103072 The Labor Content of Exports in South Africa and Botswana – a preliminary exploration January 2015 Prepared by Massimiliano Calì and Claire Hollweg1 – Trade and Competitiveness Global Practice 1 Corresponding author: please contact on: chollweg@worldbank.org This is a Working Paper of the World Bank – it is being issued in an effort to share ongoing research. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent 1. Introduction and methodology2 We use the newly developed World Bank dataset on the labor content of exports (LACEX) to explore the possible implications of exports for shared prosperity in South Africa. The LACEX dataset has been recently assembled to compute the (direct and indirect) value of the compensation of employees linked to exports for each sector/country/year (Calì et al., 2015). The data has been computed on the basis of a panel of global input-output data spanning intermittent years from 1995 to 2007 from the Global Trade Analysis Project (GTAP). This represents a form of social accounting data - a variation on the social accounting matrix (SAM) where incomes are shown in the rows of the SAM while expenditures are shown in the columns (See Hertel 2013, McDougal 2001). The structure of the data provides a comprehensive and consistent record of national income accounting relationships between different sectors and regions, including intermediate and final demand linkages. This structure of the dataset allows one to obtain the value added content of final output and exports, including its compensation of employees’ component. That includes both the direct and indirect compensation, based on the backward linkages of each sector with the rest of the economy. In order to obtain these labor value added measures, two intermediate multiplier matrixes need to be calculated (see Appendix 1 for more details on these calculations). The first is the Leontief inverse matrix, which measures the inputs contained in a unit of final output. This matrix contains both direct and indirect inputs. Next, one needs to calculate a matrix which has the compensation of employees’ shares of total output. Using these two matrixes as multipliers one can obtain the compensation of employees’ shares of exports and final outputs. These shares are also split between skilled and unskilled workers. The data are then matched with the UNIDO industrial statistics, which contains data for a number of countries, including South Africa, on employment at the 4-digit ISIC sector. This allows isolating the jobs (JOCEX) component of LACEX 2 The authors thank Thomas Farole for very useful inputs and Alen Mulabdic for excellent research assistance 1 2. South Africa The LACEX data shows that in South Africa the labor value added (in nominal terms) directly contained in exports (i.e. the wages paid to produce the exports directly) has been increasing robustly since 2001 ( Figure 1) (the CAGR between 2001 and 2011 was 7.68 percent). The growth in the total labor value added in exports (i.e. all the wages paid to produce exports including those for the domestic inputs of exports) has been more rapid (the CAGR between 2001 and 2011 was 7.89 percent), reaching $ 35 billion in 2011. Figure 1: Direct and total labor content of exports have been growing steadily in South Africa, 1995-2011 South Africa 40000 30000 US$ Mil 20000 10000 2000 2005 2010 year Direct VA Total VA Source: Authors’ elaboration on the Labor content of exports dataset This growth is not reflected in the growth of the share of total labor content in gross exports ( Figure 2), which has in fact declined since 2011 (figure 2). In 2001 each $ 100 of exports generated $41 of domestic wages, while by 2011 this figure had reduced to $33 (of which $14.5 directly and $18 via backward linkages). This $33 is well below that of China and Brazil and it is in line with India (which has followed a similar trajectory to that of South Africa). The only BRICS country with a share below that of South Africa is Russia. On the other hand South Africa has a higher labor content than other middle-income countries, including Colombia, Chile, Thailand and Turkey, while Peru has a higher content. However, the labor value added in exports has increased in these countries (except Colombia) relatively to that of South Africa, which may be a sign of concern going forward. Table A1 at the end reports also the split of the total labor value added by skilled and unskilled labor. The latter dominates in these countries to a larger extent than in the rest of the world, although South Africa has a slightly higher share of skilled labor in exports than its comparators. 2 Figure 2: South Africa has a lower share of total labor content in gross exports than most of the BRIC (panel a) 45 40 35 30 25 20 1995 2000 2005 2010 Year ZAF BRA RUS IND CHN But it has a higher share than most other comparators (panel b) 50 40 30 20 10 1995 2000 2005 2010 Year ZAF COL CHL PER THA TUR Source: Authors’ elaboration on the Labor content of exports dataset 3 Not only has the labor contained in exports been declining relative to gross exports, but it has also declined relative to the total domestic labor value added (Figure 3). The slow growth in the labor content of exports is mainly due to the fact that the export sector has been slowly shedding jobs both directly and indirectly. The total number of jobs associated to exports was just below 3 million in 2011 (of which 1 million are the direct jobs) from just above 3 million in 2004. Figure 3: Total labor value added in exports has grown less strongly than that in total production ($ million) South Africa 200000 150000 100000 50000 0 2000 2005 2010 year Total Labor Value Added Total Labor Value Added in Exports This trend has translated into a reduction of the jobs contained in exports both directly and total (Figure 4). South Africa moved from having 22 jobs per $ million exports in 2004 to around 17 in 2011. While this is a sizable reduction, it is less dramatic than most other comparator countries. This generalized decrease in the number of jobs in exports is consistent with labor saving technological changes across developing countries. Despite this smaller drop in jobs content, South Africa remains one of the middle income countries in the sample with the lowest number of jobs associated with exports. That is partly a composition effect as South African exports are dominated by capital intensive mining sector. We will control for this effect in the analysis below by comparing similar sectors across countries. 4 Figure 4: Direct and total number of jobs contained in exports have been falling, 2004 and 2011 Direct Number of Jobs 40 30 20 10 0 1995 2000 2005 2010 year ZAF BRA RUS COL PER THA TUR Total Number of Jobs 80 60 40 20 0 1995 2000 2005 2010 year ZAF BRA RUS COL PER THA TUR Source: Authors’ elaboration on the Labor content of exports dataset and UNIDO On the other hand South Africa has a higher share of jobs contained in exports in total employment than any other comparator countries barred Thailand ( Figure 5). That is despite the reduction in this share. While in 2004 more than 1 in 4 jobs was directly or indirectly associated with exports, by 2011 this ratio declined to just above 1 in 5 jobs ( Figure 5). 5 Figure 5: Direct and total number of jobs in exports as a share of total employment, 2004 and 2011 Direct Number of Jobs 15 10 5 1995 2000 2005 2010 year ZAF BRA RUS COL PER THA TUR Total Number of Jobs 50 40 30 20 10 1995 2000 2005 2010 year ZAF BRA RUS COL PER THA TUR Source: Author’s elaboration on the Labor content of exports dataset and UNIDO What sectors explain these results? Figure 6 shows that in 2011 the leading export sectors in terms of labor content were minerals, metals, machinery and ferrous metals. With over $ 6 billion minerals is by far the largest contributor to wages among export sectors, reflecting the large value of gross exports and a relatively high labor content in exports (more on this below). However the other main sectors all have a higher share of labor content explained by domestic backward linkages than the mineral sector, suggesting higher extent of interconnectedness with domestic suppliers. For example around three quarter of the total labor contained in exports of machinery and equipment is due to labor used in domestic suppliers of this sector. In the case of minerals that share is less than one third. 6 Figure 6 also shows how the labor intensity of exports varies also across both directly and indirectly. For example each $100 of wearing apparel exports generates $43 of wages, of which $18 go the workers directly producing these services. At the lower end of the spectrum $100 of metal exports generate only $21 in labor value added, split equally between direct and indirect content. Figure 6: Direct and total labor value added in exports in 2011 ($ milion) South Africa 2011, backward .8 6000 Labor VA-export share .6 4000 .4 2000 .2 0 0 PubAdmin/Defence/Health/Educat Paper Products, Publishing Wearing Apparel Dwellings Chemical, Rubber, Plastic Products Machinery and Equipment nec Electricity, Gas, Water Mineral Products nec Metal Products Transport Equipment Minerals nec Wood Products Energy Extraction Metals nec Textiles Construction Other Private Services Leather Products Manufactures nec Ferrous Metals Processed Foods Beverages and Tobacco Products Trade and Transport Services Agr, Forestry, Fisheries Direct Labor VA in Exports Indirect Labor VA in Exports Total Labor VA in Exports to Gross Exports In Figure 8 we split the total labor content of exports for 2011 into skilled and unskilled labor. While the unskilled component is larger in most sectors, there is some cross-sectoral variation. In public administration, defense, health and education and in other private services exports the split is even or almost even (either when considering total value added computed through the backward or forward linkages method). At the other extreme agriculture, forestry and fishing have the lowest shares of skilled labor contained in total exports. Figure 7 shows the counterpart of these numbers, i.e. how much the labor in each sector is used by the other exporting sectors (i.e. the forward linkages). The picture is very different here, highlighting an asymmetry that is typical of many economies. The services sectors, while being relatively small contributors to the direct labor content of exports, are the most important contributor of labor to exports when accounting also for the forward linkages. Trade and transport services as well as other private services pay for over $8 and $7 billion in wages mostly for supplying services to export sectors. On the other hand labor in neither minerals nor metals is employed to produce inputs for other exporting sectors suggesting very limited downstream connectivity with the rest of the economy. Perhaps surprising also labor in machinery and equipment is not involved in the production to supply other exporting sectors. The majority of its wages are paid to employees producing equipment to be directly exported. 7 In Figure 8 we split the total labor content of exports for 2011 into skilled and unskilled labor.3 While the unskilled component is larger in most sectors, there is some cross-sectoral variation. In public administration, defense, health and education and in other private services exports the split is even or almost even (either when considering total value added computed through the backward or forward linkages method). At the other extreme agriculture, forestry and fishing have the lowest shares of skilled labor contained in total exports. Figure 7: Direct and total labor value added in exports in 2011, forward linkages ($ million) South Africa 2011, forward 8000 3 Labor VA-export share 6000 2 4000 1 2000 0 0 PubAdmin/Defence/Health/Educat Paper Products, Publishing Wearing Apparel Dwellings Chemical, Rubber, Plastic Products Machinery and Equipment nec Electricity, Gas, Water Mineral Products nec Metal Products Transport Equipment Minerals nec Wood Products Energy Extraction Metals nec Textiles Construction Other Private Services Leather Products Manufactures nec Ferrous Metals Processed Foods Beverages and Tobacco Products Trade and Transport Services Agr, Forestry, Fisheries Direct Labor VA in Exports Indirect Labor VA in Exports Total Labor VA in Exports to Gross Exports Figure 8: Skilled and unskilled total labor value added in exports in 2011 ($ milion) South Africa 2011, backward .8 6000 Labor VA-export share .6 4000 .4 2000 .2 0 0 PubAdmin/Defence/Health/Educat Paper Products, Publishing Wearing Apparel Dwellings Chemical, Rubber, Plastic Products Machinery and Equipment nec Electricity, Gas, Water Mineral Products nec Metal Products Minerals nec Wood Products Transport Equipment Energy Extraction Metals nec Textiles Construction Other Private Services Leather Products Manufactures nec Ferrous Metals Processed Foods Beverages and Tobacco Products Trade and Transport Services Agr, Forestry, Fisheries Total VA Unskilled Total VA Skilled Total Labor VA in Exports to Gross Exports 3 We use 2007 as the split for 2011 is not compatible with previous years. 8 Figure 9 shows how these sectoral numbers compare to other comparators. In general South Africa has a lower share of labor in exports than its BRIC comparators (first panel), while it has a generally higher share across sectors relative to the other comparator countries. In Figure 10 we show this comparison across all countries for the 3 largest South African exporters for 2001 and 2011. In machinery and equipment, South Africa’s labor content as a share of exports declined relative to the average in the other countries, although it remains still slightly above the level expected given its income per capita. In metals the country has moved from being above to being below the expected share of labor share in export and in 2011 it was also below most comparator countries. In minerals on the other hand South Africa’s position has improved and by 2011 it had the highest share of labor in exports among the comparator countries. Finally Figure 11 shows the number of direct and indirect jobs in exports for the macro sectors of the economy in 2011. Manufacturing dominates the ranking with almost two third of the jobs associated with exports, the vast majority of them in sectors providing inputs to it. Figure 9: Total labor value added in $1 export across sectors and countries Total Labor VA in Export (backward) to Gross Exports (2011) Agr, Forestry, Fisheries Beverages and Tobacco Products Chemical, Rubber, Plastic Products Construction Dwellings Electricity, Gas, Water Energy Extraction Ferrous Metals Leather Products Machinery and Equipment nec Manufactures nec Metal Products Metals nec Mineral Products nec Minerals nec Other Private Services Paper Products, Publishing Processed Foods PubAdmin/Defence/Health/Educat Textiles Trade and Transport Services Transport Equipment Wearing Apparel Wood Products 0 .2 .4 .6 .8 Percent ZAF BRA RUS IND CHN 9 Total Labor VA in Export (backward) to Gross Exports (2011) Agr, Forestry, Fisheries Beverages and Tobacco Products Chemical, Rubber, Plastic Products Construction Dwellings Electricity, Gas, Water Energy Extraction Ferrous Metals Leather Products Machinery and Equipment nec Manufactures nec Metal Products Metals nec Mineral Products nec Minerals nec Other Private Services Paper Products, Publishing Processed Foods PubAdmin/Defence/Health/Educat Textiles Trade and Transport Services Transport Equipment Wearing Apparel Wood Products 0 .2 .4 .6 .8 Percent ZAF COL CHL PER THA TUR Figure 10: Total labor value added in $1 export (by backward linkages) across sectors and countries Total Labor VA in Export (backward) to Gross Exports (2011) Machinery and Equipment nec, 2001 .6 Machinery and Equipment nec, 2011 .8 ZAF COL RUS .4 .6 BRA PER CHN CHL BRA RUS TUR .4 CHNZAF IND COL CHL .2 IND .2 TUR THA PER THA 0 0 4 6 8 10 12 4 6 8 10 12 GDP per capita (current USD) (2011) GDP per capita (current USD) (2011) 10 Total Labor VA in Export (backward) to Gross Exports (2011) Metals nec, 2001 Metals nec, 2011 1 1 .8 .8 .6 .6 COL PER .4 .4 CHN ZAF RUS BRA CHN BRA RUS IND PER COL CHL CHL ZAF .2 .2 TUR IND THA TUR THA 0 0 4 6 8 10 12 4 6 8 10 12 GDP per capita (current USD) (2011) GDP per capita (current USD) (2011) Minerals nec, 2001 .6 Minerals nec, 2001 .8 COL ZAF .4 PER .6 CHN BRA TURRUS BRA IND TUR COL .4 CHN RUS THA ZAF CHL .2 IND PER CHL .2 THA 0 0 4 6 8 10 12 4 6 8 10 12 GDP per capita (current USD) (2011) GDP per capita (current USD) (2011) Figure 11: Number of jobs in exports in 2011 across macro sectors (in ‘000) South Africa 2011 Manufacturing Mining and Energy Other commercial services Agriculture Transport and communication Trade, sales and accomodation Public services Financial services Electricity, gas and water Real estate, renting Construction 0 500 1,000 1,500 2,000 Direct Jobs in Export Indirect Jobs in Export 11 As noted above, the low labor content of exports in South Africa relatively to other comparator countries may be due to a composition effect. To check that, we compare a number of narrowly defined sectors across countries. In particular we focus on the main Global Value Chain-oriented sectors in the country, i.e. wearing apparel and motor vehicles, other food and electronic equipment (Figure 12). The labor contained in South African exports of wearing apparel has increased only slightly over the 2000s, the product of a substantial increase in the indirect content and a decrease in the direct content of exports. The increase brings South African total content of exports ahead of most other comparator countries. In 2011 only Brazil and China had a higher content. The direct labor content drives the large total content in Brazil apparel exports, while it is the indirect content behind China’s labor intensity. Not surprisingly the labor content is generally lower in motor vehicles exports. South Africa has experienced a decrease between 2001 and 2011, when just below $30 of labor was contained in $100 of motor vehicle exports. This figure is well below that of Brazil, China, where the labor content has increased and India, where it has remained stable. The drop in South Africa is entirely explained by the direct content of exports, which more than halved in a decade. A similar picture applies to the ‘other food’ sector, where the labor value added in South African exports has slightly declined in 2001-2011 and it lags behind Brazil, China and India. While most of the value added is indirect across countries, most comparator countries have a higher indirect content per $100 of exports than South Africa. Finally the labor contained in electronic equipment exports has dropped substantially in South Africa between 2001 and 2011, with a dramatic drop in the direct content and an increase in indirect content. Despite this drop South African labor content of electronic exports remain relatively high and behind only Peru and Brazil. This analysis suggests that the relatively low labor content of South African exports seems to be due not only to a composition effect, but also to a generally low use of domestic labor in production. Figure 12: Direct and indirect labor value added in each $100 of exports (including backward linkages) Wearing Apparel Motor Vehicles 2001 2001 BRA 2011 BRA 2011 2001 2001 CHL 2011 CHL 2011 2001 2001 CHN 2011 CHN 2011 2001 2001 COL 2011 COL 2011 2001 2001 IND 2011 IND 2011 2001 2001 PER 2011 PER 2011 2001 2001 RUS 2011 RUS 2011 2001 2001 THA 2011 THA 2011 2001 2001 TUR 2011 TUR 2011 2001 2001 ZAF 2011 ZAF 2011 0 20 40 60 0 20 40 60 80 Direct VA Indirect VA Direct VA Indirect VA 12 Other Food Electronic Equipment 2001 2001 BRA 2011 BRA 2011 2001 2001 CHL 2011 CHL 2011 2001 2001 CHN 2011 CHN 2011 2001 2001 COL 2011 COL 2011 2001 2001 IND 2011 IND 2011 2001 2001 PER 2011 PER 2011 2001 2001 RUS 2011 RUS 2011 2001 2001 THA 2011 THA 2011 2001 2001 TUR 2011 TUR 2011 2001 2001 ZAF 2011 ZAF 2011 0 10 20 30 40 50 0 20 40 60 Direct VA Indirect VA Direct VA Indirect VA In Figure 13, we replicate the same graphs but with the skilled-unskilled instead of the direct-indirect split. The share of skilled workers appears to be generally higher in South Africa than in most other comparators in all 4 sectors except motor vehicles. However this share is somewhat declined in electronic exports, while it has increased in wearing apparel. Figure 13: Unskilled and skilled indirect labor value added in each $100 of exports (including backward linkages) Wearing Apparel Motor Vehicles 2001 2001 BRA 2011 BRA 2011 2001 2001 CHL 2011 CHL 2011 2001 2001 CHN 2011 CHN 2011 2001 2001 COL 2011 COL 2011 2001 2001 IND 2011 IND 2011 2001 2001 PER 2011 PER 2011 2001 2001 RUS 2011 RUS 2011 2001 2001 THA 2011 THA 2011 2001 2001 TUR 2011 TUR 2011 2001 2001 ZAF 2011 ZAF 2011 0 20 40 60 0 20 40 60 80 Total VA Unskilled Total VA Skilled Total VA Unskilled Total VA Skilled Other Food Electronic Equipment 2001 2001 BRA 2011 BRA 2011 2001 2001 CHL 2011 CHL 2011 2001 2001 CHN 2011 CHN 2011 2001 2001 COL 2011 COL 2011 2001 2001 IND 2011 IND 2011 2001 2001 PER 2011 PER 2011 2001 2001 RUS 2011 RUS 2011 2001 2001 THA 2011 THA 2011 2001 2001 TUR 2011 TUR 2011 2001 2001 ZAF 2011 ZAF 2011 0 10 20 30 40 50 0 20 40 60 Total VA Unskilled Total VA Skilled Total VA Unskilled Total VA Skilled 13 3. Botswana We first focus on the total labor value added contained in exports, i.e. the wages paid to produce the exports directly as well as the wages paid to produce the domestic inputs used by the exports. Our data shows that this has been increasing in the late 1990s through 2001 but it has grown very little since then, hovering around US$ 1.5 billion (Figure 14).4 Most of this evolution is determined by that of the direct labor value added in exports (the wages to directly produce the exported goods), which now comprise over 90% of the total labor value added. This suggests that backward linkages are very weak in Botswana. Figure 14: Direct and total labor content of exports in Botswana, 1995-2011 Botswana 2000 1500 US$ Mil 1000 500 0 1995 2000 2005 2010 year Direct VA Total VA Source: Authors’ elaboration on the Labor content of exports dataset This growth through peaks and troughs is accentuated when looking at the evolution of the share of total labor content in gross exports (Figure 15). At the end of the 1990s it grew rapidly to reach 43% in 2001 (i.e. each $ 100 of exports generate $43 in domestic wages), the highest share among the set of comparator middle-income countries that we use.5 The labor share in gross exports eventually declined equally rapidly and by 2011 it was at 23%, the lowest share among all comparator countries but Costa Rica. Table A1 at the end reports also the split of the total labor value added by skilled and unskilled labor. The latter dominates in the set of countries we consider to a larger extent than in the rest of the world for which we have data. That is unsurprising as the coverage of countries in the dataset is skewed against low income countries, thus middle income countries are at the lower end of the income distribution in this sample. Botswana has a lower skilled share in labor value added of exports vis-à-vis all the comparator countries except Namibia. The growth in the total labor value added in exports is considerably less sustained than the overall growth in labor value added (Figure 16). Thus the share of labor value added explained by exports has been declining especially since 2001. 4 The CAGR between 1997 and 2011 for direct VA is: 13.25 percent. The CAGR between 1997 and 2011 for total VA is: 11.96 percent. The CAGR between 2001 and 2011 for direct VA is: 2.12 percent. The CAGR between 2001 and 2011 for total VA is: -0.15 percent. 5 This includes South Africa, Namibia, Chile, Costa Rica, Kazakhstan, Malaysia and Mauritius. 14 Figure 15: The rise and fall of Botswana’s labor content in exports 50 40 30 20 10 1995 2000 2005 2010 Year BWA ZAF NAM CHL CRI KAZ MYS MUS Source: Authors’ elaboration on the Labor content of exports dataset Figure 16: Total labor value added in exports and has grown less than that in total production ($ million) Botswana 6000 4000 2000 0 1995 2000 2005 2010 year Total Labor Value Added Total Labor Value Added in Exports Source: Authors’ elaboration on the Labor content of exports dataset What sectors drive these results? Figure 17 shows that in 2011 the leading export sectors in terms of total labor content (computed on the basis of the direct content and backward linkages) were minerals and metals (at around US$ 370 million). While they both have a similar total labor value added in exports, minerals stimulate a larger labor value added among domestic suppliers. Metals have virtually no domestic backward linkages. The third largest contributor to labor content in export is manufacturing, with half of its contribution coming from backward linkages. This makes manufacturing a more important sector in terms of overall labor value added in exports despite having smaller direct labor content of exports than private services exports. 15 Figure 17 also shows that the labor intensity of exports varies also across sectors both directly and indirectly. Each $100 of paper product exports generate close to $45 of labor value added.6 At the other end of the spectrum minerals generate very little value added, less than $10 every $100 of exports. The median and average value of labor share in export across export sectors are around 40%. That is well below the overall share of labor in exports in 2011 (23%) as that is brought down by the low share of the largest export, i.e. minerals. Figure 17: Direct and total labor value added (backward linkages) in exports in 2011 ($ million) Botswana 2011, backward 400 1 .8 Labor VA-export share 300 .6 200 .4 100 .2 0 0 PubAdmin/Defence/Health/Educat Paper Products, Publishing Wearing Apparel Dwellings Chemical, Rubber, Plastic Products Machinery and Equipment nec Electricity, Gas, Water Mineral Products nec Metal Products Transport Equipment Minerals nec Energy Extraction Wood Products Metals nec Textiles Construction Other Private Services Leather Products Manufactures nec Ferrous Metals Processed Foods Beverages and Tobacco Products Trade and Transport Services Agr, Forestry, Fisheries Direct Labor VA in Exports Indirect Labor VA in Exports Total Labor VA in Exports to Gross Exports In Figure 18 we re-compute the total labor value added in exports on the basis of forward linkages, i.e. how much labor in each sector is utilized in its own export and of those of other sectors. Metals and minerals are again the largest contributors due to their relatively large direct labor content. However they are virtually not used by other domestic exporters. Their entire production is just exported. On the other hand around half of the labor value added of services, both other private and trade and transport, is contained in other sectors’ exports (i.e. close to $ 100 million each). Next in Figure 19 we show the split in skilled and unskilled labor. While unskilled labor dominates in all sectors, the share of skilled labor in other private services is larger than that in the other sectors. Among the large exporters, metals and minerals have similar split. 6 The large labor contained in metals is most likely due to an error in the data. 16 100 200 300 400 0 100 200 300 400 Dwellings 0 Energy Extraction Leather Products Ferrous Metals Mineral Products nec Paper Products, Publishing Dwellings Wood Products Ferrous Metals Metal Products Electricity, Gas, Water Leather Products Construction Agr, Forestry, Fisheries Transport Equipment Beverages and Tobacco Products Total VA Skilled Wearing Apparel Labor VA-export share Beverages and Tobacco Products Total VA Unskilled Textiles Wood Products Chemical, Rubber, Plastic Products Transport Equipment Textiles Processed Foods Machinery and Equipment nec Wearing Apparel PubAdmin/Defence/Health/Educat Paper Products, Publishing Trade and Transport Services Other Private Services Mineral Products nec Botswana 2011, backward Manufactures nec Metals nec Metal Products Total Labor VA in Exports to Gross Exports Minerals nec Energy Extraction Chemical, Rubber, Plastic Products Labor VA-export share 0 .2 .4 .6 .8 1 Machinery and Equipment nec Construction Agr, Forestry, Fisheries Direct Labor VA in Exports Electricity, Gas, Water Indirect Labor VA in Exports Processed Foods PubAdmin/Defence/Health/Educat Manufactures nec Trade and Transport Services Botswana 2011, forward Other Private Services Minerals nec Total Labor VA in Exports to Gross Exports Metals nec 0 50 100 150 Figure 18: Direct and total labor value added (forward linkages) in exports in 2011 ($ million) Figure 19: Skilled and unskilled total labor value added in exports (backward linkages) in 2011 ($ mln) 17 Figure 20 and Figure 21 show how these sectoral numbers compare to other countries. The first thing to notice in Figure 20 is that Botswana has the lowest total labor value added per dollar of mineral exports. That is partly due to the very low backward linkages of the sector relative to the other countries (Figure 21). On the other hand manufacturing in Botswana has a larger domestic labor content in exports than all other countries but Kazakhstan, and it has also the largest labor content in textiles, leather products and metals. However by and large the total over the direct labor content of exports (an indication of the relative importance of the backward linkages) in Botswana is lower than its comparators. This suggests little interconnectedness between the exporters and the rest of the economy in all sectors. Figure 20: Total labor value added in $1 export across sectors and countries Total Labor VA in Export (backward) to Gross Exports (2011) Agr, Forestry, Fisheries Beverages and Tobacco Products Chemical, Rubber, Plastic Products Construction Dwellings Electricity, Gas, Water Energy Extraction Ferrous Metals Leather Products Machinery and Equipment nec Manufactures nec Metal Products Metals nec Mineral Products nec Minerals nec Other Private Services Paper Products, Publishing Processed Foods PubAdmin/Defence/Health/Educat Textiles Trade and Transport Services Transport Equipment Wearing Apparel Wood Products 0 .2 .4 .6 .8 1 Percent BWA ZAF NAM CHL CRI KAZ MYS MUS 18 Figure 21: Total over direct labor value added in exports across sectors and countries Total (backward) over Direct Labor VA in Export (2011) Agr, Forestry, Fisheries Beverages and Tobacco Products Chemical, Rubber, Plastic Products Construction Dwellings Electricity, Gas, Water Energy Extraction Ferrous Metals Leather Products Machinery and Equipment nec Manufactures nec Metal Products Metals nec Mineral Products nec Minerals nec Other Private Services Paper Products, Publishing Processed Foods PubAdmin/Defence/Health/Educat Textiles Trade and Transport Services Transport Equipment Wearing Apparel Wood Products 0 5 10 15 Ratio BWA ZAF NAM CHL CRI KAZ MYS MUS In Figure 22 and Figure 23, we focus on the main Global Value Chain-oriented sectors in the country, i.e. wearing apparel, motor vehicles, other food and electronic equipment. 19 Figure 22: Direct and indirect labor value added in each $100 of exports (including backward linkages) Wearing Apparel Motor Vehicles 2001 2001 BWA BWA 2011 2011 2001 2001 CHL CHL 2011 2011 2001 2001 MYS MYS 2011 2011 2001 2001 ZAF ZAF 2011 2011 0 10 20 30 40 50 0 10 20 30 40 Direct VA Indirect VA Direct VA Indirect VA Other Food Motor Vehicles 2001 2001 BWA BWA 2011 2011 2001 2001 CHL CHL 2011 2011 2001 2001 MYS MYS 2011 2011 2001 2001 ZAF ZAF 2011 2011 0 20 40 60 0 10 20 30 40 Direct VA Indirect VA Direct VA Indirect VA Figure 23: Unskilled and skilled indirect labor value added in each $100 of exports (including backward linkages) Wearing Apparel Motor Vehicles 2001 2001 BWA BWA 2011 2011 2001 2001 CHL CHL 2011 2011 2001 2001 MYS MYS 2011 2011 2001 2001 ZAF ZAF 2011 2011 0 10 20 30 40 50 0 10 20 30 40 Total VA Unskilled Total VA Skilled Total VA Unskilled Total VA Skilled Other Food Electronic Equipment 2001 2001 BWA BWA 2011 2011 2001 2001 CHL CHL 2011 2011 2001 2001 MYS MYS 2011 2011 2001 2001 ZAF ZAF 2011 2011 0 20 40 60 0 20 40 60 Total VA Unskilled Total VA Skilled Total VA Unskilled Total VA Skilled 20 References Calì, M., J. Francois, M. Manchin, D. Oberdabernig, H. Rojas-Romagosa and P. Tomberger (2015). The Labor Content of Exports, mimeo, The World Bank. Francois, J., M. Manchin and P. Tomberger (2013). Services Linkages and the Value Added Content of Trade, World Bank Policy Research Working Paper. Hertel, T. W. (2013). Global Applied General Equilibrium Analysis Using the Global Trade Analysis Project Framework, in P. Dixon and D. Jorgenson eds., Handbook of Computable General Equilibrium Modeling, Elsevier: North Holland. McDougall, R., ed. (2001). The GTAP database – version 5, Global Trade Analysis Center: Purdue University. 21 Appendix 1 – calculating direct and indirect value added in exports This section illustrates how to generate the (direct and indirect) value added associated with each sector- country from Input-Output tables in GTAP (based on Francois et al., 2013). That forms the basis for the computation of the compensation of employees. We begin by denoting a representation of intermediate and final demands as follows: Y = Z - AZ (1) where the term Y denotes a final demand vector, Z denotes a gross output vector, and A denotes a matrix of intermediate use coefficients. Equation (1) therefore defines final output with respect to intermediate input requirements. With some manipulation we arrive at the Leontief inverse matrix, also known as the multiplier matrix M. Z = (I - A)-1 Y = MY (2) The multiplier matrix M measures the inputs contained in a unit of final output. If one assigns the sector indexes i,j to the A and M matrices, then a representative element of the M matrix Mij gives the direct and indirect inputs (and thus the sector i receipts) linked to each unit (for example each dollar) of sector j receipts in the data. This implies real production activities measured by value of output. For our purposes, it provides a means to trace, through these income flows, the flow of gross activity and value added from intermediate to final goods and services, ostensibly across borders as well as sectors. Because linkages will vary by industry, each industry will be characterized by different multipliers. To focus on value added, we note first that in terms of gross output values Z, some share of this involves value added within each sector. We define ̂ as the diagonal matrix indexed over i, j with diagonal elements equal to the value added shares of output Z. We then use M to provide a breakdown of the flow of value added across activities in the form of the matrix V. ̂ = (3) Similar to the Leontief inverse matrix itself, the V matrix identifies the inputs of value added in each sector related to a unit of final demand. If we multiply V by the diagonal matrix ̂ whose non-zero elements are the vector of final outputs, the matrix yields a breakdown of economy-wide value added (the primary component of Gross National Product on a source basis). Similarly, if we multiply V by the diagonal matrix ̂ whose non-zero elements are the national export vector, we can recover the value added content of exports X (both direct and indirect). ̂ = (4) ̂ = (5) The G matrix and the H matrix give us the set of linkages, both direct and indirect, between value added across sectors. We are interested in the compensation of employees’ part of the value added, which forms the basis of the computation of LACEX. Appendix 2 – additional charts Table A1: Total labor value added in each $100 of exports (including backward linkages) - South Africa 1995 1997 2001 2004 2007 2011 22 ZAF Skilled 9.03 8.11 7.55 7.69 Unskilled 31.57 25.14 23.32 24.94 Total 40.60 33.25 30.87 32.63 BRA Skilled 6.88 8.55 11.51 8.82 9.30 8.69 Unskilled 26.82 29.67 32.00 32.39 32.48 32.47 Total 33.70 38.22 43.51 41.21 41.78 41.16 RUS Skilled 6.83 6.12 4.87 5.15 5.15 Unskilled 29.82 27.10 18.71 18.87 19.54 Total 36.65 33.22 23.58 24.02 24.69 IND Skilled 8.61 5.97 6.83 7.47 7.62 7.04 Unskilled 32.97 27.82 28.96 25.77 23.81 25.34 Total 41.58 33.79 35.79 33.24 31.43 32.38 CHN Skilled 4.15 5.60 5.89 5.55 5.87 7.43 Unskilled 22.77 31.43 31.88 27.10 27.92 34.22 Total 26.92 37.03 37.77 32.65 33.79 41.65 COL Skilled 7.20 8.74 7.82 4.30 4.43 4.55 Unskilled 35.26 35.52 33.97 20.37 20.26 21.87 Total 42.46 44.26 41.79 24.67 24.69 26.42 CHL Skilled 5.28 5.63 6.30 6.44 6.33 6.56 Unskilled 21.12 22.63 22.48 21.46 20.60 21.79 Total 26.40 28.26 28.78 27.90 26.93 28.35 PER Skilled 5.50 5.44 8.78 8.39 8.13 Unskilled 20.39 19.92 38.16 36.43 37.45 Total 25.89 25.36 46.94 44.82 45.58 THA Skilled 2.85 3.68 2.91 3.43 4.00 3.51 Unskilled 15.47 13.93 12.74 13.36 15.11 14.15 Total 18.32 17.61 15.65 16.79 19.11 17.66 TUR Skilled 3.84 5.13 6.78 5.62 5.32 4.73 Unskilled 17.06 18.57 26.31 19.72 18.94 18.04 Total 20.90 23.70 33.09 25.34 24.26 22.77 WLD Skilled 15 13 13 12 12 Unskilled 30 27 26 25 24 Total 45 41 39 37 35 36 Table A2: Non labor value added in each $100 of exports – South Africa versus peers 1997 2001 2004 2007 2011 ZAF 38.8 44.6 42.5 44.9 BRA 50.1 39.3 43.3 44.1 47.4 RUS 53.7 53.5 64.5 65.8 66.5 IND 49.5 47.5 45.3 43.1 42.5 CHN 36.4 37.0 37.1 38.7 36.7 COL 44.6 44.0 62.5 61.9 61.2 CHL 51.4 47.1 46.1 48.6 49.6 PER 59.6 60.1 43.9 44.2 43.3 THA 46.4 41.5 38.2 39.8 37.3 TUR 50.1 41.8 50.4 47.9 45.8 23 Table A3: Total labor value added in each $100 of exports (including backward linkages) - Botswana 1995 1997 2001 2004 2007 2011 BWA Skilled 3.13 7.91 5.29 6.63 Unskilled 11.79 35.47 18.29 25.92 Total 14.92 43.38 23.58 32.55 22.82 ZAF Skilled 9.03 8.11 7.55 Unskilled 31.57 25.14 23.32 Total 40.60 33.25 30.87 32.63 NAM Skilled 5.55 5.26 Unskilled 22.03 21.32 Total 27.58 26.58 27.29 CHL Skilled 5.28 5.63 6.30 6.44 6.33 Unskilled 21.12 22.63 22.48 21.46 20.60 Total 26.40 28.26 28.78 27.90 26.93 28.35 CRI Skilled 4.82 5.16 Unskilled 17.18 18.14 Total 22.00 23.30 22.05 KAZ Skilled 8.17 8.23 Unskilled 25.97 26.40 Total 34.14 34.63 39.10 MYS Skilled 2.55 7.17 6.17 4.21 4.63 Unskilled 12.58 25.25 21.59 15.44 16.90 Total 15.13 32.42 27.76 19.65 21.53 23.06 MUS Skilled 6.73 6.03 7.28 Unskilled 24.27 21.98 23.82 Total 31.00 28.01 31.10 33.48 WLD Skilled 15 13 13 12 12 Unskilled 30 27 26 25 24 Total 45 41 39 37 35 36 Table A4: Non labor value added in each $100 of exports – Botswana versus peers 1997 2001 2004 2007 2011 BWA 69.7 33.9 65.7 56.8 69.9 ZAF 38.8 44.6 42.5 44.9 NAM 47.3 45.6 49.3 CHL 51.4 47.1 46.1 48.6 49.6 CRI 42.8 42.9 43.1 KAZ 42.5 43.7 47.2 MYS 31.9 28.0 34.0 35.1 36.8 MUS 36.8 34.9 37.3 24