Relationship Between Skill, Technology and Input–Output Indicators

Abstract

This chapter uses the data from the firm survey (2010) to examine skill indicators, their implications and relationships between skill, technology and input-output indicators at the micro/firm level. We illustrate the low skill levels – due to the excessive share of unskilled workers and the implications on skills mismatch, industrial performance indicators and productivity decline across firms. These results are consistent with the micro-macro findings in Chap. 5 above, which indicate the low share of high skilled in total population and employment – measured by both educational and occupational levels – and the serious implications on skills mismatch and the macro-micro duality with respect to upskilling efforts. We find that an increase in skill level and firm size lead to improved relationships between actual and required education, between actual education, experience and wages and between required education, experience and wages. Our results concerning the positive complementary relationships between skill, technology (ICT) and upskilling (ICT training) and between computers, telecommunications and ICT training are consistent with the findings in the new growth literature. We illustrate that an increase in skill level and firm size lead to an improvement in the complementary relationships between skill, upskilling and technology (ICT) at the micro level.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Earlier findings in Chap. 5 show the serious implications of the deficient educational system and excessive use of low educated workers, and illustrate one surprising contradicting macro–micro view regarding the transfer of knowledge/external schooling effects. This surprising result from Chap. 5 motivates our research to attempt a more comprehensive analysis of skill problem and implications of unskilled workers at the micro level/across firms. Hence, the aim of this chapter is to broaden our earlier analysis in Chap. 5 by providing an indepth analysis of skill and technology indicators and the relationship between them and the implications of the prevalence of low-skilled workers at the micro level. In addition, we examine the relationships between: skill indicators (education/actual education and occupation/required education respectively and experience) and average wages; between skill, upskilling (spending on ICT training) and technology (spending on ICT); and between technology (spending on ICT) and input–output indicators across firms. We also compare the relevance of our results to the theoretical framework in Chap. 3 and the findings concerning these relationships in the new growth literature.

Prior to investigating the relationships between skill, upskilling, technology and input–output indicators across firms, it is convenient to begin with explaining the importance of the industrial sector across firms, because understanding the importance of the industrial sector from the perspective of the industrial firms can help in supporting the potential contribution of the industrial sector in enhancing economic development in Sudan. Beginning with the importance of the industrial sector for economic development in Sudan, our results from the firm survey (2010) imply that the respondent firms seem to be highly optimistic regarding the potential contribution of the industrial firms in achieving not only the traditional microeconomic aim of maximising private industrial profit but also in achieving the macroeconomic development aims, provided that the appropriate conditions for industrial development is created. For instance, the potential contribution of the industrial sector in: increasing output and income; increasing employment opportunities for present and future labour force (in response to potential population increase); satisfying domestic consumption and achieving self sufficiency by offering the basic and necessary goods for the Sudanese; achieving industrial profit; improving production relationships between workers; and enhancing local technological capability building by adaptation of imported technologies to fit with local needs. This is in addition to: contribution to economic growth through enhancing industrial linkages; reforming the structural imbalances in Sudan economy; decreasing imports; and enhancing optimal and full utilisation of natural resources and local raw materials; enhancing local capability; enhancing development of local technologies to fit with local development needs; supporting development and urbanisation of all regions in Sudan; enhancing local industrialisation using local raw materials; and enhancing economic growth by increasing industrial exports. Finally, meeting the needs and enhancing linkages with other sectors specially agriculture, are also mentioned but are of somewhat less importance (see Table 7.1 below).¹

Table 7.1 The importance of the industrial sector for economic development in Sudan (2008)

The rest of this chapter is organised as follows: Sect. 7.2 defines the variables used in our analysis and the general characteristics of firms; Sect. 7.3 presents our hypotheses and discusses differences in prevalent skill levels and requirements and the implications of low skill levels on skills mismatch, industrial performance indicators and productivity decline across firms; Sect. 7.4 examines the relationships between actual and required education, experience and wages; Sect. 7.5 shows the relationships between skill, technology (spending on ICT) and upskilling (spending on ICT training) and between technology (ICT) and input–output indicators; Sect. 7.6 concludes.

2 Data, Definition of Variables and General Characteristics of Firms

Before commencing with the empirical analysis, it will be useful to briefly explain the data used in our analysis and the general characteristics of firms.

2.1 Data and Definition of Variables

Our analysis in this chapter uses the data from the firm survey (2010), which provides us with three sets of micro variables.² The first set includes skill variables, while the second and third sets include both technology and input–output and performance related variables respectively. We define skill variables by educational attainment, occupational level (measured by the required qualifications/schooling years) and average years of experience.³ We use the total spending on machinery and equipment to define “old technology” and also we use the total spending on ICT⁴ to define “new technology”, the share of spending on ICT training as a percentage of total spending on ICT to define “upskilling”, total sales value to define “output”, total profit and total value added to define “performance”, in addition we use economic, productivity, activity and profitability indicators to define industrial performance indicators, and total employment and net worth to define “labour” and “capital” inputs, respectively.⁵

We use the first set of skill variables in Sect. 7.3 to discuss hypotheses 3.b. and 4.a. in Chap. 1 above regarding the implications of unskilled workers across firms. We use input–output and performance indicators to illustrate the decline in industrial performance and productivity indicators and ratios. Next, in Sect. 7.4, we test hypothesis 4.b. in Chap. 1 above about the relationships between actual and required education and experience and wages. In Sect. 7.5, we use the first and second sets of variables including skill, ICT and the share of spending on ICT training to test hypothesis 4.c. in Chap. 1 above regarding the relationship between skill, technology (ICT) and upskilling. Next, we use the second and third sets of technology and input–output variables to test the fifth hypothesis in Chap. 1 above about the relationship between technology (ICT) and input–output indicators.⁶

2.2 General Characteristics of Firms

Table 7.2 presents the main general characteristics of firms and economic indicators such as the share of firms in total employment, capital, profit and output (total sales value), and their differences defined by firm size and industry level. We observe that the market size or structure (defined by the share in total employment, raw materials, profit, fixed capital and value added) seems biased toward large size and chemical and food firms respectively. For instance, on average, the large size and chemical firms respectively employ 74 % and 50 % of total workers, absorb 99 % and 73 % of total raw materials, and therefore, it is not surprising that they constitute 99 % and 72 % of total profit. While small size and food industries absorb 99 % and 99 % of total capital, large size and food industries absorb 84 % and 83 % of total fixed capital in the form of machinery and equipment, hence, it is not surprising that they constitute 84 % and 84 % of total value added respectively.⁷ In addition, medium size and food industries constitute 63 % and 75 % of total output (total sales value). These differences in market size leads to several implications, as we explain below and in the next sections.

Table 7.2 Main characteristics of firms in the Sudan (2005–2008)

From Table 7.2 we observe the limited contribution of public sector and high share of private sector in the metal, food, chemical and textile industries and medium, small and large size firms respectively. We also note the high share of local ownership and also a limited share of foreign and mixed ownership, which implies the limited dependency on foreign capital and foreign workers. We find that the share of firms in local ownership decreases and so the share in foreign ownership increases with firm size and to some extent with industry level. But despite the presence of foreign capital, there is limited contribution of multinational companies; however, such contribution is diversified as the sources of foreign capital of multinational companies originates from different countries and increases to some extent with industry level and to less extent with firm size. We also observe limited changes in the general structure of firms during the period 2005–2008, which may indicate a lack of dynamism, particularly with respect to the distribution of economic indicators, i.e. total employment, capital and output/sales value across firms. The reported change since establishment in ownership, nationality of main owner and length of years in operation (age) varies across firms and generally increases with firm size and industry level; it was observed only in some of the chemical industries and large and small size firms. In addition, the geographical distribution of firms indicates that most are clustered in two main locations and only a few of the chemical, food and metal industries and large and medium size firms have branches in cities other than the main location, though the probability of clustering to some extent increases with firm size and industry, and the probability of having branches increases with firm size but to lesser extent increases with industry. Moreover, we realise the limited scope for diversification as measured by sales and employment indices across firms.⁸ The average diversification index increases to some extent with firm size but only to a lesser extent increases with industry; this implies that metal and chemical industries and large size firms have more interest in diversification, whereas food and textile industries and medium size firms have less interest in diversification and more interest in concentration and specialisation. As expected, large size firms reported more interest in diversification than medium and small size firms. Somewhat surprising and in contrast to our expectations, the findings across firms indicate that metal firms reported more interest in diversification more than chemical, food and textile firms, moreover, somewhat surprising was that small size firms indicated more interest in diversification than medium size firms.

3 Differences in Skill Level and Requirements and the Implications Across Firms

Our earlier findings in Chap. 5 indicate that the share of high skilled workers in total employment, the total number of full time equivalent researchers (FTER), R&D and ICT expenditure, patent, product and process innovations are higher within large size and chemical firms when compared to medium and small size and food, metal and textile firms. Our result with respect to R&D and chemical sector is consistent with the standard classification developed by the OECD in the mid-1980s, which distinguishes between industries in terms of R&D intensity (cf. OECD 1997). For instance, in the mid-1980s, the OECD classification distinguished between industries in terms of R&D intensity, considering pharmaceutical and ICT as high-technology, chemical and vehicle as medium-technology and food and textile as low technology (cf. OECD 1997). Our findings with respect to firm size are consistent with the literature and the Schumpeterian hypothesis, which indicates that large size/market concentration is conducive to R&D investment (cf. Braga and Willmore 1991). For instance, Kumar and Saqib (1994) suggest that the probability of undertaking R&D increases with firm size only up to a certain level, while R&D intensity increases with it linearly. However, one should also expect that these results could imply a possibility for reversed causality, mainly because R&D is a fixed cost that requires high financial capacity, which is most likely to be strong amongst large size firms.

In addition to earlier findings, we observe that skill levels and requirements (actual and required education and experience) and skills mismatch are not homogenous across firms and vary with industry and size. As we explained in Sect. 7.3, these findings can be used to test the hypotheses 3.b. and 4.a. in Chap. 1 above that, irrespective of these differences, high skill requirements and low skill levels – due to high share of unskilled workers – lead to skills mismatch and also contribute to industrial performance indicators and productivity decline across firms. In Sects. 7.4 and 7.5, we then examined hypotheses 4.b. and 4.c. in Chap. 1 above that an increase in skill levels and firm size lead to improved relationships between actual and required education and experience, between actual education, experience and wages and between skill, upskilling and technology (ICT). Finally, in Sect. 7.5, we also investigated the fifth hypothesis in Chap. 1 above concerning the relationships between technology (the use of ICT) and input–output indicators at the micro/firm level.

3.1 Differences in Skill Level and Requirements (Education and Experience) Across Firms

Prior to investigating the first hypothesis on the extended implications of low skill levels as presented above, it is convenient to begin with explaining differences in skill levels and requirements across firms because understanding why and how they vary with industry and firm size can help in investigating both the first and second hypotheses.

In Figs. 7.1, 7.2, and 7.3 below we explain differences in skill levels and requirements and low skill levels (defined by education and occupation groups) across firms (defined by size and industry).⁹ Figures 7.1 and 7.2 show the low share of high skilled – high educated and white collar – workers, differences in skill levels according to education and occupation definitions and differences across firms. For instance, Fig. 7.1 indicates that for 55 % of all respondent firms, the share of high skilled (educated) represents 1–30 % of total employed workers. For a further 20 % of all respondent firms, the share of high skilled (educated) represents 31–50 % of total employed workers, but for the remaining 25 % the share is more than 50 % of the workforce. Figure 7.2 shows, for example, that for 66 % of all respondent firms, the share of white collar (WC) represents 1–30 % of total employed workers; for 21 % of all respondent firms the WC share is 31–50 % and for 13 % the figure stands at 50 % of total employed workers. The results show that the incidence of high educated and white collar workers constituting more than half of total employment is observed only within 25 % and 13 % of all respondent firms respectively. They also indicate that the share of high skilled – measured by education – is less than one third of total workers for 55 % of all firms and the share of high skilled – white collar measured by occupational level – is less than one third of total workers for 66 % of all firms. That means that across all firms the share of high educated and white collar respectively are less than one third and two thirds; therefore, the majority of employed workers are low and medium skilled.

Fig. 7.1

Differences in the distribution of workers by educational level across firms (% share) 2008 (Source: Firm Survey (2010))

Fig. 7.2

Differences in the distribution of workers by occupational level across firms (% share) 2008 (Source: Firm Survey (2010))

Fig. 7.3

Differences in the educational requirements by occupational level across firms (% share) 2008 (Source: Firm Survey (2010))

Figures 7.3 and 7.4 show that skill requirements – average required years of schooling – vary and increase with occupational level across firms.¹⁰ For instance, Fig. 7.3 indicates that for 26 % of all respondent firms the average required years of education for white collar (WC) is 12 and above; 68 % of all respondent firms require an average of 16 years; whilst 6 % of all respondent firms put this figure at 18 and above. Moreover, Fig. 7.4 indicates that for 16 % of all respondent firms the average required years of education for white collar high (WCH) is 14 years (diploma degree), for 47 % the requirement is 16 years (university degree) and for 37 % the requirement is 17–19 years and above (postgraduate degree). The figures show that the university degree is the major preferred required qualification only within the first and second occupational groups, while for the other occupational groups either a diploma or secondary or less than secondary schooling is required.

Fig. 7.4

Average required years of schooling defined by occupation classes across firms (2008) (Source: Firm Survey (2010))

Figure 7.5 below indicates the variation in skill requirements (required years of experience), defined by educational and occupational levels. For instance, for 36 % of all respondents firms the average required years of experience for high education is 2–5 years; for 39 % the experience requirement stands at 5–10 years, for 17 % the experience requirement stands at 10–15 years and for 8 % the figure is 16 years and above. Moreover, for 19 % of all respondent firms the average required years of experience for white collar high (WCH) is 2–5 years, for 37 % the experience requirement stands at 5–10 years, for 26 % the experience requirement stands at 10–15 years and for 18 % the figure is 16 years and above. Figure 7.5 illustrates that average years of experience are increasing in educational and occupational levels respectively. In the next section, we explain the relationships between required education/actual education, occupation/required education and experience and wages across firms.

Fig. 7.5

Average years of experience defined by education and occupation classes across firms (2008) (Sources: Firm Survey (2010))

3.2 The Implications of Low Skill Levels Across Firms

In this section we examine the first hypothesis that, irrespective of the observed differences in skill levels and requirements and as we explained above, the low skill levels – due to high share of unskilled workers – lead to skills mismatch and probably contribute to industrial performance indicators and productivity decline across firms.

3.2.1 Low Skill Levels and Skills Mismatch (Differences in Required and Attained Education)

When comparing the required schooling with the actual/attained schooling, we find that differences in schooling requirements across firms have caused considerable variations between the required and actual/attained schooling for high, medium and low skilled groups. When we interpret the required schooling as the demand for skills and the actual/attained schooling as the supply of skills, we observe that the inconsistency between the required and actual/attained schooling indicates an inconsistency between the demand for and supply of skills, which can be interpreted as skills mismatch.¹¹ For instance, Fig. 7.6 below illustrates the differences between the required and actual/attained schooling across firms, defined by firm size and industry level and skill levels. We observe that the inconsistency between the demand for and supply of skills, or skills mismatch, is particularly higher/serious within both high and low skilled groups respectively and across medium, small, chemical, food and metal firms respectively. We find mismatch amongst all employment categories, especially within high, medium and low skilled labour; for instance, we observe that for all firms, on average, the intensity of mismatch for high, medium and low skill groups accounts for 40 %, 31 % and 45 % respectively. This implies that the educational attainment amongst high, medium and low skilled labour does not match the required skills/educational level for high, medium and low skilled jobs across approximately 40 %, 31 % and 45 % of total respondents firms respectively. The mismatch is highest for high, medium and low skills, probably because of both insufficient educational attainment and high educational requirements for high, medium and low skills (see Fig. 7.3 above). Moreover, the intensity of mismatch is more prevalent across small size and medium size and chemical, metal and food firms compared to large size and textile firms. For instance, for medium size firms, on average the mismatch intensity for high, medium and low skill groups accounts for 44 %, 38 % and 39 % respectively, while for small size firms the figures are 22 %, 44 % and 71 % respectively, whereas for large size firms the figures are 38 %, 24 % and 39 % respectively. Moreover, for the chemical industries, on average the mismatch intensity for high, medium and low skill groups accounts for 60 %, 20 % and 60 % respectively, while for food industries the figures are 58 %, 30 % and 43 % respectively, whereas for metal industries the figures are 36 %, 40 % and 53 % respectively, while for textile industries the figures are 17 %, 13 % and 38 % respectively. Hence, our results in this section concerning the presence of serious skills mismatch due to the excessive share of unskilled foreign workers at the micro level are consistent with our earlier findings in Chap. 5 above, which indicates the presence of serious skills mismatch at the macro level.

Fig. 7.6

Skills mismatch defined by high medium and low skill levels across firms (%) (2008) (Source: Firm Survey (2010))

3.2.2 Low Skill Levels and the Declining Trend of Labour Productivity (Output/Labour Ratio)

The low skill levels may contribute to productivity decline across firms.¹² Table 7.3 below illustrates considerable variation in the value and trend of labour productivity (total output/labour ratio) in physical term, in particular, considerable decline in labour productivity (output/labour ratio) for numerous firms over the period 2005–08.^{13 , 14 , 15}

Table 7.3 Assessment of industrial performance: labour productivity: output/labour ratio measured in physical term across firms (2005–2008)

The declining labour productivity across many firms may not be surprising since the majority of employed workers are low skilled/educated workers – see our result above – and a low skill level may lead to further decline in productivity. For instance, Table 7.3 below shows that over the periods 2005–2006, 2006–2007 and 2007–2008, the declining trend of labour productivity is reversed across 8 out of 37 (22 %) of all respondent firms and the increasing trend continues across 16 out of 37 firms (43 %); however, the increasing trend turns into a declining one across 11 out of 37 firms (30 %), or the declining trend continues across 2 out of 37 (5 %) of all respondent firms. Hence, for the majority 24 out of 37 (65 %) of all respondent firms either the declining trend turns into an increasing one or the increasing trend continues, but for the remaining 13 out of 37 (35 %), i.e. for more than one third of all firms either the increasing trend turns into a declining one or the declining trend continues. For chemical firms over the periods 2005–2006, 2006–2007 and 2007–2008, the declining trend of labour productivity is reversed across 5 out of 18 (28 %) of the chemical firms and the increasing trend continues across 9 out of 18 firms (50 %); however, the increasing trend turns into a declining one across 2 out of 18 firms (11 %), or the declining trend continues across 2 out of 18 (11 %) of the chemical firms. Thus, for the majority 14 out of 18 (68 %) of the chemical firms either the declining trend turns into an increasing one or the increasing trend continues, but for the remaining 4 out of 18 (22 %), i.e. for more than one fifth of the chemical firms either the increasing trend turns into a declining one or the declining trend continues. For food firms over the periods 2005–2006, 2006–2007 and 2007–2008, the declining trend of labour productivity is reversed across 1 out of 12 (8 %) of the respondent firms and the increasing trend continues across 6 out of 12 firms (50 %); however, the increasing trend turns into a declining one across 5 out of 12 firms (42 %). Therefore, for more than half or the majority 7 out of 12 (58 %) of the food firms either the declining trend turns into an increasing one or the increasing trend continues, but for the remaining 5 out of 12 (42 %), i.e. for more than one third and near to one half of the food firms either the increasing trend turns into a declining one or the declining trend continues. For metal firms over the periods 2005–2006, 2006–2007 and 2007–2008, the declining trend of labour productivity is reversed across 1 out of 3 (33 %) of the metal firms; however, the increasing trend turns into a declining one across 2 out of 3 (67 %) of the metal firms. Hence, for the majority 2 out of 3 (67 %), i.e. for more than two third of the metal firms the increasing trend turns into a declining one, but for the remaining 1 out of 3 (33 %) the declining trend turns into an increasing one. For textile firms over the periods 2005–2006, 2006–2007 and 2007–2008, the declining trend of labour productivity is reversed across 1 out of 4 (25 %) of the textile firms and the increasing trend continues across 1 out of 4 firms (25 %); however, the increasing trend turns into a declining one across 2 out of 4 firms (50 %). Thus, for the first half (2 out of 4 or 50 %), i.e. for one half of the textile firms either the declining trend turns into an increasing one or the increasing trend continues, while for the other half (2 out of 4 or 50 %) the increasing trend turns into a declining one For large size firms over the periods 2005–2006, 2006–2007 and 2007–2008, the declining trend of labour productivity is reversed across 4 out of 15 (27 %) of the large size firms and the increasing trend continues across 4 out of 15 firms (27 %); however, the increasing trend turns into a declining one across 6 out of 15 firms (40 %), or the declining trend continues across 1 out of 15 (7 %) of the large size firms. Thus, for the majority 8 out of 15 (53 %), i.e. for more than one half of the large size firms either the declining trend turns into an increasing one or the increasing trend continues, but for the remaining 7 out of 15 (47 %) either the increasing trend turns into a declining one or the declining trend continues. For medium size firms over the periods 2005–2006, 2006–2007 and 2007–2008, the declining trend of labour productivity is reversed across 3 out of 10 (30 %) of the medium size firms and the increasing trend continues across 5 out of 10 firms (50 %); however, the increasing trend turns into a declining one across 2 out of 10 (20 %) of the medium size firms. Thus, for the majority 8 out of 10 (80 %) of the medium size firms either the declining trend turns into an increasing one or the increasing trend continues, but for the remaining 2 out of 10 (20 %), i.e. for one fifth of the medium size firms the increasing trend turns into a declining one. For small size firms over the periods 2005–2006, 2006–2007 and 2007–2008, the declining trend of labour productivity is reversed across 1 out of 12 (8 %) of the small size firms and the increasing trend continues across 7 out of 12 firms (58 %); however, the increasing trend turns into a declining one across 3 out of 12 (25 %) of the small size firms, or the declining trend continues across 1 out of 12 (8 %) of the small size firms. Thus, for the majority 8 out of 12 (67 %) of the small size firms either the declining trend turns into an increasing one or the increasing trend continues, but for the remaining 4 out of 12 (33 %), i.e. for one third of the small size firms either the increasing trend turns into a declining one or the declining trend continues.

Therefore, our results in this section concerning the declining labour productivity are consistent with our results regarding the declining industrial performance indicators that we measure by three sets of economic-productivity, activity and profitability indicators at the micro level as we show in the next section (see Tables 7.4 and 7.5 below).

Table 7.4 Assessment of the value, trend and growth rates of industrial performance: economic, activity, labour productivity, output/labour and capital/labour ratios and other productivity indicators across firms (2005–2008)

Table 7.5 Assessment of the value, trend and growth rates of industrial performance: Activity, other productivity and profitability indicators across firms (2005–2008)

3.2.3 Low Skill Levels and the Declining Trend of Other Industrial Performance Indicators

The low skill levels may contribute to the decline of industrial performance indicators across firms. The trend of these indicators show considerable variation across firms and in most cases seem to be more sensitive to differences in firm size, industry and sector, in particular, the average performance ratio for different indicators for numerous firms show a considerable decline over the period 2005–2008. Tables 7.4 and 7.5 below illustrate an assessment of the value and trend of industrial performance indicators across firms over the period 2005–2008, which we measure by three different sets of economic and productivity indicators, activity indicators and profitability indicators. Using Al-Quraishi’s (2005) definition of industrial performance, first we measure the first set of economic indicators by three indicators including first the degree of industrialisation that is indicated by the ratio of total value added as a percentage of total output measured by total sales value, and second the capital intensity level indicators that we measure by the ratios of capital and fixed capital – measured by total spending in machinery and equipment – as percentages to total labour respectively. We define the third economic indicator by a set of productivity indicators that we measure by: labour productivity indicator measured by the ratio of total value added as a percentage to total labour; capital productivity indicator measured by the ratio of total output measured by total sales value as percentages of total capital; fixed capital productivity indicator measured by the ratio of output measured by total sales value as a percentage of fixed capital or machinery and equipment; wage productivity indicator that we measure by the total output measured by total sales value as a percentage of total wage; and raw materials productivity indicator measured by the ratio of total output measured by total sales value as a percentage to total spending on raw materials. Second, we measure the second set of activity indicators or ratios by first the fixed capital turnover ratio that we measure by the ratio of total sales value as a percentage of fixed capital, and second the capital turnover ratio that we measure by the ratio of total sales value as a percentage of total capital. Third, we measure the third set of profitability indicators by three indicators including first the rate of return on labour that we measure by profit/labour ratio, second the rate of return on capital that we measure by the ratio of profit as a percentage to capital and third profit margin indicator that we measure by the ratio of profit as a percentage to total sales value (Al-Quraishi 2005: 249–277).

Beginning with the first set of economic indicators, we find that for all firms the trend of value and growth rate of the economic indicator as measured by the degree of industrialisation as measured by the value added/sales value (output) ratio, showed a negative decreasing trend over the periods 2005–2006 and 2005–2008 but that again turned into a positive increasing trend over the periods 2006–2007 and 2007–2008. In particular, we find that the economic indicator as measured by the degree of industrialisation as measured by the value added/sales value (output) ratio varied across firms over the period 2005–2008, for instance, either the declining trend continued or the increasing trend turned into a declining trend for metal and textile industries, small size and mixed firms, while by contrast either the increasing trend continued for food industries or the declining trend turned into an increasing trend for all firms, chemical industries and medium size and large size and private firms. Moreover, as for the second economic indicator of capital intensity and productivity indicator as measured by capital/labour productivity indicator or ratio, we find that for all firms the trend of value and growth rate of capital/labour ratio showed a negative decreasing trend over the period 2005–2006 that turned into a positive increasing trend over the periods 2006–2007, 2007–2008 and 2005–2008. In particular, we find that the capital intensity and productivity indicator measured by capital/labour ratio varied across firms over the period 2005–2008, for instance, either the declining trend continued for the chemical industries or the increasing trend turned into a declining trend for textile industries and medium size firms, whereas by contrast either the increasing trend continued or the declining trend turned into an increasing trend for all firms, food and metal industries and small size and large size and private and mixed firms. Moreover, we find that for all firms the trend of value and growth rate of the second economic indicator of capital intensity and productivity indicator measured by fixed capital/labour ratio measured by machinery and equipment/labour ratio showed a negative decreasing trend over the periods 2005–2006, 2006–2007 and 2005–2008 that turned into a positive increasing trend over the period 2007–2008. In particular, we find that the capital intensity and productivity indicator measured by fixed capital/labour ratio measured by machinery and equipment/labour ratio varied across firms over the period 2005–2008, for instance, either the declining trend continued for the chemical industries or the increasing trend turned into a declining trend for the medium size firms, whereas by contrast either the increasing trend continues or the declining trend turned into an increasing trend for all firms, food, metal and textile industries, small size and large size, private and mixed firms. In addition, we find that for all firms the trend of value and growth rate of raw materials/labour ratio showed a positive increasing trend over the periods 2005–2006 and 2005–2008 that turned into a negative decreasing trend over the periods 2006–2007 and 2007–2008. In particular, we find that the raw materials/labour ratio varied across firms over the period 2005–2008, for instance, either the declining trend continues or the increasing trend turned into a declining trend for all firms, food and textile industries and medium size and mixed firms, while by contrast either the increasing trend continued for the chemical industries, large size and private firms or the declining trend turned into an increasing trend for metal industries and small size firms. Moreover, we find that for all firms the trend of value and growth rate of wages/labour ratio showed a negative decreasing trend over the periods 2005–2006 and 2005–2008 that turned into a positive increasing trend over the periods 2006–2007 and 2007–2008. In particular, we find that wages/labour ratio varied across firms over the period 2005–2008, for instance, either the declining trend continued or the increasing trend turned into a declining trend for food and textile industries and medium size firms, while either the increasing trend continued for the chemical and metal industries and small size and private firms, or the declining trend turned into an increasing trend for all firms, large size and mixed firms. Moreover, we find that for all firms the trend of value and growth rate of sales value (output)/labour ratio showed a negative decreasing trend over all the periods 2005–2006, 2006–2007, 2007–2008 and 2005–2008. In particular, we find that sales value (output)/labour ratio varied across firms over the period 2005–2008, for instance, either the declining trend continued for all firms, food industries, medium size, and mixed firms or the increasing trend turned into a declining trend for private firms, while by contrast either the increasing trend continued or the declining trend turned into an increasing trend for the chemical, metal, textile industries and small size and large size firms. Moreover, we find that for all firms the trend of value and growth rate of value added/labour ratio showed a negative decreasing trend over the periods 2005–2006 and 2005–2008 that turned into a positive increasing trend over the periods 2006–2007 and 2007–2008. In particular, we find that the value added/labour ratio vary across firms over the period 2005–2008, for instance, either the declining trend continued or the increasing trend turned into a declining trend for chemical and food industries, medium and large size and mixed firms, while by contrast either the increasing trend continued for metal industries and small size firms or the declining trend turned into an increasing trend for all firms, textile and private firms (see Table 7.4 above). Moreover, we find that for all firms the trend of value and growth rate of other productivity indicators as measured by the wage productivity ratio as measured by sales/wage ratio showed a negative decreasing trend over the period 2005–2006 that turned into a positive increasing trend over all the periods 2006–2007, 2007–2008 and 2005–2008. In particular, we find that the other productivity indicators as measured by the wage productivity ratio as measured by sales/wage ratio varied across firms over the period 2005–2008, for instance, either the declining trend continued or the increasing trend turned into a declining trend for food industries and small size and mixed firms, while by contrast either the increasing trend continued for metal industries or the declining trend turned into an increasing trend for all firms, chemical and textile industries and medium size and large size and private firms. Moreover, we find that for all firms the trend of value and growth rate of other productivity indicators as measured by the raw materials productivity as measured by the sales/raw materials ratio showed a positive increasing trend over the period 2005–2006, that turned into a negative decreasing trend over the period 2006–2007 but that again turned into a positive increasing trend over the periods 2007–2008 and 2005–2008. In particular, we find that the other productivity indicators as measured by the raw materials productivity as measured by the sales/raw materials ratio varied across firms over the period 2005–2008, for instance, either the declining trend continued or the increasing trend turned into a declining trend for food, metal and textile industries and small size and large size and mixed firms, while by contrast either the increasing trend continued or the declining trend turned into an increasing trend for all firms, chemical industries and medium size and private firms. Moreover, we find that for all firms the trend of value and growth rate of the second set of indicators (the activity indicators as measured by fixed capital turnover ratio as defined by the sales/fixed capital (machinery and equipment) ratio) showed a positive increasing trend over the periods 2005–2008 and 2005–2006, that turned into a negative decreasing trend over the period 2006–2007 but that again turned into a positive increasing trend over the period 2007–2008. In particular, we find that the activity and other productivity indicators as measured by the fixed capital turnover ratio as measured by the sales/fixed capital ratio as measured by machinery and equipment varied across firms over the period 2005–2008, for instance, either the declining trend continued or the increasing trend turned into a declining trend for food industries and small size firms, while by contrast either the increasing trend continued for mixed firms or the declining trend turned into an increasing trend for all firms, chemical, metal and textile industries and medium size and large size and private firms. Moreover, we find that for all firms the trend of value and growth rate of activity and other productivity indicators, defined by capital turnover ratio, defined by sales/capital ratio showed a positive increasing trend over the period 2005–2006, that turned into a negative decreasing trend over the periods 2006–2007, 2007–2008 and 2005–2008. Particularly, we find that the activity and other productivity indicators, defined by capital turnover ratio, defined by sales/capital ratio vary across firms over the period 2005–2008; for instance, either the declining trend continued for medium size firms or the increasing trend turned into a declining trend for all firms, food industries and small size and private firms, while by contrast either the increasing trend continued for textile industries or the declining trend turned into an increasing trend for chemical and metal industries and large size and mixed firms (see Table 7.5 below).

As for the third set of profitability indicators from Table 7.5, we find that for all firms the trend of value and growth rate of profitability that we measure by the rate of return on labour or profit/labour ratio showed a positive increasing trend over the periods 2005–2006, 2006–2007 and 2005–2008 that turned into a negative declining trend over the period 2007–2008. In particular, we find that profit/labour ratio varied across firms over the period 2005–2008, for instance, either the declining trend continued for mixed firms or the increasing trend turned into a declining trend for all firms, chemical, food and textile industries, large size and private firms, while by contrast either the increasing trend continues or the declining trend turned into an increasing trend for metal industries, small and medium size firms. In addition, we find that for all firms the trend of value and growth rate of profitability as measured by the rate of return on capital as measured by profit/capital ratio showed a positive increasing trend over the periods 2005–2006 and 2006–2007 that turned into a negative decreasing trend over the periods 2007–2008 and 2005–2008. In particular, we find that profitability as measured by the rate of return on capital measured by profit/capital ratio varied across firms over the period 2005–2008, for instance, either the declining trend continued for medium size firms or the increasing trend turned into a declining trend for all firms, food and textile industries and private firms, while by contrast either the increasing trend continues or the declining trend turned into an increasing trend for chemical and metal industries and small size and large size and mixed firms. Moreover, we find that for all firms the trend of value and growth rate of profitability measured by profit margin that we measure by profit/sales ratio showed a negative decreasing trend over all the periods: 2005–2006, 2007–2008 and 2005–2008. In particular, we find that profitability as measured by profit margin as measured by profit/sales ratio varied across firms over the period 2005–2008, for instance, either the declining trend continued for chemical and food industries, large size, medium size and private firms or the increasing trend turned into a declining trend for all firms, metal and textile industries and small size, while by contrast the declining trend turned into an increasing trend only for mixed firms.

We find that in most cases the trend of these indicators seem to be more sensitive to differences in firm size, industry and sector. In particular, the industrial performance indicators that seem to be more sensitive to differences in firm size, industry and sector include the economic indicator as measured by the degree of industrialisation that we measure by the ratio of total value added as a percentage of total output measured by total sales value. Moreover, other industrial performance indicators that seem to be more sensitive to differences in firm size, industry and sector include three productivity indicators: capital productivity indicator (total output (measured by total sales value)/total capital); the fixed capital productivity indicator (total output (measured by total sales value)/fixed capital (machinery and equipment)); and the wage productivity indicator (total output (measured by total sales value)/total wage). In addition to the activity indicators or ratios measured by fixed capital turnover ratio, measured by the ratio of total sales value as a percentage of fixed capital, and capital turnover ratio measured by the ratio of total sales value as a percentage of total capital, in addition to the profitability indicator measured by the rate of return on capital measured by the ratio of profit as a percentage to capital. We find that the industrial performance indicators that seem to be to some extent sensitive to differences in firm size but less sensitive to industry and sector include the economic or capital intensity level indicator measured by both the ratio of total capital as a percentage to total labour and the ratio of fixed capital or total spending in machinery and equipment as a percentage to total labour. Moreover, we find that the industrial performance indicator that seems to be sensitive to only differences in industry is the raw materials productivity indicator measured by the ratio of total output measured by total sales value as a percentage to total spending on raw material. We find that the industrial performance indictors that seem to be insensitive to differences in firm size, industry and sector include the labour productivity indicator measured by the ratio of total value added as a percentage to the total labour and profitability indicators that we define by profit/labour ratio and profit margin indicator measured by the ratio of profit as a percentage to total sales value. These results imply that in most cases an increase in skill level – share of high skill in total employment – firm size and industry most probably leads to an improvement in most of industrial performance indicators (see Tables 7.4 and 7.5 below).

3.2.4 Low Skill Level and Declining Performance of Manufacturing Industrial Firms

The findings from the firm survey (2010) and Table 7.6 below support our argument that the low skill levels may contribute to declining industrial performance indicators: economic, activity, profitability and labour productivity across firms as we explained above. Table 7.6 below shows that the low skill level is indicated by firms among the important problems that are hindering industrial performance and contribution towards economic development in Sudan.¹⁶ For instance, we find that from the perspective of all respondent firms the most important problems are: inadequate finance and inappropriate conditions for industrial development, spread of routine and bureaucracy and slow procedures related to the industrial needs, interruption and inadequate availability and high costs of electricity and water, lack of raw materials, inadequate infrastructure, weak maintenance capability and lack of spare parts, inadequate skills and lack of trained labour force, weak industrial awareness, weak and narrow marketing opportunities, weak and inadequate economic visibility studies, inadequate management and organisational facilities and inadequate transportation equipment respectively (see Table 7.6 below).¹⁷ Moreover, from the firms’ perspective other extremely important factors hindering contribution of the industrial sector in economic development in Sudan include the lack of support from Ministry of Industry and the government, and high production costs caused by the imposition of high taxes, fees, levies and customs for clearance of imported raw materials, machines, machinery and equipment imposed on the industrial firms in Sudan.¹⁸ For chemical industries the most important problems are: interruption and inadequate availability and high costs of electricity and water, spread of routine and bureaucracy and slow procedures related to industrial needs, lack of raw materials, inadequate finance and inappropriate conditions for industrial development, inadequate infrastructure, weak industrial awareness, inadequate skill and lack of trained labour force, weak maintenance capability and lack of spare parts, weak and narrow marketing opportunities and inadequate management and organisational facilities respectively.¹⁹ For food industries the most important problems are: spread of routine and bureaucracy and slow procedures related to industrial needs, interruption and inadequate availability and high costs of electricity and water, inadequate finance and inappropriate conditions for industrial development, weak maintenance capability and lack of spare parts, inadequate infrastructure, inadequate skills and lack of trained labour force, weak industrial awareness, weak and narrow marketing opportunities and lack of raw materials respectively.²⁰ For metal industries the most important problems are: inadequate skills and lack of trained labour force, inadequate finance and inappropriate conditions for industrial development, lack of raw materials, inadequate infrastructure, weak maintenance capability and lack of spare parts, weak industrial awareness, inadequate management and organisational facilities and spread of routine and bureaucracy and slow procedures related to industrial needs respectively.²¹ For textile industries the most important problems are: inadequate finance and inappropriate conditions for industrial development, lack of raw materials, interruption and inadequate availability and high costs of electricity and water, weak and narrow marketing opportunities, inadequate skills and lack of trained labour force, spread of routine and bureaucracy and slow procedures related to industrial needs, weak maintenance capability and lack of spare parts and weak industrial awareness respectively.²² For large size firms the most important problems are: inadequate skills and lack of trained labour force, weak maintenance capability and lack of spare parts, inadequate finance and inappropriate conditions for industrial development, lack of raw materials, interruption and inadequate availability and high costs of electricity and water, spread of routine and bureaucracy and slow procedures related to industrial needs, weak industrial awareness, inadequate management and organisational facilities, inadequate infrastructure and weak and narrow marketing opportunities respectively.²³ For medium size firms the most important problems are: inadequate finance and inappropriate conditions for industrial development, spread of routine and bureaucracy and slow procedures related to industrial needs, interruption and inadequate availability and high costs of electricity and water, inadequate infrastructure, lack of raw materials, narrow marketing opportunities, weak industrial awareness, inadequate skills and lack of trained labour force and weak maintenance capability and lack of spare parts respectively.²⁴ For small size firms the most important problems are: inadequate finance and inappropriate conditions for industrial development, spread of routine and bureaucracy and slow procedures related to industrial needs, interruption and inadequate availability and high costs of electricity and water, inadequate infrastructure, lack of raw materials, weak maintenance capability and lack of spare parts, weak industrial awareness, weak and narrow marketing opportunities, weak and inadequate economic visibility studies and inadequate skills and lack of trained labour force respectively.²⁵

Table 7.6 The factors constraining improvement of industrial firms performance and economic development in Sudan (2008)

Hence, our results from Table 7.6 and the firm survey (2010) are consistent with the findings in developing countries and the Sudanese literature that indicate several problems of industrialisation in Sudan (El-Sayed 1998; Abdel-Salam 2006) similar to those in the typically developing countries (Ismail 1994). Different from the studies in the Sudanese literature (El-Sayed 1998; Abdel-Salam 2006) which provide a somewhat general overview concerning the problems of industrialisation in Sudan, an interesting and novel element in our analysis is that our findings are based on recent micro primary data based on the firm survey (2010) and the follow-up interviews with firm managers, and we present a new and more elaborate interpretation of the main problems of industrialisation in Sudan from the perspective of the different industrial firms considering the opinions of a more diversified sample of industrial firms, defined by industry and size as we explained in Table 7.6 below.²⁶

Therefore, our findings in this section verify the first hypothesis that high skill requirements and low skill levels – due to high share of unskilled workers – lead to skills mismatch and probably contribute to industrial performance and productivity decline across firms. In the next sections we examine the second and third hypotheses.

4 Upskilling, Improving Industrial Performance and Relationships Between Required Education (Occupation), Attained/Actual Education, Experience and Average Wages

Before examining the second and third hypotheses, it is useful to briefly show the importance of upskilling, because explaining this can be used to prevent the decline in labour productivity and industrial performance indicators and to enhance the complementary relationships between skill, technology and upskilling across firms.

4.1 Upskilling and Improving Performance of Manufacturing Industrial Firms

The findings from the firm survey (2010) presented in Tables 7.3, 7.4, 7.5, and 7.6 above, support our argument that low skill levels may contribute to the declining of labour productivity and other industrial performance indicators including economic, productivity, activity and profitability indicators across firms as we explained above. These findings imply that improving skill level is an important factor for facilitating improvement of labour productivity and other industrial performance indicators. Table 7.7 below indicates upskilling or improving skill level and adequate availability of skill and trained labour force to be amongst the important factors facilitating improvement of industrial firms performance and contributing towards economic development in Sudan.²⁷ For instance, we find that from the perspective of all respondent firms the most important factors facilitating improvement are: improving and enhancing adequate availability of finance and appropriate conditions for industrial development, improving and enhancing adequate availability of raw materials, improving and enhancing adequate availability of industrial awareness, improving and enhancing adequate availability of maintenance capability and spare parts and avoidance of routine and bureaucracy and speeding up of the procedures related to industrial needs. In addition to improving and enhancing adequate availability of infrastructure, improving and enhancing adequate availability of electricity and water with cheap and subsidised price, improving and enhancing adequate availability of skill and trained labour force, improving and enhancing adequate availability of marketing opportunities, improving and enhancing adequate availability of management and organisational facilities, improving and enhancing adequate availability of transportation equipment and improving and enhancing adequate availability of economic visibility studies (see Table 7.7 below).²⁸ Furthermore, from the firms’ perspective other extremely important enhancing factors for the development of the performance of the industrial firms include lowering or cancellation of fees, taxes and levies imposed by the government, establishment of databases, reduction of government intervention in the industrial activities and improving and accelerating the procedures for customs clearance of imported raw materials and speeding up of the process of final export of industrial products. From the perspective of chemical firms the most important factors are: improving and enhancing adequate availability of finance and appropriate conditions for industrial development, improving and enhancing adequate availability of raw materials, improving and enhancing adequate availability of industrial awareness, improving and enhancing adequate availability of infrastructure, improving and enhancing adequate availability of maintenance capability and spare parts and improving and enhancing adequate availability of marketing opportunities. In addition to improving and enhancing adequate availability of management and organisational facilities, improving and enhancing adequate availability of skills and trained labour force, avoidance of routine and bureaucracy and speeding up of the procedures related to industrial needs, improving and enhancing adequate availability of electricity and water with cheap and subsidised prices.²⁹ From the perspective of food firms the most important factors are: improving and enhancing adequate availability of finance and appropriate conditions for industrial development, avoidance of routine and bureaucracy and speeding up of the procedures related to industrial needs and improving and enhancing adequate availability of industrial awareness. In addition: improving and enhancing adequate availability of electricity and water with cheap and subsidised prices, improving and enhancing adequate availability of maintenance capability and spare parts, improving and enhancing adequate availability of raw materials, improving and enhancing adequate availability of infrastructure, improving and enhancing adequate availability of marketing opportunities and improving and enhancing adequate availability of skill and trained labour force respectively.³⁰ From the perspective of metal firms the most important factors are: improving and enhancing adequate availability of skills and trained labour force, improving and enhancing adequate availability of raw materials, improving and enhancing adequate availability of maintenance capability and spare parts, improving and enhancing adequate availability of industrial awareness, improving and enhancing adequate availability of finance and appropriate conditions for industrial development, improving and enhancing adequate availability of infrastructure and avoidance of routine and bureaucracy and speeding up of the procedures related to industrial needs. In addition: improving and enhancing adequate availability of electricity and water with cheap and subsidised prices, improving and enhancing adequate availability of management and organisational facilities and improving and enhancing adequate availability of transportation equipment’s respectively.³¹ From the perspective of textile firms the most important factors are: improving and enhancing adequate availability of finance and appropriate conditions for industrial development, improving and enhancing adequate availability of raw materials, improving and enhancing adequate availability of electricity and water with cheap and subsidised prices and improving and enhancing adequate availability of skills and trained labour force. In addition to improving and enhancing adequate availability of industrial awareness, improving and enhancing adequate availability of maintenance capability and spares part, improving and enhancing adequate availability of marketing opportunities, improving and enhancing adequate availability of management and organisational facilities, avoidance of routine and bureaucracy and speeding up of the procedures related to industrial needs and improving and enhancing adequate availability of infrastructure respectively.³² From the perspective of large size firms the most important factors are: improving and enhancing adequate availability of skills and trained labour force, improving and enhancing adequate availability of management and organisational facilities, improving and enhancing adequate availability of raw materials, improving and enhancing adequate availability of finance and appropriate conditions for industrial development, improving and enhancing adequate availability of maintenance capability and spare parts, improving and enhancing adequate availability of industrial awareness and improving and enhancing adequate availability of infrastructure. In addition: avoidance of routine and bureaucracy and speeding up of the procedures related to industrial needs, improving and enhancing adequate availability of electricity and water with cheap and subsidised prices, improving and enhancing adequate availability of marketing opportunities, improving and enhancing adequate availability of transportation equipment and improving and enhancing adequate availability of economic visibility studies respectively.³³ From the perspective of medium size firms the most important factors are: improving and enhancing adequate availability of industrial awareness, avoidance of routine and bureaucracy and speeding up of the procedures related to industrial needs, improving and enhancing adequate availability of finance and appropriate conditions for industrial development, improving and enhancing adequate availability of raw materials and improving and enhancing adequate availability of skills and trained labour force. In addition to improving and enhancing adequate availability of infrastructure, improving and enhancing adequate availability of maintenance capability and spare parts, improving and enhancing adequate availability of electricity and water with cheap and subsidised prices, improving and enhancing adequate availability of marketing opportunities and improving and enhancing adequate availability of management and organisational facilities and availability of transportation equipment.³⁴ From the perspective of small size firms the most important factors are: improving and enhancing adequate availability of finance and appropriate conditions for industrial development, improving and enhancing adequate availability of electricity and water with cheap and subsidised prices, improving and enhancing adequate availability of maintenance capability and spare parts, improving and enhancing adequate availability of marketing opportunities, improving and enhancing adequate availability of raw materials and improving and enhancing adequate availability of infrastructure. In addition: avoidance of routine and bureaucracy and speeding up of the procedures related to industrial needs, improving and enhancing adequate availability of industrial awareness, improving and enhancing adequate availability of management and organisational facilities, improving and enhancing adequate availability of economic visibility studies, improving and enhancing adequate availability of transportation equipment and improving and enhancing adequate availability of skills and trained labour force respectively.³⁵

Table 7.7 The factors facilitating improvement of industrial firms performance and economic development in Sudan (2008)

4.2 Relationships Between the Required Education (Occupation), Attained/Actual Education, Experience and Average Wages

Based on the above findings, in this section we examine a part of the second hypothesis that an increase in skill levels and firm size leads to improved relationships between actual and required education, and between actual education, required education, experience and wages across firms.

We begin with the relationship between occupation and education. Using the above definitions of occupation and education/actual and required education respectively, we translate the required qualifications for each of the occupation groups into average years of schooling and use the OLS regression, assuming that the required schooling in each occupation class is dependent on the actual/attained education. Our findings in Table 7.8 and Fig. 7.7 below illustrate that improvement in occupational status (measured by the required education) is positively and significantly correlated with education (measured by actual/attained education) across all firms. In addition, Table 7.8 illustrates that an increase in firm size and industry level leads to improved relationships between required and actual education. For instance, the required education appears to be more sensitive to and increasing in attained/actual education within both large size and chemical and food firms, and more sensitive within all firms. This result is plausible since the skill level – share of high skilled measured by educational attainment – is higher within large size and chemical and food firms compared to metal and textile, medium and small size firms (see Fig. 7.1 above). This is also probably because large size firms are more prevalent in the chemical and food industries (see Table 7.2 above) and may have more consistent recruitment strategies. These results confirm our earlier observations that skill levels and requirements (actual and required education) are non-homogenous across firms and are determined by size and industry.

Table 7.8 Required and actual/attained education and experience across firms (2008)

Fig. 7.7

The distribution of occupation classes according to the translated average years of schooling across firms (2008) (Source: Firm Survey (2010))

Concerning the relationship between education, occupation and experience, Table 7.8 above shows that average years of experience are positively correlated and increasing in education (i.e. attained/actual education) and occupation (i.e. required education) respectively. This result is consistent with Fig. 7.5 above, and probably implies that skill indicators – education and experience – are complementing rather than substituting each other.

Table 7.9 below illustrates a considerable variation in the distribution of average wages amongst high, medium and low skill – educational and occupational – levels across firms. When using the occupational rather than the educational definition the distribution of wages shows less fluctuation across firms. Therefore, the effect of occupation/required education on the distribution of average wages across firms seems to be less sensitive to differences in firm size and industry. In contrast, when using the educational definition, we observe that the effect of the actual/attained education on the distribution of average wages across firms seems to be more sensitive to differences in firm size and industry. Our interpretation of the observed differences across firms implies the presence of a significant wage differential, the lack of a coherent, homogeneous, unified and sound wage policy and the lack of systematic and consistent recruitment strategies across firms that most probably related to the lack of systematic regulations to organise the labour market in Sudan

Table 7.9 Differences in the distribution of average wages defined by firm size and industry level and sector (2008)

The above results are consistent with the OLS regression reported in Table 7.10 below, which indicates that the average wages are positively and significantly correlated with and more sensitive to attained/actual education. For instance, Table 7.10 below illustrates that the average wages are increasing in actual/attained education, experience and its square (cf. Mincer 1974) and therefore, is biased against less educated and experienced workers. These findings support our results from the firm survey, which indicate that wages are increasing in education and biased against low educated workers because the ratio of high skilled to low skilled wages, which can be interpreted as wages/skills premium, exceeds one (see Fig. 7.8 below).³⁶ These results are consistent with the findings in the new growth literature, particularly skilled biased technical change theorems (cf. Aghion and Howitt 1992, 1998; Acemoglu 1998; Autor et al. 1998). Our results from Table 7.11, which indicate that required education also has significant impact on wages, are plausible and consistent with our expectation in view of the results of the overeducation literature (Hartog 2000; Muysken and ter Weel 1998; Muysken and Ruholl 2001; Muysken et al. 2002a, b; Muysken et al. 2003). We find that the positive correlations between actual education, experience, its square and wages seem more sensitive to firm size and industry level and are particularly significant for large and medium size firms and chemical and food industries, which may not be surprising since these firms have sufficient scope for a coherent wage policy (Nour 2005; Muysken and Nour 2006). This is also probably because large size and medium size firms and chemical and food industries may have more consistent recruitment strategies and high skill levels – share of high skilled workers in total employment (see Fig. 7.1 above and Fig. 7.9 below). These results imply that an increase in skill level/actual education and firm size and industry leads to an improved relationship between actual education, experience and wages.

Table 7.10 Correlation between wages (log) actual and required education and experience (2008) (education definition)

Fig. 7.8

Differences in wage/skill premium (the ratio of high skilled wages/low skilled wages) defined by education levels across firms (2008) (Source: Firm Survey (2010))

Table 7.11 Correlation between wages (log) actual and required education and experience (2008) (occupation definition)

Fig. 7.9

Differences in skill level (share of high skilled) defined by education and occupation classes across firms (2008) (Sources: Firm Survey (2010))

One interesting observation from the firm survey data (2010) and the follow-up interviews with firms managers and the results presented in Tables 7.9, 7.10, and 7.11 is that in most cases, the OLS regression results seem to be more significant when using the education definition as compared to occupation definition. This finding seems to be consistent with the observations from Table 7.9 above but seem to be opposite to the observations from the follow-up interviews and the wide belief among firm managers, which probably implies that across the majority of the respondent firms, the structure of wage policy is most likely structured to be more consistent based on occupation definition compared to education definition. This also implies that from the firms’ perspective the decision of determining wages levels for workers is most probably determined by the nature of jobs that the workers will do in the firms rather than the years of schooling the workers have already obtained. This also most probably implies the positive but weak return and incentives for additional years of schooling to compensate the costs of additional years of schooling. Another interesting observation is that for all groups of firms when using both education and occupation definitions the OLS regression reported in Tables 7.10 and 7.11 below indicate that the correlations between wages levels and years of education variable are more significant as compared to the correlations between wages levels and average years of experience variable. This result implies that the rate of return to the worker’s average years of education is higher and more significant than the average years of experience. This finding is also opposite to the observations from the follow-up interviews and the wide belief among some firm managers which probably implies that across some firms and from some firms’ perspective, the decisions of hiring and offering wages are largely determined by experience in the practice of work which is measured by a worker’s average years of experience, which is more important than average years of education for some firms that prefer to hire more experienced than educated workers for specific fields.

Therefore, our findings in this section corroborate the first part of the second hypothesis that an increase in skill levels and firm size leads to an improved relationship between actual and required education and experience and between actual education, required education, experience, its square and wages. In the next section we proceed to examine the second part of the second hypothesis that an increase in skill levels and firm size lead to improved relationships between skill, upskilling and technology (ICT). Finally, we test our third hypothesis on the relationship between technology (ICT) and input–output indicators at the micro/firm level.

5 Skill, Upskilling (ICT Training), Technology (ICT) and Input–Output Indicators

Based on the above results, in this section we examine the other part of the second hypothesis that an increase in skill levels and firm size leads to improved relationships between skill, upskilling and new technology (ICT) across firms. Before examining this hypothesis, it is useful to briefly show the variations in the use of new technology (spending on ICT) and upskilling (spending on ICT training) across firms, because the observed differences in skill and spending on ICT and ICT training can be used to interpret the complementary relationships between skill, technology and upskilling across firms.

5.1 Skill and the Share of Spending on Technology (ICT) and Upskilling (ICT Training)

Table 7.12 shows considerable variations in the share and trend of total spending on ICT including computers, telecommunications, training, Internet, maintenance and other items, defined by firm size and industry. The share of telecommunications exhibits a continuous increasing trend for all firms, while that of training shows an opposite declining trend. Table 7.2 above shows that, on average, the share of large size and food and chemical firms represents about 48 %, 53 % and 23 % of total spending on ICT respectively and about 75 %, 73 % and 2 % of total spending on ICT training respectively. However, despite the big share of spending on ICT and ICT training, large size and food firms experienced declining trends of ICT and ICT training (cf. Figs. 7.10 and 7.11). The decline in total ICT spending can be interpreted as being due to a lack of plan for critical expansion in ICT sector or probably due to a general cutback in total spending across food and large size firms. The declining expenses on both ICT training and computers follow the general decline in total ICT spending, which can also be attributed to a lack of plan for critical expansion and a possible change in the strategy of firms that, having already established a sound basis for these components, probably need to shift priority to increase spending on both telecommunications and maintenance.

Table 7.12 Spending on ICT defined by firm size and industry (2005–2008) (% share in total spending)

Fig. 7.10

The share and trend of total spending on ICT across firms (2005–2008) (Source: Firm Survey (2010))

Fig. 7.11

The share and trend of spending on ICT training across firms (2005–2008) (Source: Firm Survey (2010))

We now proceed to examine the second part of our second hypothesis that an increase in skill levels and firm size leads to improved complementary relationships between skill, technology (ICT) and upskilling (ICT training) (see Table 7.13 below). For instance, we observe the complementary relationship between the share of high education and the share of expenditure on ICT, which can be seen and understood as complementarity between skill and technology (cf. Goldin and Katz 1998; Acemoglu 1998). We find a complementary relationship between the share of high education and the share of expenditure on ICT training, which can be interpreted as complementarity between skill and upskilling. Tables 7.13 and 7.14 show complementary relationships between the share of expenditure on ICT and ICT training, and between spending on computers, telecommunications, Internet and training, which can be read as complementarity between technology and upskilling (cf. Colecchia and Papaconstantinou 1996; Bresnahan et al. 1999). Our findings, that these complementarities are particularly significant for large size firms, are plausible since these firms have more spending on ICT and ICT training (see Table 7.2 above) and have high skill levels – share of high skilled workers in total employment (see Fig. 7.1 above). These results are consistent with the second part of our second hypothesis that an increase in skill levels and firm size lead to improved complementary relationships between skill, upskilling and technology (ICT) (cf. Acemoglu 1998). The results also imply the importance of a good education/high skill level for the enhancement of skill, technology and upskilling complementarity at the micro level. That also seems consistent with the endogenous growth framework and stylised facts concerning the relationships between human capital, technical progress and upskilling (see our theoretical framework in Chap. 3 above).

Table 7.13 The relationship between ICT, skill and upskilling across firms (2008) (2005–2008)

Table 7.14 The relationship between computers, training, internet and telecommunications expenditures across firms (2005–2008)

5.2 The Use of Technology, ICT, Skill and the Demand for Skilled Workers Across Firms

One implication of the above complementary relationship between skill and technology is that the demand for skilled workers has changed in response to the increasing uses of ICT and other technologies. For instance, during the period 2006–2008 the uses of ICT (85 %) increased faster than that of other technologies (70 %); similarly, the corresponding rise in the demand for skilled workers needed for ICT (65 %) was more than that for other technologies (61 %) across all respondents firms (see Fig. 7.12 below). This trend may reflect the fact that the real demand for skilled workers needed for ICT is more than that of other technologies across firms, which may not be surprising given the recent rapid increasing trend of IT diffusion despite the recent history of IT diffusion in Sudan. For instance, according to the World Development WDI database (2005), before 2000 the number of users of both mobile phone and Internet per 1,000 population were zero and up until the year 2000 both were only one; in recent years, Sudan has shown a growing telecommunication network and Internet services but still the highest price/most expensive Internet services as compared to other African and Arab and developing countries.

Fig. 7.12

The increasing use of technology, ICT and the demand for high skilled workers across firms, 2006–2008 (Source: Firm survey (2010))

According to the respondent firms, the increasing use of new technologies caused an increase in both the demand for more skilled workers and the required skill levels of the respective workers involved with them. Table 7.15 indicates that the increasing use of new technologies has important effects on increasing the general skill levels and the demand for skilled workers amongst 88 % and 83 % of the respondent firms respectively.³⁷ However, it has relatively less important effects on increasing skill levels mainly for unskilled workers, and decreasing and substituting the demand for unskilled workers due to reduction and elimination/substitution of some unskilled jobs. This implies change in the structure of employment/demand for workers in response to the increasing uses of new technologies and is also evidence of skilled-biased technical change theorem.³⁸

Table 7.15 The effects of new technologies on skill level and the demand for workers in the Sudan, 2008

Moreover, from the firm survey we find that the increasing use of new technologies has not only raised the demand for high skilled workers in the past years, but also encouraged firms to predict a future/long run increase in the demand for high skilled workers. For instance, for 68 % of the respondent firms the interpretations of the predicted long run increase in the demand for skilled workers are related to planned/expected expansion of production, product diversification, implementation of new process, output technologies, purchases of new machines and equipment and increasing R&D activities.³⁹ This result seems consistent with the assumption made by Aghion and Howitt (1992) that an expectation of more research in the next period must correspond to an expectation of higher demand for skilled labour in research in the next period.

5.3 The Share of Spending on ICT and Input–Output Indicators

Finally, in this section we investigate the third hypothesis on the positive relationships between new technology (total expenditures on ICT) and input–output indicators across firms and over time. For instance, when investigating the relationship between ICT and input variables, we find from Table 7.16 that the total spending on ICT is positively correlated and more sensitive to labour (firm size), and industry level throughout the period 2005–2008 and also became sensitive to capital (net worth), notably, throughout the period 2007–2008. Both the total spending on ICT and ICT training (upskilling) are positively and significantly correlated and more sensitive to labour (firm size), and capital (net worth) throughout the period 2005–2008. The relationship between ICT and labour (firm size) is particularly more significant for the large size, chemical and textile firms. The different results across chemical and textile or large size firms is plausible and can be attributed to differences in the skill levels – share of high skilled workers in total employment (see Fig. 7.1 above). This is also because large size firms are more prevalent in the textile and chemical industries, they have high share in total ICT spending, employment, fixed capital, value added and profit (see Table 7.2 above) and probably have more consistent entrepreneurial/organisational strategies.

Table 7.16 Total spending on ICT, labour and capital across firms (2005–2008)

We examine the relationship between the use of new technology as measured by total spending on ICT, profit and output. Table 7.17 illustrates plausible positive though not significant correlations between the use of new technology (as measured by total spending on ICT) and capital, labour, total output (as measured by total sales value), output diversification (as measured by sales diversification), and productivity (as measured by total sales value/labour ratio) over the period 2007–2008. Moreover, Table 7.17 shows positive significant correlations between the use of new technology as measured by total spending on ICT and total profit and value added over the period 2007–2008.⁴⁰ In addition to positive significant correlations between old technology measured by total spending on machinery and equipment and total output measured by total sales value, profit and value added, between value added and old technology measured by total spending on machinery and equipment, spending on raw materials and capital. For old technology measured by total spending on machinery and equipment, the correlation coefficients are more significant than traditional inputs (labour-capital) over the period 2005–2008. These results prove our third hypothesis regarding the positive correlation between ICT and input–output indicators at the micro/firm level. However, our results should be interpreted carefully as they probably have two-ways causality and may leave open the possibility for reversed causality. Mainly because more profit and output would imply more financial capacity that permits more spending on ICT, on the other hand, more spending on ICT implies higher costs and lower profit (see Table 7.17 below).

Table 7.17 The correlation between, firm performance, output and profit and labour, capital, total spending on ICT, machinery and equipment and raw materials across firms across firms, (2005–2008)

Our findings concerning the significant positive correlations between ICT and profit and value added and the insignificant correlation between ICT and output imply an inconclusive effect at the micro level. These results agree with our observations at the aggregate level, which imply that the growing expenditures on ICT in Sudan raises the shares of the population using the Internet, enhances e-business, e-education and e-government. However, despite the growing ICT expenditures, their effects are inconclusive at the aggregate level, probably due to low spending on ICT, high poverty and illiteracy rates, low skill levels and inadequate investment in education.⁴¹ The macro observations are consistent with the recent literature indicating the growing but limited effects of ICT diffusion in the developing countries due to a lack of sufficient investment in the complementary infrastructure such as education, skills and technical skills (cf. Pohjola 2002; Kenny 2002). Therefore, these results prove the third hypothesis in Chap. 1 above about the inconclusive effect of ICT at the micro level.

6 Conclusions

In this chapter we use the data from the firm survey (2010) to examine skill indicators, their implications and relationships with average wages, and with upskilling (ICT training) and technology (ICT), ICT and input–output indicators at the micro/firm level.

Our findings in Sect. 7.3 illustrate the low skill levels – due to the excessive share of unskilled workers (Figs. 7.1 and 7.2) – and the implications on skills mismatch (Fig. 7.6), industrial performance indicators and productivity decline across firms (Tables 7.3, 7.4, 7.5, and 7.6). These results are consistent with the micro–macro findings in Chap. 5 above, which indicate the low share of high skilled in total population and employment – measured by both educational and occupational levels – and the serious implications on skills mismatch and the macro–micro duality with respect to upskilling efforts. These findings together with those in Chap. 5 above verify hypotheses 3.b and 4.a in Chap. 1 above regarding the implications of the interaction between the deficient educational system and high use of unskilled workers. These findings then confirm our first hypothesis, which we proved in Chap 2 above, concerning the pressing need for upskilling, particularly within the private sector.

Our results in Sect. 7.4 show positive correlations between actual and required education, experience and average wages (Tables 7.8, 7.9, 7.10, and 7.11). We verify hypothesis 4.b. in Chap. 1 above that an increase in skill level and firm size lead to improved relationships between actual and required education (Table 7.8), between actual education, experience and wages (Table 7.10) and between required education, experience and wages (Table 7.11).

In Sect. 7.5 our findings with respect to the positive complementary relationships between skill, technology (ICT) and upskilling (ICT training) and between computers, telecommunications and ICT training (Tables 7.13 and 7.14) are consistent with the findings in the new growth literature. We illustrate and corroborate hypothesis 4.c. in Chap. 1 above that an increase in skill level and firm size lead to an improvement in the complementary relationships between skill, upskilling and technology (ICT).

Taken together, all these results imply the importance of a good education for bridging differences between firms and also for enhancing skill, technology and upskilling complementarity at the micro level. These findings seem consistent with the endogenous growth framework and stylised facts concerning the relationships between human capital, technical progress and upskilling and our theoretical framework in Chap. 3 above.

Finally, our results in Sect. 7.5 indicate positive significant correlations between total spending on ICT and profit and value added, but insignificant correlations between total spending on ICT and output at the micro/firm level (Table 7.17). This result confirms the fifth hypothesis in Chap. 1 above, which implies an inconclusive effect of ICT at the micro level and supports the observations at the macro level in Sudan and the recent literature in the developing countries.

Moreover, our results in Sects 7.4 and 7.5 show the relationships between actual and required education, experience and wages and between skill, technology (ICT) and upskilling (ICT training), defined by firm size and industry level. These results are consistent with our findings in Chap. 5 above, which imply that both skill and technology indicators vary across firms and increase with firm size and industry level.

Therefore, our findings in this chapter are consistent with hypotheses 3.b. and 4.a. in Chap. 1 above with respect to the implications of the excessive use of unskilled workers at the micro level. In addition, our results verify hypotheses 4.b. and 4.c. in Chap. 1 above concerning the relationships between actual and required education and experience and between actual education, required education, experience and wages and the relationships between technology (ICT), skill and upskilling (ICT training). Finally, we corroborate the fifth hypothesis in Chap. 1 above regarding the inconclusive effect of ICT at the micro level.