Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
1. Driving factors of male employment in African countries John C. Anyanwu1,2 Abstract The issue of employment is currently one of the greatest development challenges facing countries globally, including those in Africa. In 2011, the global male employment-to-population was estimated at about 72.7% compared to the female employment-to-population ratio of only 47.9%. For Africa as a whole, the employment-to-population ratio for males was estimated at about 69.2% compared to a ratio for females of only 39.2%. In addition to analyzing the characteristics of male employment in Africa, this paper empirically studies its key drivers (proxied by the male employment-to-population ratio for the age group 15–64 over the period 1991–2009), using cross-sectional data. The results suggest that for all-Africa estimation, quadratic (squared) levels of real per capita GDP, greater access to credit by the private sector, more education, and higher male share of the population increase male employment while higher level of real GDP per capita, higher government consumption expenditure, urban share of the population, and being a net oil-exporting country tend to lower it. In Sub-Saharan Africa, our results indicate that higher domestic investment, greater access to credit by the private sector, more education, and higher male share of the population increase male employment, while higher government consumption expenditure, greater openness of the economy, urban share of the population, and being a net oil-exporting country tend to lower it. However, North Africa presents a different picture. The North Africa sample results indicate that while the quadratic (squared) element of democracy, greater access to credit by the private sector, inflation, and greater openness of the economy increase male employment; the level of real GDP per capita, the level of democracy, increased inflow of foreign direct investment, and being a net oil-exporting country tend to lower male employment in the subregion. The policy implications of these results are discussed. Key words: Male employment-to-population ratio; female employment-topopulation ratio; determinants; Sub-Saharan Africa; North Africa. JEL classification: J16, J21, J64, J68, I32, C33, E24, O55. 1. Professor John C. Anyanwu is a Lead Research Economist in the Development Research Department, Chief Economist Complex, The African Development Bank (AfDB), Temporary Relocation Agency, BP 323, 1002 Tunis-Belvedere, Tunisia. Email: [email protected]. 2. The views expressed here are those of the author and in no way reflect those of the AfDB or its Executive Directors. 12 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries Résumé La question de l’emploi est actuellement l’un des grands défis au développement que doivent relever les pays du monde, y compris ceux d’Afrique. En 2011, tandis que le ratio emploi masculin-population était estimé à 72,7 % au niveau mondial, le ratio emploi féminin-population plafonnait à 47,9 % seulement. Dans toute l’Afrique, le ratio emploi masculin-population n’est que de 69,2 % alors que le ratio emploi féminin-population s’établit à 39,2 %. L’article analyse les caractéristiques de l’emploi chez les hommes en Afrique et en étudie empiriquement les principaux déterminants (sur le ratio emploi masculinpopulation du groupe d’âge 15–64 ans et sur la période 1991–2009) à partir d’une base de données transversale. Nos résultats montrent que, sur l’ensemble de l’Afrique, des niveaux quadratiques de PIB réel par habitant, un plus grand accès au crédit pour le secteur privé, un meilleur niveau d’éducation et une proportion plus importante d’hommes dans la population améliorent la situation de l’emploi chez les hommes ; par contre, la situation a tendance à se dégrader avec l’accroissement du PIB réel par habitant, l’augmentation des dépenses de consommation du gouvernement, une plus forte proportion de citadins dans la population et le fait que le pays soit exportateur net de pétrole. Nos résultats se présentent comme suit : en Afrique subsaharienne, la situation de l’emploi chez les hommes s’améliore en fonction d’un certain nombre de facteurs – augmentation des investissements dans le pays, facilité d’accès au crédit pour le secteur privé, qualité du système éducatif, forte proportion d’hommes dans la population – et a tendance à se dégrader lorsque les dépenses de consommation du gouvernement sont élevées, l’économie plus ouverte, la proportion de citadins plus importante dans la population totale, et quand le pays est un exportateur net de pétrole. En Afrique du Nord cependant, l’élément quadratique de la démocratie, un plus grand accès au crédit de la part du secteur privé, l’inflation, ainsi qu’une ouverture accrue de l’économie améliorent la situation de l’emploi chez les hommes, mais celle-ci se dégrade avec le niveau du PIB réel par habitant, le niveau de démocratie et l’afflux accru d’investissements étrangers directs, et lorsque le pays est un exportateur net de pétrole. L’auteur examine les conséquences de ces résultats en termes de politiques. Mots clés : Ratio emploi masculin–population ; ratio emploi féminin–population ; déterminants ; Afrique ; Afrique sub-saharienne ; Afrique du Nord. Classification JEL : J16, J21, J64, J68, I32, C33, E24, O55. The African Statistical Journal, Volume 16, May 2013 13 John C. Anyanwu 1.INTRODUCTION The issue of employment has grown in prominence on national and global development agendas in recent times, given its socio-economic cum political implications. Though the employment challenge has its own dimensions, it scourges countries worldwide regardless of their stage of socio-economic development. Thus, employment is currently a global policy issue: employment increases economic growth; promotes political and social stability; positively affects progress toward the Millennium Development Goals (MDGs) and poverty reduction generally. It also leads to greater social equality and more efficient resource allocation; increased productive potential; and low dependency ratio (Anyanwu 2012; Anyanwu and Erhijakpor 2012; World Bank 2012a). In 2011, the male employment-to-population ratio, globally, was estimated at about 72.7% compared to a female employment-to-population ratio of only 47.9%. For Africa as a whole, the male employmentto-population ratio was estimated at about 69.2% compared to a female employment-to-population ratio of only 39.2%. While estimates for SubSaharan Africa stood at 70.4% (male) to 58.8% (female), the gender gap was more pronounced in North Africa. Women in North Africa faced an employment rate of only 19.6% (compared to the global average of 47.9%), the second lowest of all regions and subregions in the world – and against a figure of 68% for the men in the subregion during the same year. What factors are driving male employment in Africa and its key components? In addition to analyzing the characteristics of male employment in Africa, this paper empirically studies the key drivers (proxied by the male employment-to-population ratio for the age group 15–64 over the period 1991–2009), using cross-sectional data. It also draws out important policy implications for African countries. The model is estimated by feasible generalized least squares (FGLS) method, with subregional and oil-fixed effects. A deeper understanding of the key determinants of male employment in Africa is crucial for implementing effective policies to make Africa’s labor market more inclusive and promote gender equality in employment so as to reap its benefits in the shortest time possible. The next section of the paper summarizes the evidence on the characteristics of male employment-to-population ratios, an indicator of how effectively a country utilizes its male productive potential. The third section reviews some relevant empirical literature. The fourth section presents the model and data, while section five presents the cross-country regressions of the key determinants of male employment for the entire continent, Sub-Saharan Africa and North Africa. The last section concludes with policy recommendations. 14 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries 2. CHARACTERISTICS OF MALE EMPLOYMENT IN AFRICA Low employment is a global problem As noted above, the employment problem is currently a global phenomenon. Figure 1 demonstrates the relationship between male (malesq22012) and female employment ratios in OECD countries during the second quarter of 2012. It shows that the problem is most acute in countries like Greece, Spain, Hungary and Italy, for both male and female employment – and also Turkey for female employment. However, the problem is also serious in Belgium, France, the Euro Area and EU generally, among others. Yet, employment, including for males, is critical for economic development (see Figure 2 for the positive correlation between male employment (empmrnew) and economic growth in Africa, for example). 80 Figure 1: Relationship between male and female employment ratios in OECD countries, Q2-2012 Switzerland 70 Korea Norway New Zealand Chile Australia Netherlands Japan Canada Israel OECD - Total Austria United States G7 Czech Republic United Kingdom Sweden Luxembourg Germany Denmark Estonia Slovak Republic Finland European Union Poland Slovenia Euro area Portugal France Belgium Ireland Italy Hungary 60 Turkey 50 Male employment ratio in Q2 2012 Iceland Mexico Greece 30 Spain 40 50 60 Female employment ratio in Q2 2012 malesq22012 70 Fitted values Source: OECD database. The African Statistical Journal, Volume 16, May 2013 15 John C. Anyanwu 80 70 60 Burkina Faso Ethiopia Madagascar Nigeria São Tomé & Príncipe Eritrea Seychelles Benin Togo Zimbabwe Uganda Central African Republic Cape Verde Burundi Côte d’Ivoire Gambia Malawi Namibia Guinea Tunisia Chad Comoros Cameroon Mauritius South Africa Guinea...Bissau Angola Swaziland Congo. Dem. Rep. Ghana Mozambique Zambia Egypt Libya Kenya Botswana Congo. Rep. Algeria Equatorial Guinea Rwanda Liberia Mali Lesotho Tanzania Gabon Niger Mauritania 50 Male employment ratio 90 Figure 2: Positive correlation between male employment ratios and economic growth in Africa, 1991–2011 Sudan 0 5 10 Real GDP Growth Rate (mean) empmrnew 15 20 Fitted values Sources: ILO and World Bank databases, ILO (2012a), World Bank (2012b). Male employment ratio relatively high in Africa but far below that of South Asia; however, it fell in all regions between 1991 and 2011 In 2011, the South Asia region recorded the highest average male employment ratio (at about 78.5%) (Figure 3). While the male employment ratio for the whole of Africa was relatively high at 69.3%, that of Sub-Saharan Africa was slightly higher, at 70.8%. Africa’s performance was pulled down slightly by North Africa’s average of only 67.8%. However, as the figure shows, male employment ratios fell between 1991 and 2011 in all regions. 16 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries Figure 3: Male employment ratios by global regions, 1991 and 2011 South-East Asia & the Pacific Developed Economies & European Union Sub-Saharan Africa East Asia Africa World Latin America & the Caribbean South Asia Middle East North Africa 0 20 2011 40 60 80 100 1991 Sources: ILO database, ILO (2012a). Figure 4 presents the average employment ratio for men and women in African countries in 1991 and 2011. It shows that employment was significantly higher for men. For the continent as a whole in 2011, the male employment ratio stood at about 69%, compared to 39% for women (1.8 times higher than for women). However, these average figures hide huge disparities between North Africa and Sub-Saharan Africa (Figure 5). In addition, there is greater gender inequality in employment in North Africa than in Sub-Saharan Africa, across all age groups (Figures 6 and 7). The African Statistical Journal, Volume 16, May 2013 17 John C. Anyanwu Figure 4: Male and female employment ratios in Africa, 1991 & 2011 Figure 5: Male and female employment ratios in North Africa & Sub-Saharan Africa, 2011 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 1991 2011 0 Females (age 15+) Males (age 15+) Females (age 15+) North Africa Males (age 15+) Sub-Saharan Africa Sources: ILO database, ILO (2012a) Figure 6: Male and female employment ratios in North Africa, 1991 & 2011 Figure 7: Male and female employment ratios in SubSaharan Africa, 1991 & 2011 100 100 80 80 60 60 40 40 20 20 0 1991 2011 0 1991 2011 Females (age 15+) Males (age 15+) Females (age 15+) Males (age 15+) Females (age 25+) Males (age 25+) Females (age 25+) Males (age 25+) Females (age 15-24) Males (age 15-24) Females (age 15-24) Males (age 15-24) Sources: ILO database, ILO (2012a, b). 18 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries As Figure 8 demonstrates, the average male employment-to-population ratio in North Africa (57%) is far greater than that for women (18%). This huge gender gap in employment in North African countries largely explains the generally low levels of employment ratios in the subregion. For the African continent as a whole, the employment ratio averaged 45% for males against 38% for females between 1991 and 2011. In Sub-Saharan Africa, the employment ratio averaged 71.3% for males compared to 57.3% for females. In 2011, the employment ratio in North Africa was 68% for men, compared to just 20% for women, aged 15 and above. In Sub-Saharan Africa in the same year, the employment ratio was 71% for men, compared to 59% for women. Male adult employment ratios are also higher than those for the youth, but much more so in North Africa, where it stands at more than double that of the youth male (Figure 9). Indeed, in 2011, while the ratio of male adult–to-youth employment in Sub-Saharan Africa was 1.7, it was 2.2 in North Africa. Figure 8: Male and female employment ratios in Africa’s major subregions, 1991–2011 (%) 80 70 60 50 40 30 20 North Africa – females Sub-Saharan Africa – females 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992 1991 10 North Africa – Males Sub-Saharan Africa – Males Sources: ILO database, ILO (2012). The African Statistical Journal, Volume 16, May 2013 19 John C. Anyanwu Substantial variation in male employment ratios across African countries There were substantial variations in male employment ratios across African countries between 1991 and 2011. Figure 10 also shows that a number of smaller African economies have relatively higher male and female employment ratios compared to richer, oil-exporting and North African economies. Figure 9: Male adult and youth employment ratios in Africa, 1991– 2011 100 80 60 40 North Africa – male youth North Africa – male adult 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992 1991 20 Sub-Saharan Africa – male youth Sub-Saharan Africa – male adult Sources: ILO database, ILO (2012a, b). 20 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries 60 70 80 Burkina Faso Ethiopia Madagascar Equatorial Guinea Nigeria São Tomé & Príncipe Seychelles Benin Eritrea Togo Central African Republic Zimbabwe Uganda Côte d’Ivoire Burundi Gambia Malawi Cape Verde Namibia Chad Mauritius Tunisia Guinea Comoros Cameroon Guinea...Bissau South Africa Swaziland Zambia Angola Mozambique Ghana Egypt Kenya Congo. Dem. Rep. Libya Botswana Congo. Rep. Somalia Rwanda Algeria Mali Liberia Lesotho Tanzania Gabon Mauritania Niger 50 Male employment ratio 90 Figure 10: Relationship between average male and female employment ratios in African countries, 1991–2011 Sudan 0 20 40 Female employment ratio (mean) empmrnew 60 80 Fitted values Sources: ILO database, ILO (2012a). Falling male employment trend in Africa generally Male employment has fallen generally since 1991 across the continent, but the decline has been dramatic in some countries. These include Niger, Benin, Rwanda, Lesotho, Kenya, and Burundi (Figure 11). A similar picture is evident in the OECD countries between 2009 and 2012 (Figure 12). The African Statistical Journal, Volume 16, May 2013 21 John C. Anyanwu Ethiopia Burkina Faso São Tomé & Príncipe Ethiopia African Republic Guinea...Bissau Uganda Central Malawi Madagascar Burkina Côte d’Ivoire Guinea Namibia Benin Swaziland Faso Equatorial Guinea Burundi Zimbabwe Comoros Tunisia Chad EritreaGambiaNigeria São Tomé Angola Libya Mauritius Seychelles & Príncipe Togo Cape Verde Zambia Egypt South Africa Central African Republic Botswana Guinea...Bissau Ghana Cameroon Uganda Malawi Congo. Rep. Côte d’Ivoire Mozambique Guinea Namibia Benin Swaziland Somalia Congo. Dem. Rep. Burundi Comoros Tunisia Chad Algeria Gambia Kenya Mali Angola Libya Mauritius Cape Verde Liberia Zambia Egypt South Africa Botswana Ghana Cameroon Rwanda Congo. Rep. Mozambique TanzaniaSomalia Congo. Dem. Rep. Algeria Kenya Lesotho Mali Gabon Liberia Madagascar Equatorial Guinea Nigeria Eritrea Seychelles Togo 60 70 80 70 80 90 Zimbabwe 50 60 Mauritania Sudan Mauritania 50 Male employment ratio in 2010 Male employment ratio in 2010 90 Figure 11: Male employment ratios did not generally increase between 1991 and 2010 in African countries 50 Sudan Rwanda Tanzania Niger 60 Niger 50 Lesotho 80 90 empmrnew2010 Fitted values 60 70 80 Male employment ratio in 1991 90 Gabon 70 Male employment ratio in 1991 Sources: ILO database, ILOempmrnew2010 (2012a). Fitted values 80 70 80 Mexico OECD - Total 60 70 50 60 50 Norway Iceland Korea New Zealand Mexico Netherlands Japan Australia Turkey Male employment ratio in Q4 2009 Spain 50 Chile Canada Austria Norway United States United Kingdom Korea Czech Republic Luxembourg G7 Germany Australia New Zealand Chile OECD Total Denmark Sweden Netherlands Estonia Japan Slovak Republic Canada Poland TurkeyUnion Austria Finland European Israel United States Slovenia Euro area United Kingdom Czech Republic PortugalG7 Luxembourg Belgium Germany France IrelandSweden Denmark Hungary Estonia Slovak Republic Poland Italy Finland European Union Slovenia Euro area Spain Portugal Belgium Greece France Ireland Hungary 55Italy 60 65 70 Israel 50 Male employment ratio in Q2 2012 Male employment ratio in Q2 2012 Figure 12: OECD countries: Male employment ratios on a decline Iceland between 2009 and 2012 55 malesq22012Greece Fitted values 60 65 70 Male employment ratio in Q4 2009 malesq22012 75 75 Fitted values Sources: OECD database, ILO (2012a). 22 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries Overall, a substantial gender inequality persists in African countries. As shown in the distribution of the difference between average male and female employment ratios (malefemalediffemp) between 1991 and 2011, the gap between the two is not symmetrical but positively distributed, indicating gendered social exclusion (Figure 13). 100 0 50 Frequency 150 Figure 13: Distribution of average male minus female employment ratios in Africa, 1991 and 2011 0 20 malefemalediffemp 40 60 Source: ILO database, ILO (2012a). 3. REVIEW OF THE LITERATURE The literature suggests that the key determinants of male employment relate to a country’s stage of economic development, access to credit by the private sector, globalization, demographic factors, macroeconomic factors, education, cultural and social norms, perceptions and expectations, and political systems. It is hypothesized that as countries develop, the male labor force participation generally increases. Böserup (1970) suggests that men’s greater access to education and technologies leads to their displacing women from the labor force during the early stages of a country’s development. However, as development increases and women gain more access to education and technologies, the female labor force participation increases. The African Statistical Journal, Volume 16, May 2013 23 John C. Anyanwu Another well-established hypothesis for this phenomenon focuses on income and substitution effects. As development occurs, households’ unearned incomes rise, reducing the incentive for women to work outside the home. The negative impact of rising incomes on women’s labor force participation is termed the “income effect,” since greater household income implies that households are able to afford more female leisure time. On the other hand, the “substitution effect” works in the opposite direction – as female wages rise, more women have the incentive to enter the labor market (Goldin 1995; Mammen and Paxson 2000; Bloom et al. 2009; Chaudhuri 2009; Tam 2011). The stylized U-shaped curve holds for overall male employment in African countries. It can be seen in Figure 14, which depicts the relationship between economic development (as captured by real GDP per capita) and the male employment ratio across African countries between 1991 and 2011. 80 70 60 Male employment ratio 90 Figure 14: Africa: U-shaped correlation between male employment ratio and GDP per capita, 1991–2011 Ethiopia Burkina Faso Madagascar Nigeria Benin Eritrea Central African Republic Togo Zimbabwe Uganda Côte d’Ivoire Burundi Gambia Cape Verde Malawi Guinea Namibia Chad Comoros Guinea...Bissau Cameroon Swaziland Ghana Zambia Angola Egypt Mozambique Kenya Congo. Rep. Congo. Dem. Rep. Liberia Rwanda Mali Equatorial Guinea Seychelles Tunisia Mauritius South Africa Libya Botswana Algeria Lesotho Tanzania Gabon 50 Niger Mauritania Sudan 5 6 7 Log of Real GDP Per Capita (mean) empmrnew 8 9 Fitted values Sources: ILO and World Bank databases, ILO (2012a), World Bank (2012b). The male employment ratio in Africa partly reflects the natural resource endowment structure, being low in fossil fuel-rich and mineral-rich economies. For example, Algeria, Egypt, Libya, and Sudan have very low employment ratios relative to their income levels. The same applies to countries like Gabon, Mauritania, and Tanzania, among others. 24 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries Recent empirical work reveals that foreign direct investment (FDI) and international trade can generate employment opportunities for both men and women (Richards and Gelleny 2007; see also ILO 2012c). According to Oostendorp (2009), in the long term, FDI may increase male employment at the expense of females; alternatively, women may be pushed down the production chain into subcontracting work. Furthermore, FDI may widen the gender gap as technical training is offered primarily to men, thereby “improving male technical knowledge and reducing women’s access to technology and employment” (Parpart et al. 2000). Also, by benefiting men, FDI may reinforce existing gender inequalities (Ward 1984; Ernesto 2011). Policies designed to increase trade and FDI inflows reduce state revenues, and therefore reduce the government’s capacity to provide social services. Because women are often the key beneficiaries of these services, economic integration can act against men’s employment in many dimensions. In a recent study, Tseloni, Tsoukis, and Emmanouilides (2011) reveal a greater participation of women in paid employment than men in more populous countries, with a greater share of women in their populations, more equal income distribution, and higher growth rates. However, such countries also evince a lower level of economic development, democracy ratings or international capital mobility (i.e., current account surplus or deficit/GDP). The banking sector is the key conduit for financial intermediation in an economy. Thus, access to credit by the private sector enhances the productive capacity of businesses. Businesses and enterprises with adequate credit access have greater potential to grow in terms of sales, revenues, and operations; to expand investments; and to create more employment. Indeed, greater access to credit by the private sector fosters growth because it means that fewer firms will be financially constrained and more investment projects will be undertaken. This in turn should foster employment and economic growth. Financial development also contributes to building and improving productivity of assets held by poor people, creating opportunities for entrepreneurship and new investments, improving efficiencies in product and factor markets, while stimulating private sector development and job creation (see Gandelman and Rasteletti 2012). Also, as IFC (2012) notes, improving access to finance helps firms expand their operations, which can have a positive effect on the quality and number of jobs created. The effects tend to be greatest for smaller firms. In a recent review of seven evaluations that focused on the provision of loans and advisory services to MSMEs and households, the IFC (2012) found that private sector access to credit has mainly positive The African Statistical Journal, Volume 16, May 2013 25 John C. Anyanwu effects on job creation. The report concludes that increasing access to loans helped firms to expand, facilitating the hiring of more workers. According to the findings of Niemi and Lloyd (1981), inflation has an independent, positive impact on female labor force participation. As a result of women’s lower cash holdings relative to men, it is posited that it possible that women are less adversely affected than men by increases in inflation (Cardoso 1992). Another macroeconomic variable is domestic investment. Domestic investment is a key source of employment, wealth creation and innovation. Without it, countries are unable to spur the growth of their economies or to sustain the reduction of poverty over the long term. Where domestic investment is low, the productive capacity of the economy fails to increase. This results in lower rates of economic growth, fewer opportunities for the poor to improve their livelihoods, and lower rates of job creation. As the ILO (2011) shows, investment growth has a strong and positive effect on employment creation. The results show that a 1 percentage point increase in the investment growth rate produces a 0.12 percentage point increase in employment growth. In addition, results from ILO (2012c) show that both private and public investment rates positively and strongly affect employment. The literature on the relationship between government consumption expenditure and employment focuses on the transmission mechanisms and the results of policy actions. According to the real business cycle (RBC) model, an increase in government consumption expenditure (financed by current or future lump-sum taxes) has a negative wealth effect which should lead to a decline in consumption and hence a rise in labor supply. The rise in labor supply leads, in equilibrium, to a lower real wage, higher employment and higher output (Ramey 2011a, 2011b; Wilson 2012). However, the New Keynesian models differ in that they generally assume some degree of stickiness, that is, that wages and prices do not adjust instantly to economic shocks (Christiano, Eichenbaum, and Rebelo 2011). Earlier, the Keynesian IS-LM model had asserted that consumption rises in response to an increase in government consumption expenditure. According to the proponents of this position, consumers exhibit non-Ricardian behavior in the IS-LM model and consumption is a function of current disposable income. Thus, the impact of an increase in government consumption expenditure depends on how the government consumption expenditure is financed, with the fiscal multiplier increasing in line with the extent of deficit financing (see Wilson 2012). Indeed, Wilson (2012) concluded that 26 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries though the impact depends on the environment in which expenditures are made, the effects of government investment are potentially greater than other types of government expenditure, a finding which was also attested by Kehoe and Serra-Puche (1982). Aiyagari, Christiano, and Eichenbaum (1990) also find that persistent changes in government consumption have contemporaneous employment and output effects which are larger than those due to transitory changes. Fatás and Mihov (2002) maintain that increases in government consumption lead to increases in employment. Also, the ILO (2012c) finds that government expenditures on public wages and salaries, social benefits, and social transfers positively and significantly affect employment, while government expenditures on interest payments significantly reduce employment. As Sakellariou (2011) explains, changes in educational attainment and in the demographic profile of the population, explain changes in the female–male gap in labor force participation, especially in rural communities. Changes in education and literacy help to explain the variation in male and female labor force participation within a country (Ogawa and Akter 2007; World Bank 2010; Gallaway and Bernasek 2004) and in employment in US multinational enterprises (MNEs) in Africa (Asiedu 2004). Campa et al. (2011) analyze the extent to which the gender culture affects the gender gap in employment. They show that the index of gender culture, based on firms’ attitudes as well as female literacy and education, is significant in explaining the gender gap in employment in Italian provinces. As Forsythe et al. (2000) have noted, “rapid development is particularly likely to be accompanied by greater gender rigidity in a country with a tradition of patriarchal institutional arrangements.” Indeed, Böserup (1970), Moghadam (1994), Shukri (1996), Psacharopoulos and Tzannatos (1989) have found that Muslim and Latin American countries – i.e. countries with strong socio-religious views about women’s role in the public sphere and the workplace – are more likely to be characterized by entrenched patriarchal institutions (see also Antecol 2000; Fernández 2010; Fernández and Fogli 2005; Fernández, Fogli, and Olivetti 2004). In addition, the seminal work by Hegre et al. (2001) supports a quadratic relationship between democracy and employment. Democracy could also unleash women’s labor market potential and open up the decision-making process to the less privileged, including women, resulting in redistributive policies that would benefit these groups. Democracy could also increase women’s employment by increasing expenditures on social programs. We The African Statistical Journal, Volume 16, May 2013 27 John C. Anyanwu therefore expect that democracy will not have significant positive effect on male employment. 4. THE MODEL AND DATA This section focuses on the econometric analyses of the determinants of male employment in Africa. We use the cross-sectional time series data covering 48 African countries to empirically study the key drivers of male employment in the continent during the period 1991 to 2009. The variable that proxies male employment (male employment-to-population ratio for the age group 15-64 over the period) was used as dependent variable. The level of economic development, along with other control variables, acted as independent variables. Independent Variables Level of economic development To control for the level of economic development, we include a nation’s real gross domestic product (GDP) per capita measured in terms of constant 2000 dollars. We also include the square of real GDP per capita in order to determine whether a non-monotonic relationship exists between development and male employment. The quadratic term tests Böserup’s (1970) assertion that the gap between men and women increases at intermediate levels of economic development but subsequently narrows after a nation has achieved a certain level of economic development. We also include economic growth (real GDP growth rate) separately to control for the possibility that an economic decline or slowdown might have adverse effects on male employment independent of the level of development. Institutionalized democracy It has been hypothesized that democracy increases equity in gender relations as women and men become empowered through the political process. This is because it is assumed that democratic regimes have greater respect for human rights, relative to authoritarian regimes. We use the measure democracy from the Polity IV Project, in which a country’s level of democracy is ranked along a 21-point spectrum, ranging from -10 for fully institutionalized autocracies to +10 for fully institutionalized democracies, based on research conducted at the Center for International Development and Conflict Management, University of Maryland. Since it is intuitively plausible that democratic countries encourage male employment, we expect 28 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries that increasing levels of democracy will act to increase women’s employment more than men’s. Democracy is fitted as a quadratic function for capturing possible average across country non-linear effects. Access to credit by the private sector As noted earlier, access to credit by the private sector enhances the productive capacity of businesses, leading to greater potential to grow in terms of sales, revenues, and operations. Moreover it expands investments and creates more employment. We therefore include banking system’s credit to the private sector as a percentage of GDP. The higher the value of credit to the private sector, the more resources firms have at their disposal for investment, expansion and entrepreneurship, all of which increase employment. Macroeconomic factors Three indicators are used to measure macroeconomic conditions: (i) the inflation rate (percentage change in CPI), (ii) domestic investment rate, and (iii) government consumption expenditure relative to GDP. With respect to inflation’s effect on employment, the current literature is inconclusive. Some experts contend that inflation hurts women more than men, since women are disproportionately represented among the poor and thus unable to protect their consumption levels in the presence of rising inflation. However, others assert that inflation will not harm women as much as men due to their lower cash holdings; further still, others contend that inflation may actually benefit women by increasing labor force participation. We expect inflation to have a negative effect on men’s employment. The second indicator of macroeconomic condition is a nation’s domestic investment measured as a percentage of GDP. The higher the value of investment rate, the more resources a government and the private sector ostensibly have at their disposal to spend on economic and social programs, including investments for employment creation. The third macroeconomic condition is government consumption expenditure measured as a percentage of GDP. The higher the value of government consumption expenditure, the less resources the government has at its disposal to spend on economic and social programs, including investments for employment creation for both men and women. Demographic factors To measure the effect of key demographic variables on men’s employment, three indicators are used: (i) population growth rate, (ii) ratio of male to female population of those aged 15 to 64, and (iii) the share of urban areas The African Statistical Journal, Volume 16, May 2013 29 John C. Anyanwu to total population. Increasing population growth is expected to narrow the gender equality in employment. Inclusion of the sex population ratio ensures that changes in the population ratio due to changes in the sex population ratio are properly accounted for. In light of the above, the population sex ratio is expected to have a positive effect on the male employment. More specifically, increases in the male–female population ratio, a measure of labor supply, are expected to lead to increases in male employment. Thus, the higher the proportion of men in a nation’s population, the higher will be the employment of men. On the other hand, living in an urban area is associated with an increase in access to labor markets and formal employment opportunities. Men, like their female counterparts, have access to more economic opportunities in urban areas than in rural areas. This is because urban labor markets offer a wide variety of occupations, from manufacturing and services to clerical activities. Thus, increased urbanization rate is expected to lead to higher levels of male employment. Globalization In order to control for the effect of globalization on male employment, trade openness of the economy and foreign direct investment (FDI) are included as explanatory variables and are measured as a percentage of GDP. A nation’s openness to trade is defined as the sum of net exports of goods and services as a percentage of GDP. An increase in openness is hypothesized to augment male labor force participation thereby increasing male employment. In addition, if the export sector is primarily capital intensive, then male employment is expected to decrease as a result of differential access to productive resources. As authors like Oostendorp (2009) have argued, FDI is assumed to be positively associated with higher employment, especially for women. On the other hand, some authors have argued that FDI can have a positive effect on male employment by serving to reinforce existing gender inequalities in the access to the labor market and the gender division of labor. Indeed, in the predominantly agricultural nations of Africa, men have a greater advantage in producing export crops, compared with women who predominately produce crops for subsistence and local consumption. According to this hypothesis, greater access to export channels through FDI would further widen the gender gap. Many African countries are today blessed with abundant natural resources, which have been attracting huge FDI. However, most natural resources sectors such as minerals, are enclave and capital-intensive sectors, and operate to the advantage of men, thus widening the gender gap in employment. 30 Journal statistique africain, numéro 16, mai 2013 predominantlypredominantly agricultural nations of Africa, menofhave a greater advantage in producing agricultural nations Africa, men have a greater advantage export in producing crops, compared with women who predominately produce crops for subsistence and local crops, compared with women who predominately produce crops for subsistence and ter advantage in producing export consumption. According to this hypothesis, greater access to export channels through FDI would consumption. According to this hypothesis, greater access to export channels through FDI crops for subsistence and local 1. Driving factors of male employment in African countries further widen the gender gap. African countries are today blessed with abundant natural further widen theMany gender gap. Many African countries are today blessed with abundant xport channels through FDI would resources,natural which have been attracting hugeattracting FDI. However, mostHowever, natural resources sectors such assectors s resources, which have been huge FDI. most natural resources day blessed with abundant are as enclave and sectors, and operate theoperate advantage of men, thus of me minerals, arecapital-intensive enclave and capital-intensive sectors,toand to the advantage st natural resourcesminerals, sectors such widening the gender gap in employment. widening the gender gap in employment. ate to the advantage of men, thus Level of general education Education tends to broaden one’s views, reduce ethnocentricity, and increase one’s flexibility to one’s accept newreduce customs andreduce norms. As increase such, the level Education tends to broaden views, ethnocentricity, and one’s flexibility to flexib Education tends to broaden one’s views, ethnocentricity, and increase one’s of education attained by the general population plays an important role accept new customs and customs norms. As the As level of education attained by the general accept new andsuch, norms. such, the level of education attained by the g y, and increase one’s flexibility to in increasing access toplays the an labor market and employment opportunities. population plays an important role in increasing access to the labor market and employment population important role in increasing access to the labor market and emplo ducation attained by the general Indeed, menopportunities. with higher levels of education are more likely to enter the likely opportunities. Indeed, men with higher levels of education are more likely to enter the labor Indeed, men with higher levels of education are more to enter the he labor market and employment labor market, especially in urban areas, which may reflect their higher wage market, especially in urban areas, which may reflect their higher wage premiums and higher market, especially in urban areas, which may reflect their higher wage premiums and re more likely to enter the labor premiums cost and opportunity cost of(see being inactive (see also2007). Ogawa opportunity ofhigher being inactive (see also Ogawa and Akter 2007). opportunity cost of being inactive also Ogawa and Akter igher wage premiums and higher and Akter 2007). 007). Level of general education Level of general education Subregional and oil effects and oil effects Subregional Subregional and oil effects We five subregional to capture subregional effects. In addition, We include the fivedummies subregional dummies to capture subregional effects. addition, to c We include includethethe five subregional dummies to capture subregional effects. In toIncapture the effects of net oil exporters, we exporters, add two oil dummies representing oil dummies exporters net oil and effects of net oil weexporters, add two dummies representing net and oil exporters addition, tothe capture the effects of net we addnet two nal effects. In addition, to capture importers. importers. representing net oil exporters and net oil importers. nting net oil exporters and net oil The TheModel Model The Model Based on the above review and following the frameworks posited by Tseloni, Based on the Based above on review and following the following frameworks by Tseloni, and Tsouk the above review and the posited frameworks positedTsoukis by Tseloni, Tsoukis and Emmanouilides (2011), Choudhry, Marelli, and Signorelli Emmanouilides (2011), Choudhry, Marelli, and Signorelli (2012), and Eastin and Prakash Emmanouilides (2011), Choudhry, Marelli, and Signorelli (2012), and Eastin and P posited by Tseloni, Tsoukis andEastin and Prakash (2013), the relationship that we want to (2012), and (2013), the relationship that we want to estimate can be written as: (2013), the relationship that we want to estimate can be written as: (2012), and Eastin and Prakash estimate can be written as: Kommentar [SJ3]: In the equation as: below, if possible, position the2(1) further 2 1ME log( log( credit 4 (democ log MEit i log log(it rgdp rgdp democit2)it ) 5 (demo it ) log( i itright, 1level the ) rgdp credit itit)) 45((democ 2 log( it ) 3 log( it ) to with the right hand2 3 rgdp it ) 2 margin. X it it) 7 ( Z it ) it 5(democ ) 4 (democit ) ) 7( Zit6)( 6 ( X it ) it (i 1,...., N ; t (i1 ,....., T ),........ (1) 1,...., N ; t .......( 1,.....,1T) ),...............(1) where is the measure male countryini at time t; ii is atime fixed effect ME isofthe measure maleinemployment in country i at time t; ai a fixed effect ref i is reflecting whereME ME iswhere the measure of employment maleofemployment country at time differences between countries; 1 is the elasticity of male employment with respect realrespect time differences between countries; 1 is the elasticity of male employment a fixed effect reflecting time differences between countries; b1 is the towith me t; i is a fixed is effect reflecting capitatoincome in 2000, rgdp;in22000, is thergdp; male elasticity respectintowith quadratic elasticity ofreal male employment with respect capitawith income per capita income employment 2 isto thereal maleper employment elasticity respect to qu e employment withper respect 2000, rgdp; b2 is the male employment elasticity with respect to quadreal per capita GDP; 3 is the coefficient of access to credit by the private sector, credit; 4 is the real per capita GDP; 3 is the coefficient of access to credit by the private sector, credit; elasticity with respect to quadratic ratic real per capita GDP; b3 is the coefficient of access to credit by the coefficient of democracy, domec; 5 is the coefficient of the quadratic of democracy; X is the coefficient of democracy, domec; 5 is the coefficient of the quadratic of democracy; X the private sector, credit; 4 is the private sector, credit; b4 inflation is including the coefficient of(inf), democracy, domec; b5 (% is(inv), control variables, including (inf),inflation domestic investment of GDP) trade (inv) control variables, domestic(% investment of GDP) quadratic of democracy; X is the the coefficient of the quadratic of democracy; X is the control variables, openness (open), foreign direct investment (% of GDP) (fdi), primary school enrollment ratio openness (open), foreign direct investment (% of GDP) (fdi), primary school enrollmen tment (% of GDP) (inv), trade including inflation (inf ), domestic investment (% of GDP) (inv), trade (educ), urban population share (urban), population growth rate (popg), male–female population (educ), urban population share (urban), population growth rate (popg), male–female pop , primary school enrollment ratio openness (open), e (popg), male–female population foreign direct investment (% of GDP) (fdi), primary school enrollment ratio (educ), urban population share (urban), population Page growth | 18 rate (popg), male–female population ratio (popratio); Z represents subregional and oil effects dummies used as fixed effects; and e is an error term that includes errors in the male employment measure. We use the North African dummy with its separate estimation to check if indeed, North Africa is different. The African Statistical Journal, Volume 16, May 2013 31 Page | 18 Pa John C. Anyanwu Data for these variables are largely drawn from the World Bank’s WDI Online database, except as indicated in Appendix 1. The Feasible Generalized Least Squares (FGLS) regressions with subregional and oil fixed-effects were estimated to investigate the determinants of male employment. Table 1 provides detailed descriptions of the raw dataset. Table 1: Descriptive statistics of main regression variables (excluding dummies), 1991–2009 Variable Observations Mean Standard Deviation 931 960 950 895 916 877 950 942 962 623 1007 1007 988 966 71.77 1065.43 19.40 15.82 -3.57 92.66 20.78 75.10 3.88 36.72 38.23 2.33 1.03 3.93 10.04 1573.73 21.60 8.00 23.17 1175.11 11.13 38.56 9.45 26.36 17.31 1.14 0.06 7.68 Male employment ratio Real GDP per capita Credit to private sector-GDP Govt consumption expd-GDP Democracy Inflation Domestic investment-GDP Openness FDI-GDP Education Urban population share Population growth Male–female population ratio Economic Growth Note: These are raw data before the log and other transformations. Source: Author’s calculations. 5. RESULTS AND ANALYSIS Table 2 presents the results of estimating the male employment equation (1). Level of economic development In our model, the coefficient associated with real GDP per capita is found to be negative and statistically significant in the all-Africa and North Africa samples. To test the hypothesis that real GDP per capita has a non-monotonic relationship with male employment, the squared real GDP per capita is included as an explanatory variable. The quadratic term is positive in sign and significant at the 1 percent level in only the all-Africa sample. 32 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries Table 2: FGLS estimates of the determinants of male employment (with subregional and oil-fixed effects) Variable Africa Sub-Saharan Africa North Africa Real GDP per capita Real GDP per capita2 Democracy Democracy2 Inflation Govt. consumption expd-GDP Domestic investmentGDP Openness FDI-GDP Education Urban population share Population growth Male-female population ratio Credit to private sectorGDP Net oil exporters Net oil importers North Africa Southern Africa Central Africa East Africa West Africa Real GDP Growth Constant -14.993 (-3.32***) 1.127 (3.39***) 0.036 (1.58) 1.344 (1.92) -9.729 (-3.17***) 0.030 (1.31) -0.00001 (-0.03) 0.0001 (0.17) -2.081 (-2.62**) 0.446 (2.07**) 0.200 (2.61**) -0.379 (-5.52***) -0.476 (-6.87***) -0.083 (-0.39) 0.035 (0.76) -0.007 (-0.52) 0.027 (0.36) 0.124 (7.07***) -0.254 (-5.59***) -0.968 (-1.52) 0.079 (1.65*) -0.025 (-1.80*) 0.087 (1.15) 0.114 (6.59***) -0.338 (-7.90***) -0.548 (-0.85) -0.087 (-0.95) 0.146 (3.52***) -0.246 (-1.62*) 0.176 (2.72**) 0.586 (1.39) -0.905 (-0.35) 0.216 (2.59**) 0.260 (3.11**) -0.137 (-1.17) 0.080 (4.14***) -8.210 (-7.42***) 0.067 (3.29***) -11.771 (-9.86***) 0.198 (6.36***) -11.425 (-1.75*) -5.558 (-3.25***) 5.079 (3.22***) -5.220 (-3.15**) 0.729 (0.44) 0.066 (0.83) 105.389 (6.08***) 12.881 (8.28***) -1.000 (-0.78) 6.227 (4.89***) 0.022 (0.28) 44.431 (5.16***) 0.019 (0.16) 99.162 (3.16**) Log likelihood Wald chi2 Prob > chi2 N -2008.122 287.47 0.0000 569 -1803.161 317.52 0.0000 512 -106.5155 1219.41 0.0000 57 Note: t-values are in parentheses; ***= 1% significant level; **=5% significant level; *=10% significant level. These results provide evidence of a U-shaped relationship between real GDP per capita and male employment in Africa as a whole. Thus, these results suggest that although higher levels of real GDP per capita are negatively The African Statistical Journal, Volume 16, May 2013 33 John C. Anyanwu associated with male employment, the effect is not constant. Rather, above a certain point, higher levels of real GDP per capita act to increase male employment, holding other factors constant. This relationship suggests that the marginal effect of real GDP per capita exhibits increasing returns for male employment. This finding supports Böserup’s (1970) assertion of a curvilinear relationship but this U-shaped relationship contradicts the findings of Tseloni, Tsoukis, and Emmanouilides (2011), and Eastin and Prakash (2013). We note, however, that real GDP per capita (both level and quadratic) is not significantly related to male employment in Sub-Saharan Africa, while the quadratic element is not significant in the North Africa sample – hence was dropped. Institutionalized democracy Institutionalized democracy is negative and statistically significant only in the North African sample, against earlier findings such as those of Tseloni, Tsoukis, and Emmanouilides (2011), and Eastin and Prakash (2013). The quadratic term included to determine whether democracy has a nonlinear effect on male employment is positive in sign and statistically significant in the North African sample also, indicating a U-shaped relationship between democracy and male employment in North Africa, holding other factors constant. Institutionalized democracy (both level and quadratic) does not significantly affect male employment in Africa as a whole, nor in Sub-Saharan Africa – hence the quadratic term was dropped in both results. Access to credit by the private sector The credit variable has a positive and statistically significant effect on male employment in the all-African, Sub-Saharan African and North African estimations. These results are consistent with the findings of IFC (2012), thus underscoring the huge role of access to finance in employment creation in Africa. Macroeconomic factors It is only in North Africa that rising inflation is found to be positively associated with increasing levels of male employment. This is consistent with some of the results of Eastin and Prakash (2013). This finding may be linked to the fact that a reduction in real wages increases male labor force participation, as men enter the workforce in greater numbers to supplement household earnings. As shown in Table 2, a nation’s domestic investment rate is found to be positively and significantly associated with male employment only in the Sub-Saharan African estimation. Its effect is insignificant in the all-African 34 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries and North African estimations. This could be attributed to wastages, inefficiency, and corruption associated with investment projects in many African countries. Indeed, investment in white-elephant, unproductive activities, remains a development challenge confronting the continent. Related to the issue of wastages is that of government consumption expenditures in Africa. Our results show that government consumption expenditures negatively and significantly affect male employment in both the all-African and Sub-Saharan African estimations. These results are also in conformity with the crowding-out theories prevalent in the literature. Globalization Trade openness is negative in sign and statistically significant in the SubSaharan African estimation but it is positive and statistically at the 1 percent level in the North African case. The North African results support the view that increasing levels of exports relative to imports may increase male employment and that an external market orientation may further enhance job opportunities for men. This suggests that labor-intensive export sectors dominate the capital-intensive sectors in North Africa, while the opposite is likely to be the case in Sub-Saharan Africa. On the other hand, the FDI–GDP ratio was found to have a positive but insignificant effect on male employment in both the all-African and SubSaharan African estimations. However, it has a negative but statistically significant effect on male employment in North Africa. Our findings, therefore, do not support the proposition that the inflow of foreign direct investment enhances male employment in Africa. In fact, we found that it lowers male employment in North Africa. Education The education variable has a positive and statistically significant effect on male employment in the all-African, Sub-Saharan African, and North African estimations. This is consistent with Asiedu’s findings (2004). This supports the hypothesis that education tends to broaden one’s awareness of cultures and social norms that exist in industrial countries where both men and women are in most circumstances entitled to the same freedoms and opportunities. Demographic factors Increasing urbanization rates are found to be negatively and highly significantly associated with falling male employment in the all-African and Sub-Saharan African estimations. As seen in Table 2, this effect is statistically The African Statistical Journal, Volume 16, May 2013 35 John C. Anyanwu significant at the 1 percent level in both cases. However, the variable is positive but insignificant in the North African case. Furthermore, our results suggest that rising population growth rates have a negative but statistically insignificant effect on male employment in all three sample groups. The ratio of male to female population has a positive and statistical significant effect on male employment in the all-African and Sub-Saharan African samples. Thus, the higher the proportion of men in a nation’s total population, the higher the level of male employment in that country. However, in North Africa, the male-to-female population ratio has a negative but insignificant effect on male employment. Economic growth In view of the various arguments about the effects of growth or development, we also entered the real GDP growth, in addition to real GDP per capita. The results show that real GDP growth was insignificant for three samples: all-Africa, Sub-Saharan Africa, and North Africa. These results were consistent with those of the ILO (2011) for developing economies. Subregional and oil effects The subregional fixed effects, which shift the intercepts, imply that Southern African countries, followed by those in East Africa, have lower male employment compared to Central and West Africa. Our results also show that net oil-exporting countries generally have systematically lesser male employment compared to net oil-importing countries in Africa. This suggests that, holding other factors constant, net oil-exporting countries experience lower levels of male employment than net oil-importing countries. In this sense, our results also suggest that oil-exporting nations have failed to fully utilize their huge oil revenues to create adequate jobs for their citizens. 6. CONCLUSION AND POLICY RECOMMENDATIONS Our empirical estimates, using available cross-sectional data over the period 1991 and 2009 suggest that in the all-African estimation, quadratic levels of real per capita GDP, greater access to credit by the private sector, more education, and higher male share of the population increase male employment while higher level of real GDP per capita, higher government consumption 36 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries expenditure, urban share of the population, and being a net oil-exporting country tend to lower it. In Sub-Saharan Africa, our results indicate that higher domestic investment, greater access to credit by the private sector, more education, and a higher male share of the population increase male employment while higher government consumption expenditure, greater openness of the economy, urban share of the population, and being a net oil-exporting country tend to lower it. However, North Africa is different. The North African results indicate that while the quadratic element of democracy, greater access to credit by the private sector, inflation, and greater openness of the economy increase male employment; the level of real GDP per capita, the level of democracy, increased inflow of foreign direct investment, and being a net oil-exporting country tend to lower male employment in the subregion. What are the implications of these results for African countries? Given our finding that government consumption expenditure reduces male employment across Africa and that domestic investment is not male employmentpromoting in the overall sample and particularly in North Africa, achieving investment effectiveness must remain an active goal of governments. All actors, from governments to civil society, must share responsibility for making investment more productive, efficient and effective, and for preventing one of its main breakdowns: corruption. Political will and good governance, strengthening accountability and transparency, as well as enlarging civil society space as a “watch dog” are critical in this direction. Attention should be paid to the design, implementation, and monitoring and evaluation phases of projects and programs. At the design stage, the aim should be to create achievable and quantifiable targets and to have all-stakeholder ownership through the collaboration of governments, the private sector, civil society, and other development agencies. All stakeholders must follow through to ensure that projects and programs are implemented as designed. Also, stakeholders must ensure that those projects and programs are regularly monitored and evaluated against indicators established in the design phase and agreed on by the development partners. Productive and efficient domestic investment requires the development of coordinated, objective, and transparent processes for decision-making based on thorough and rigorous cost-benefit analysis. Adoption of high-level best practice principles to inform the development of these processes will help African governments to achieve this. Those broad principles should include the following key elements: a nationally coordinated approach to the development of significant strategic projects and programs; the promotion of competitive markets; decision-making based on rigorous cost-benefit The African Statistical Journal, Volume 16, May 2013 37 John C. Anyanwu analysis to ensure the highest economic and social benefits to the nation over the long term; a commitment to transparency at all stages of the decisionmaking and project implementation processes; and a public sector financial management regime with clear accountabilities and responsibilities. To reduce waste, fraud, and corruption, all African countries should embrace and fully implement Transparency International’s (2009) “Integrity Pacts,” which embody rights and obligations to the effect that neither side in contracts will pay, offer, demand or accept bribes, or collude with competitors to obtain the contract, or while carrying it out. At the same time, efforts to reform the fiscal system for consolidation by both the executive and legislative arms of government are imperative to reduce government consumption expenditure to avoid waste, corruption and crowding out resources for public sector investment and employment creation. The current study suggests that, holding other factors constant, increasing levels of FDI are associated with decreasing male employment in North Africa while having an insignificant effect in the other samples. Thus, to promote male employment and ensure men have complete access to productive resources, African countries should regulate the inflow of foreign capital to ensure labor-intensive industries are not displaced by globalization. Further, to protect against threats to individual basic rights, the government should mandate that MNCs adhere to core labor standards, as provided by the International Labor Organization (ILO). Since labor-intensive employment represents a viable channel through which job-seekers are able to realize gains in real wages and social capital, the protection of these industries should be a policy priority for African countries. We find that access to credit by the private sector is essential to increasing men’s employment across the countries of Africa. African governments should start encouraging entrepreneurship and access to financing, for both men and women. The continent needs entrepreneurs ready and able to exploit new opportunities. Men and women need training in entrepreneurship and to be encouraged to take risks and start businesses and subsequently become employers themselves (WEF 2012a, 2012b); however, some deliberate and focused credit targeting will be of immense help in this direction. In fact, the promotion of domestic investment through improved credit conditions for small and medium enterprises, for example, would yield significant gains in employment. Effective policies that invest in the human capital of the workforce are needed. Policies that promote the upskilling, better training and education for the 38 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries low-skilled workforce are imperative. Educational reforms that conform to industry needs will also help to address the skills mismatches existing in many African countries. African governments need to dialogue with large employers as to how best to create employment for the men (and women) through strategic skills planning, skills development, and skills matching. Addressing the skills mismatch in the short-run will require improved training programs and closer links between tertiary and vocational educational institutions on the one hand, and the private sector on the other. Training programs should include on-the-job initiatives targeting those already working, as well as graduates with a general education who lack specific work skills. In addition, governments need to develop innovative public–private partnerships and the opportunities for collaboration among large employers, governments and other relevant stakeholders such as higher and vocational educational institutions to transform institutional structures and strengthen the region’s economy (Ncube and Anyanwu 2012). Indeed, stronger university–industry linkages are essential. This can be achieved by including private sector representatives in national education and training policy bodies and on academic boards involved in curriculum development. No doubt, this will also facilitate private sector funding for research, scholarships, internships, and apprenticeships. Our results also point to the need for enhanced government efforts, not just to increase human capital, especially for the men, but more importantly to reform the educational curriculum for better quality education and skills. We have shown in this study that being a net oil-exporting country decreases men’s employment in Africa. Thus, efficient management of oil and other natural resources in Africa requires actions throughout the value chain. In particular, a new natural resources management framework is needed for better governance, sectoral linkages, economic growth and human, capacity and infrastructure development – with strong parliamentary legislation, oversight, and representation throughout the resources value chain. Key effective natural resources management practices will require the following measures: • Enhanced good governance, especially as it relates to the way public money is spent, is a crucial factor in turning a natural resource boom into an opportunity for growth and development, job creation, and gender The African Statistical Journal, Volume 16, May 2013 39 John C. Anyanwu • • • • • • equality in Africa. Prioritizing the public investment management system is imperative. Checks and balances need to be maximized through parliaments. Integrating the extractive sector into national development frameworks – Revenue optimization needs to be integrated with the downstream sector. Value-addition and natural resources–industry linkages are paramount. There are many opportunities for improving positive linkages between the natural resources sector and development initiatives. Reinforcing institutional capacity and building strong and capable institutions. Sound fiscal policy and diversification of the economy, while using windfall taxes to protect against reneging on taxation. Full disclosure of terms of natural resources contracts is needed, as well as activating third-party brokers such as development partners (e.g. AfDB) and NGOs to ease information availability and reduce information asymmetry. African natural resource-rich countries should stop the practice of entering into bilateral development agreements with extractive companies for generous concessions in extraction contracts. Consequently, all contracts and terms should be legislated in the substantive law and implemented as such. EITI adherence: African countries’ company and financial laws should be reformed to require all extractive companies to use the EITI template in their annual financial reports by law. Without doubt, sustainable inclusive growth and development as well as inclusive governance should be the basis for policymakers and leaders in Africa to bridge the transformation between pain of current unemployment time bomb, especially in North Africa and net oil-exporting countries, and the promise of the future. 40 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries APPENDIX 1: Description of Variables Variable Source Employment ratios World Bank/International Labor Organization (ILO) database World Development Indicators World Development Indicators Polity IV Project World Development Indicators World Development Indicators World Development Indicators World Development Indicators World Development Indicators World Development Indicators World Development Indicators World Development Indicators Per capita GDP (constant 2000 US dollar) Credit to the private sector Democracy Trade openness ((imports + exports)/GDP) School enrollment rate Inflation (annual percentage change in CPI) Urban population ratio Population Domestic investment FDI Government consumption expenditures REFERENCES Aiyagari, R., Christiano, L., and Eichenbaum, M. (1990). “Output, Employment and Interest Rate Effects of Government Consumption,” Journal of Monetary Economics, 30: 73–86. Antecol, H. (2000). “An Examination of Cross-Country Differences in the Gender Gap in Labor Force Participation Rates,” Labour Economics, 7: 409–426. Anyanwu, J. C. (2012). “Youth Employment in Africa: Some Discussion Points.” Presentation at the High Level Policy Dialogue, Lusaka, July 2012. Anyanwu, J. C. and Erhijakpor, A. E. O. (2012). “The Effects of Youth Employment on Poverty in Africa: Cross-Country Evidence and Lessons for Nigeria.” Paper Presented at the 53rd Annual Conference of the Nigerian Economic Society, August 27–30, 2012. Asiedu, E. (2004). “The Determinants of Employment of Affiliates of US Multinational Enterprises in Africa,” Development Policy Review, 22 (4): 371–379. Blackburn, S. (2004). Women and the State in Modern Indonesia. Cambridge: Cambridge University Press. The African Statistical Journal, Volume 16, May 2013 41 John C. Anyanwu Bloom, D., Canning, D., Fink, G., and Finley, J. (2009). “Fertility, Female Labor Force Participation and the Demographic Dividend,” Journal of Economic Growth, 14: 79–101. Böserup, E. (1970). Women’s Role in Economic Development. New York: St. Martin’s. Campa, P. et al. (2011). “Gender Culture and Gender Gap in Employment,” CESifo Economic Studies, 57 (1) March: 156–182. Cardoso, E. (1992). “Inflation and Poverty,” Working Paper No. 4006. Cambridge, MA: National Bureau of Economic Research (NBER). Chaudhuri, S. (2009). “Economic Development and Women’s Empowerment.” Working Paper, University of Wisconsin – Eau Claire. Choudhry, M. T., Marelli, E., and Signorelli, M. (2012). “Youth Unemployment Rate and the Impact of Financial Crises,” International Journal of Manpower, 33 (1): 76–95. Christiano, L., Eichenbaum, M. and Rebelo, S. (2011). “When Is the Government Spending Multiplier Large?,” Journal of Political Economy, 119 (1): 78–121. Eastin, J. and Prakash, A. (2013). “Economic Development and Gender Equality: Is There a Gender Kuznets Curve?,” World Politics, 65 (1): 156–186. Ernesto, A-T. (2011). “The Impact of Trade Liberalization Policies and FDI on Gender Inequalities.” Background Paper for the World Development Report 2012. Washington, DC: World Bank. Fatás, A. and Mihov, I. (2002). “The Effects of Fiscal Policy on Consumption and Employment: Theory and Evidence,” CEPR Discussion Paper No. 2760. Fernández, R. (2010). “Does Culture Matter?”, NBER Working Paper No. 16277. Cambridge, MA: National Bureau for Economic Research. Fernández, R., and Fogli, A. (2005). “Fertility: The Role of Culture and Family Experience.” Cambridge, MA: National Bureau of Economic Research. 42 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries Fernández, R., Fogli, A., and Olivetti, C. (2004). “Mothers and Sons: Preference Formation and Female Labor Force Dynamics,” Quarterly Journal of Economics, 119 (4): 1249–1299. Forsythe, N., Korzeniewicz, R. P., and Durrant, V. (2000). “Gender Inequalities and Economic Growth: A Longitudinal Evaluation”, Economic Development and Cultural Change, 48 (3): 573–617. Gallaway, J. H. and Bernasek, H. (2004). “Literacy and Women’s Empowerment in Indonesia: Implications for Policy”, Journal of Economic Issues 38. Gandelman, N., and Rasteletti, A. (2012). “The Impact of Bank Credit on Employment Formality in Uruguay.” IDB Working paper Series No. IDB-WP-302, April. Goldin, C. (1995), “The U-shaped Female Labor Force Function in Economic Development and Economic History.” In T. P. Schultz (ed.), Investment in Women’s Human Capital. Chicago, IL: University of Chicago Press. Hegre, H., Ellingsen, T., Gates, S., and Gleditsch, N. P. (2001). “Towards a Democratic Civil Peace? Democracy, Political Change, and Civil War, 1816-1992,” American Political Science Review, 95 (1): 33–48. IFC (2012). “Towards a Democratic Civil Peace?” IFC Open Source Study (Meta-evaluation on job creation effects of private sector interventions: Summary of preliminary findings). Washington, DC: IFC. ILO (2011). World of Work Report 2011: Making Markets Work for Jobs. Geneva: ILO & IILS. ILO (2012a). Global Employment Trends: Preventing a Deeper Jobs Crisis. Geneva: ILO. ILO (2012b). Global Employment Trends for Youth 2012. Geneva: ILO. ILO (2012c). World of Work Report 2012: Better Jobs for a Better Economy. Geneva: ILO & IILS. Kehoe, T., and Serra-Puche, J. (1982). “The Employment Effects of Taxes and Government Expenditures in Mexico,” The Southwestern Review, Winter: 85–100. The African Statistical Journal, Volume 16, May 2013 43 John C. Anyanwu Mammen, K., and Paxson, C. (2000). “Women’s Work and Economic Development,” The Journal of Economic Perspectives, 14 (4): 141–164. Moghadam, V. (1994). “Introduction: women and identity politics in theoretical and comparative perspective.” In V. Moghadam (ed.), Identity Politics and Women: Cultural Reassertions and Feminisms in International Perspective. Boulder, CO: Westview. Ncube, M. and Anyanwu, J. C. (2012). “Inequality and the Arab Spring Revolutions in North Africa and the Middle East,” AfDB Africa Economic Brief, 3 (7), July. Niemi, B. T., and Lloyd, C. B. (1981). “Female Labor Supply in the Context of Inflation,” The American Economic Review, 71 (2): 70–75. Ogawa, K., and Akter, M. (2007). “Female Labor Force Participation in Indonesia,” Journal of International Cooperation Studies, 14 (3): 71–108. Oostendorp, R. (2009). “Globalization and the Gender Wage Gap,” World Bank Economic Review, Oxford University Press, 23 (1): 141–161. Parpart, J. L., Connelly, M. P., and Barriteau, V. E. (eds.) (2000). Theoretical Perspectives on Gender and Development. Ottawa: International Development Research Center. Psacharopoulos, G. and Tzannatos, Z. (1989). “Female Labor Force Participation: An International Perspective,” World Bank Research Observer, Oxford University Press, 4 (2): 187–201. Ramey, V. A. (2011a). “Identifying Government Spending Shocks: It’s All in the Timing,” Quarterly Journal of Economics, 126 (1): 1–50. Ramey, V. A. (2011b). “Can Government Purchases Stimulate the Economy?,” Journal of Economic Literature, 49 (3): 673–685. Richards, D. L., and Gelleny, R. (2007). “Women’s Status and Economic Globalization,” International Studies Quarterly 51 (4): 855–876. Sakellariou, C. (2011). “Determinants of the Gender Wage Gap and Female Labor Force Participation in EAP.” Paper commissioned for the EAP Gender Report. 44 Journal statistique africain, numéro 16, mai 2013 1. Driving factors of male employment in African countries Shukri, S. J. A. (1996). Arab Women: Unequal Partners in Development. Brookfield, VT: Avebury. Tam, H. (2011). “U-shaped Female Labor Force Participation with Economic Development: some panel data evidence,” Economic Letters, 110 (2): 140–142. Tseloni, A., Tsoukis, C., and Emmanouilides, C. (2011). “Globalization, Development, and Gender Inequality across the World: A Multivariate Multilevel Approach,” Quantitative and Qualitative Analysis in Social Sciences, 5 (1): 1–25. Ward, K. (1984). Women in the World Economic System: Its Impact on Status and Fertility. New York, NY: Praeger. WEF (2012a). World Economic Forum on the Middle East, North Africa and Eurasia: Bridging Regions in Transformation, Regional Agenda, Istanbul, Turkey, June 4–6, 2012. WEF (2012b). Addressing the 100 Million Youth Challenge: Perspectives on Youth Employment in the Arab World in 2012, Regional Agenda. Geneva: World Economic Forum. Wilson, D. J. (2012). “Government Spending: An Economic Boost?”, Federal Reserve Bank of San Francisco (FRBSF) Economic Letter 201204, February 6. Online at: http://www.frbsf.org/publications/economics/ letter/2012/el2012-04.html World Bank (2010). Women, Business and the Law, 2010. Washington, DC: World Bank. World Bank (2012a). World Development Report 2013: Jobs. Washington, DC: World Bank. World Bank (2012b). World Development Indicators 2012, Washington, DC: World Bank. The African Statistical Journal, Volume 16, May 2013 45 2. Alcohol consumption patterns: Do neighborhoods matter? A multilevel analysis of alcohol consumption patterns in Uganda Nazarius Mbona Tumwesigye,1 Richard Muwonge, 2 and Rogers Kasirye3 Abstract This paper presents a re-analysis of data on patterns of alcohol consumption with adjustment for cluster effects. The outcomes of interest were the consumption of alcohol at least three times a week and consuming 12 or more drinks in a single day. Multilevel regression analysis was applied using STATA V.10. Results show that there is a significant variation in the likelihood of drinking frequently at village level (p<0.01), but that this is confounded by the average proportion of drinkers. The variation in the likelihood of heavy drinking remained significant after controlling for background characteristics of respondents and village-level prevalence of drinking and unemployment. Cluster effects have an influence on the frequency and amount of alcohol consumption but their inclusion in statistical models does not affect the influence of established factors. Cluster analysis provides information on other correlates of frequent and heavy alcohol consumption. Key words: cluster effects, village effects, logistic regression, intracluster correlation, multilevel models, rho. Résumé Cet article présente une nouvelle analyse des données sur les schémas de consommation d’alcool, avec ajustement pour effets de grappe, et compare ces résultats avec ceux de la précédente analyse. Les résultats concernent la consommation régulière d’alcool au moins trois fois par semaine et l’absorption de douze boissons ou plus en une seule et même journée. L’analyse de régression multi-niveaux a été appliquée à l’aide de STATA V.10. Les résultats révèlent l’existence d’une variation significative dans la probabilité de boire fréquemment au niveau du village (p<0.01), mais cette variation se perd dans la proportion moyenne des consommateurs. La variation dans la probabilité d’une forte consommation d’alcool reste significative après neutralisation de certains facteurs comme le niveau d’instruction des personnes 1. Department of Epidemiology and Biostatistics, School of Public Health, Makerere University College of Health Sciences, Mulago Hill, P. O. Box 7072, Kampala, Uganda. Tel: +256-782-447771, e-mail: [email protected] 2. International Agency for Research on Cancer (IARC), 150 cours Albert Thomas, 69008 Lyon, France. Email: [email protected] 3. Uganda Youth Development Link, P.O. Box 12659, Kampala, Uganda. Email: [email protected] 46 Journal statistique africain, numéro 16, mai 2013