were no data for Dominica, Grenada, Antigua, Seychelles, Saint Lucia, Saint Vincent, Saint Kits, Belize, South Africa, Oman, Jordan, Gabon, Solomon Islands, Sao Tome, Congo, Rwanda, Bhutan, Angola, Mauritania, Somalia, Gambia, Germany and Israel, we had to substitute these missing values with averages for the socio-economic groups concerned: a) the industrialized democracies (gender development index average 0.87) b) developin!
g countries with a Human Development Index above 0.6 (in our 123 nations analysis: Barbados to Tunisia, gender development index average 0.721) c) developing countries with a Human Development Index from 0.599 to 0.389 (Oman to Egypt, gender development index average 0,542) d) developing countries with a Human Development Index under 0.388 (except for the very least developed countries; Kenya to Sierra Leone; gender development index average 0.33) and e) the very least developed countries Benin, Guinea Bissau, Chad, Mali, Niger, Burkina Faso and Sierra Leone with a gender development index of 0.2. This procedure can be regarded only as a first approximation and should be substituted in future research
gender empowerment measure (UNDP, 1995. The index weights seats held by women in parliament, the percentage share of women and managers, the share of women in the professional and technical workforce, and the share of women in total earned income
greenhouse index, 1989 (greenhouse index per 10 million people, UNDP, 1994. The greenhouse index measures the net emissions of three major greenhouse gases: carbon dioxide, methane and chlorofluorocarbons. The index weights each gas according to it's heattrapping quality in carbon dioxide equivalents and expresses them in metric tons of carbon per capita) - dimension environmental quality/degradation
deforestation rate (UNDP, 1993-95; World Bank, 1995; World Resources Research Institute). The index measures annual rates of deforestation in the 1980s in %. Because of missing data, a number of countries had to be coded by the LDC average or the OECD average
forest area per total land area (UNDP, 1993-95). In contrast to the above indicator, that measures flows, this measure rather captures stocks of already existent forest destruction. Agricultural land per total land area has to be taken into consideration as an independent variable, because else the regression equations would be biased by a desert-factor
employment (UNDP, 1994) labor force as % of total population
mean years of schooling of the population aged 25 and > (UNDP, 1993 and 1994)
income distribution (Moaddel, 1994. The measure focuses on the share of the top 20% in total incomes in over 80 countries. Wherever possible, Moaddel's data were updated by World Bank WDR, 1994 and 1995)
ethno-warfare (Gurr, 1994) magnitude of ethno-political conflict. The scores are country sums of the squared roots of the deaths (in 10s of thousands) from ethno-political conflict 1993-94 plus refugees (in 100s of thousands). Countries with no entries according to Gurr's main research results, 1994, are coded as '0'
destabilization index (coded according to Weltalmanach, 1995. The indicator codes all those countries as '1' (destabilized), that have entries about serious armed internal or external conflict in 1994. The rest is coded as '0'