CONCENTRATION AND SPECIALIZATION OF ECONOMIC ACTIVITIES IN THE KINGDOM OF SAUDI ARABIA

This research presents several indicators for understanding the structure of economic activities in terms of concentration and specialization in the Kingdom of Saudi Arabia. Subsequently to the theoretical description of the different notions of concentration relating to geographical concentration and productive concentration as well as the spatial distribution and specialization of territories, a comparative analysis is presented for all of these indicators. The interpretations in this study are based mainly on statistics of the number of employees and the remuneration paid by economic enterprises. The results obtained are interpreted according to two distinct methods, firstly, geographically, by analyzing the structure of the productive structure of the areas (regions and employment areas), and secondly, sectoral, by studying the spatial distribution of activities in a given sector, which will make possible local comparisons between sectors and areas and with international areas of other countries. Finally, we have proposed approaches and indicators that facilitate the comprehension of specialization and economic concentration. and repair of motor vehicles & motorcycles; Manufacture of other non-metallic mineral products; Specialized construction activities et Warehousing and support activities for transportation.


ISSN: 2320-5407
Int. J. Adv. Res. 8(02), 527-554 529 relation to the average wages paid to all employees in the economic sector. Consequently, the slope of the Lorenzon curve is given by the specificity index defined by the following formula: . As the Lorenz curve has an increasing slope, the income (wages) paid will be ordered in descending order of specificity. The different points of the curve are given on the abscissa by the accumulation of the weights relative to the number of the given sector (zr z ), and, on the ordinate, the cumulative relative contribution of the variable under consideration (x r x) which corresponds to the average wages paid by sectors to employees in the Kingdom of Saudi Arabia.
Graph 1:-The curve of Lorenz income (Distribution wages by sector in Saudi Arabia in 2017).

Source: Authors
For the different sectors studied, the 30% of employees in the smallest strata cover a total of 10 sectors, i.e. 83% of the entire economic sector. The last 70% of the wages are concentrated alone in the trade and food sector and the industry sector. This curve is increasing, as well as its first derivative (Saporta, 1990& Jayet, 1993. If the distribution of wages is perfectly uniform between individuals, this Lorenz curve will merge with the main diagonal. On the contrary, it will deviate from this main diagonal for a differentiated distribution of wages between employees according to the different economic sectors. The Lorenzon curve identifies the average wage observed for each employee, representing both locals and foreigners, and male and female. Consequently, the situation for both sectors (industry and trade) should be monitored by the Commercial Registry Centre with regards to the creation and deregistration of enterprises.
In order to measure the spatial distribution of wages (in the different regions) of employment for all sectors, we will use the share of the area in the total income of the reference territory as the relative weight of the area (zr z) (in ordinate), and the part of the cumulative area as a relative part of the interest variable (x r x) (in abscissa). The Lorenz curve below (graph 2) is constructed to measure the distribution of employees (incomes) according to the different regions, which is established on the basis of the data from table 2 in annex 1 for all employees without distinction of gender.
And it emerges from this curve that employees in 10 out of 13 areas receive 10% of all wages paid. This situation is justified by the presence of 21.06% of employees in these 10 areas, which is essentially caused, spatially, by the vastness of the desert and, on the other, by the occupation of the land according to climatic conditions.

Source: Authors
Conversely, if we want to measure the concentration of employment by region in the economic sectors and the wages paid by gender (male and female), we can have the following ordinates (zr z) the part of all wages paid by all sectors in total employment for the reference territory and in the different areas, and on the abscissa (x r x) the contribution of each sector to employment in the area under consideration. The Lorenz curve below (graph 3) is drawn to measure the distribution of employees (income) according to the different areas, which is established on the basis of the data from table 2 in annex 1 for all employees by gender (male and female). Source: Authors A situation that remains the same in all three areas (North bord, Al-Baha et Al-Jouf) which, as a result of the population level of employment will receive only 3 and 4% of the wages paid by the economic sectors for male and female employees respectively, the situation is improving for female workers compared to male workers in the  regions of Jazza, Tabuk, Aseer, Al-Qaseem and in Al Madinah El-Mounaoura the wages received by women are double those of men (accumulation of regions), and to Estearn region, the wage situation is the same for men and women. For the rest of the areas, employees in Makkah El Moukarama and Riyadh receive 40% of the wages paid by economic sectors to employees. These two cities have the following characteristics, one is the capital of the Kingdom of Saudi Arabia and the other is a saint city, a destination for more than a billion Muslims.
Conversely, if we seek to measure the concentration of employment and income paid by economic sectors by age group, we can have an ordinate of (zr z) the part of all wages paid by all sectors in total employment by reference age group at the national level, and on the abscissa (x r x) the part of each age group in the employment of the age range under consideration. The Lorenz curve below (graph 4) is constructed to measure the distribution of employees (income) according to the different age groups and which is based on the data from Table 3 in Annex 1.
Graph 4:-The curve of Lorenz (wages by age group in Saudi Arabia in 2017).

Source: Authors
These age groups made up of the following fringes: the under 20 years (15-19 and 20-24) and 55+ (55-59, 60-64 and 65+) receive 11% of the wages paid by the economic sectors. The rest of the age groups receive an almost equitable distribution of wages. Nevertheless, the labor force remains the key element in wages.

The Gini Index in the Kingdom of Saudi Arabia:
The Gini Index (Gini, 1947;Gini, 1965) is used to summarize the information read on the Lorenz curve. It is twice the area between the curve and the first bisector. This coefficient has values between zero, when there is a uniform distribution of the variable between the different individuals, and the unit (=1), when a single individual has the entire variable under consideration. Between these two extremes, there is a positive value that increases when the Lorenz curve moves downwards and there is an increase in inequality. To be noted that (Silber, 1989) and  have Gini coefficients disaggregated into distinct and defined contributions of total inequality.
From a practical point of view, a first method of calculation is by ordering the N observations in ascending order and using the formula: Gini= is the mean of the variable considered. A second method is to calculate the surface under the Lorenz curve and to deduct the Gini coefficient accordingly.  0  10  20  30  40  50  60  70  80  90  100  532   The calculation of the Gini index for the case where individual data have been grouped by areas or sectors, the  relative  weight  of  the  area  or  sector  is  used  according to: For purposes of further comparison, we will adopt the most widely used formula given in the World Bank's explanatory memorandum for the calculation of inequality in the incomes (Brown, 1994), which is as follows: Where X is the cumulative part of the variable to be studied and Y is the cumulative part of the mass to be distributed.
The Gini index is an indicator of the distribution of a mass (wages or income, wealth, etc.) within a population in order to get an idea of the more or less egalitarian nature of the distribution of this mass within the population and to compare the results of the distribution with other countries.
It is academically recognized that the Gini coefficient is between 0 and 1, where 0 represents perfect equality (everyone has the same resources) and 100 represents perfect inequality (resources are monopolized by a single person or category). The Gini coefficient is the ratio of the surface area between the diagonal of perfect equality and the Lorenz curve to the total Surface below the line of perfect equality. We deduce that, in terms of the sectoral distribution of wages, the Gini index measures 40.56, a level that can be classified between the European and US levels, in terms of the distribution of wealth. The Gini index for the distribution of wages across age groups is of the order of 1, which makes it possible to attribute a perfect distribution of wages in this respect (a perfect level). And with regard to the distribution of wages by region, for the overall case of employees (men and women) or only employees (men), the index varies between 55.98 and 56.93, similar to Brazil, but much better than those attributed to the Latin American region and the world average, and therefore, this index represents an appreciable level. And as regards women employees, the distribution as measured by the Gini index is equivalent to 35.78, a level that is comparable to that of the European Union countries, including France.

Concentration measures in the Kingdom of Saudi Arabia:
The purpose of measuring concentration will be to give a summary of the distribution of the size of enterprises in terms of employment or income in a sector or area. The following concentration indicators must satisfy the Lorenz conditions. First, if there is an increase in the dispersion of the distribution with the same mean, this should increase the concentration index. The second condition establishes that if all individuals are of the same size, an increase in the number of individuals must lead to a decrease in the concentration index. The measures of concentration were proposed like (Hall & Tideman, 1967) and the Entropy Concentration Index was proposed by (Jacqueminn, 1975;1079).

The proportion of the largest enterprises in the Kingdom of Saudi Arabia:
The N firms are ranked in descending order according to the variable in question. We thus note the relative size of enterprise i as measured by employment: = ∑ =1 such as: 1 ≥ 2 ≥ ⋯ ≥ ≥ and we then define the portion of the m largest enterprises by the formula: We take into account the contribution of the 4 and 10 largest sectors with the largest number of large enterprises referred to respectively as C4 et C10 . These indicators allow calculating the concentration of wages in the 4 and 10 largest sectors with large firms with more than 20 employees. These indicators C4 et C10 allow us to determine the proportion of employment in the 4 or 10 sectors (the largest enterprises in the Kingdom of Saudi Arabia) constituting a strategic volume requiring monitoring by a strategic information system. We take the same approach but by taking into account the 4 and 10 areas with the most employment (depending on the sector). These indicators C4 et C10 allow us to consider the geographical areas (regions) and sectors in which the most employees are working.
Note that the ranking in descending order of enterprises by economic sector employing more than 20   We deduce that the number of enterprises (companies with more than 20 employees) in 2018 is 85134, or 2.62% of the total number of enterprises. This number varies according to economic sectors, ranging from 7 to around 611,000 enterprises with an average of around 34,000 enterprises. Ranking the number of companies in descending order leads us to conclude that the 10 sectors with more enterprises employing more than 20 employees represent a number 1111619 and this number rises to 664606 for the first 4 sectors employing more than 20 employees for the year 2018. The top 10 as well as the top 4 sectors employing more than 20 employees represent respectively 56.40% and 33.56% of all employment in 2010.

The Herfindahl Index in the Kingdom of Saudi Arabia
Various measures of diversification have been proposed in academic work. The most widely used category, whether for studies in industrial economics (Jacquemin et Berry, 1979) or regional economic studies (Attaran et Zwick, 1987) which is based on the calculation of Herfindahl-Hirschmann concentration index and whose measure is the sum of the squares of the parts of all the individuals: = ∑ = (Nutter, 1968). The HHI results, for a given economic sector group, from the sum of the squared employment shares.

534
∈ Represents corporate employment i and N represents the total number of enterprises in the economic sector group. This index can be as high as 10000.
The inverse of this HHI index can be interpreted as the equivalent number of firms of equal size that would share the employment equally. The current guidelines suggest that a market with HHI less than 1,500 is "unconcentrated" while an HHI of greater than 2,500 is highly concentrated (Bruce and al, 2014).
We opt for the case where IHH< 1000, that the distribution of wages according to the chosen parameter (regions, economic sectors or age groups) is uniform: the N parameters are of similar size and no risk of dependency or dominance is presented. And under this condition, employment is not at all concentrated in the area (or economic sector or at the level of an age group). Thus, from Table 1  For IHH by sector for 2017 = 1957, a value ranging between 1000 and 2000, it indicates that employment is not highly concentrated in one economic sector compared to other sectors. Nevertheless, a predominance of the sector (Trade, Accommodation and Food) over overall employment, the regulator (legislation and regulations currently applicable on trade, taxation, immigration and vocational training) must ensure that a more favorable business environment is provided for this employment sector, particularly for enterprises employing between 1 and 5. Using the same estimation method, the HHI index by age group is calculated from Annex 1, Table 3, which equals: IHH by age group in 2017 = (0.14) 2 + (1.00) 2 + (2.99) 2 + (29.77) 2 + (17.45) 2 + (54.12) 2 + (221.13) 2 + (147.11) 2 + (272.88) 2 + (291.54) 2 + (372.94) 2 = 1411 For l'IHH by age group in 2017 = 1411, a value ranging between 1000 and 2000 which indicates that employment is low concentrated in one age group in comparison to other groups. Employment by age category is more equal for 6 out of 12 categories. However, the 34-44 age group alone represents 52.91% of the total employment, which is distributed around 17% for each of the 30-34 and 40-45 age groups. While the age group from 35 to 39 years old stands out with the higher employment rate which is equal to 19.31%. However, the 34-44 age group alone represents 52.91% of the total employment, which is distributed around 17% for each of the 30-34 and 40-45 age groups. While the age group from 35 to 39 years old has the highest employment rate which is equal to 19.31%. However, we recommended that l'IIH by age group in 2017 does not increase more than 250 (equivalent to 16% in overall employment) so that the situation will not be concentrated from one category to another, with the exception of a favor granted to the 20-29 age group for better para-tax participation (more contributions from this category to the pension and insurance fund) which makes it possible to ensure the financial equilibrium of the fund and long-term sustainability.
After calculating the percentage rates of each area's contribution to employment, the HHI employment index for the areas is derived from Table 2  Three areas (region) are the most job-seeking regions and these concerns Eastern Region with 19.63% followed by Makkah with 22.04% and Riyadh with 37.27%.
Thus, the Herfindahl-Hirschmann index by area IHH by region in 2017 = 2323 denotes a concentration of employment in the capital Ryadh. In order to ensure a balance of concentration, the regulators must ensure that this index does not increase more than 150 (about 12.2% of total employment).
In conclusion, independently of the parameter chosen (sector, area, or age group), there is no distinction between a borderline case and a deregulation of concentration (all employees working) in a sector or area or age group haven't a dominance that presents a risk concentration.

The Herfindahl index normalized in the Kingdom of Saudi Arabia:
There is also a standardized HHI index. While the HHI varies between 1/n and 1, the normalized HHI index varies between 0 and 1. Its formula is: From this formula, the Herfindahl-Hirschmann concentration index normalized for sector, area, and age category parameters is calculated, resulting in the following values : As a deduction to these calculations, the sector-standardized Herfindahl index is the only one that lagged above 2000 and is joined to the same analysis attributed to the regions. The table below shows the analysis according to Herfindahl-Hirschmann Index (HHI) and Herfindahl-Hirschmann Normalized Index (HHI*). These indices remain sensitive to high ∈ .

Comments
The employment is not concentrated in the area or sector or age group.
Employment is low Concentrated in area or sector or age group.
The job is concentrated in area or sector or age group.  This Theil entropy index also has a decomposition property. When the population is divided into several groups (j=1..n), The general entropy index is subdivided into two elements: intra-group entropy, measuring inequalities within each group, and inter-group entropy, measuring inequalities between different groups. We will use this peculiarity in the calculations of the Theil index by region by decomposing employees by gender (male and female), which indicates the relative importance of gaps in the degree of income concentration for each category  Table No. 7 in Annex 1 with salaries corresponding to each "male and female" gender and decomposed) The first of the two addenda in the penultimate column (of Table 7 in Annex 1) represents the contribution of inequality between male employees to the total Theil index. The second is the inequality resulting from the distribution among female employees. For its own part, the last column calculates the contribution to the total inequality of inequality between the two categories of female employees by region. The total Theil is the sum of the previous components.
Finally, the result of the sum of T l and T 2 is equal to (1.277) which does not correspond exactly to the value of the total inequality coefficient calculated previously (1.0374) because of the values used. For the first result, the average wages between the male and female category for each region (group), for the second result, the use in the calculations of the wages relating to each category (male and female) for each region (group).
Thus, the results of T 1 =1.02109 and T 2 = 0.25597, reveal that inequality between the two groups (male and female) of wage is more important in total inequality than inequality between regions.
As in the case of the Herfindhal index, a minimum and maximum value of this entropy index can be calculated. Thus, if all the parameters are of equal size with the same size 1/N ratio, the entropy index will be equal to log(N), whereas the maximum value (0) is obtained if only one parameter holds all the employment (all incomes are equal zero except one).

Source: Authors
As a result, no classification admits any concentration of wages or employment, a satisfactory situation; nevertheless, it is advisable to improve the wages of the women's category by tax moderation.

The Gini employment index by sector and region:
For (Davezies & Pech, 2014), Spatial concentration is measured by considering the totality of the spatial structure(s), i.e. using a synthetic indicator. The synthetic indicator most often used is arguably the Herfindahl index for absolute spatial concentration and the Gini index for relative concentration.
The distribution coefficient is also a measure of the concentration of sectors in geographical areas, i.e. the distribution of employment in a sector among the different areas of the territory. It is a Gini index for each geographical area weighted according to its contribution to total employment in the territory. It takes into account the weight of the areas in total employment. From Table No. 8 in Annex 1, the relative Gini coefficients of employment by area and by sector are calculated according to the method formulated above and represented in the following graph: Graph 5:-Gini coefficients by area (all sectors combined) and by sector (all areas combined) for employment.

Source: Authors
The coefficients of variation in employment defined by the formula : = x by area (all sectors combined) and by sector (all regions combined) are all above 100% and they vary respectively from 111% to 150 and 133% to 358%, which indicates that they do not have the same dispersion and they remain heterogeneous, but it should be noted that it would be easier to reach a consensus (more repeatability in the data) by area than by sector and the area remains the most representative is the East Province. An area can take on the character of a pilot area where any observation reported on a sector is significant for the whole of the Kingdom of Saudi Arabia and from which; it is recommended that an observatory on employment, productivity and competitiveness be established).

Economic density
Economic density is used to evaluate the potential of an area to attract employment and expresses productive character than residential. Hence the economic density, noted and wicth based on the hypothesis developed by (Dreier et al., 2001) and that where we live influences, the quality of life and the manner in which that place operates impacts the quality of our society. Economic density is a measure of the importance of economic activity in an area. Economic density of an area j is calculated in a manner analogous to the population density such as the number of employees per km².
These two densities are highly correlated.
The economic density in the analysis also provides an estimate of the benefits and constraints resulting from the relative concentration of actors. And for greater visibility; an employment rate for the area can also be constructed.

Source: Authors
Development increases density which in its turn enhances attractiveness and stimulates expansion, which further improves attractiveness (Ciccone et Hall, 1996). We can note that areas with a high economic density offer greater support for the competitiveness of businesses. Nevertheless, work on French territories (Binet et al., 2010) Point up the statistical problems involved in measuring economic density, given its very uneven distribution, which leads to its abandonment. There remains a direct relationship between the territorial context and entrepreneurial activity. And it goes in the same direction as most of the work concerning the local determinants of business creation (

Measures of specialization and specificity:
We note that all calculations relating to specialization and specificity measures are based on the data in Table 8 in Annex 1.

539
The specialization coefficient by area: The notion of specialization applies to a spatial unit. Specialization is relative in nature. The concept of specialization is based on the comparison of two sectoral structures. Thus, the specialization of the production or employment of an area in relation to the country is understood by comparing the sectoral structure of production or employment of the area with the sectoral structure of production of the country, and by using the production or employment of the country as a reference element (Vermaut, 2003). Specialization indicates whether an area's production or employment is more or less oriented towards a particular activity. It is also an indicator of diversification. The employment specialization coefficient measures the concentration of an area's employment in a greater or lesser number of sectors. It is a relative Gini index, i.e. each sector is weighted by its proportion of total employment. Thus, this coefficient reflects the importance of each of the sectors at the aggregate level of the reference territory.
Each relative specialization coefficient is the expression of the ratio of the weight of sectors i in the total economic activity of a specific employment area j in relation to the relative weight of the same activity at national level (Aiginger, 1999). It is obtained from the following report: That is to say:

=
The results of sectoral specialization measure the ratio between the sectoral structure of the employment area under study and that of the entire Kingdom of Saudi Arabia. Three classes were defined: over-representation (specialization index >1.25), in the average (between 0.75 and 1.25) and under-representation <0.75. The following table is derived after the calculations according to the above formula.

Krugman Specificity Index
The disadvantage of the indices of sectoral specificity of areas stems from the fact that there are as many of them as there are sectors considered for each area of the territory. These can be analysed in order to find the sectoral specialities of the area, but this does not provide an indication of the overall specificity of the area. In order to 541 remedy this inconvenience and to measure the overall specificity of the zone, the Krugman global specificity index will be used (Krugman, 1991).
This index calculates the difference between the industrial structure of the area and that of the other areas of the reference territory (Fujita & Thisse, 1996). Thus, by avoiding comparing the area to the whole territory, one avoids biasing the measurement for large areas that are always nearer to the average industrial structure. The Krugman index is therefore the sum of the differences in absolute value between the industrial structure of the area and that of the rest of the reference territory defined according to the following formula: For our case study, the economic activities or industrial sector (structure) is measured across its employment base, consequently, is the proportion of employment in sector i of administrative area k and ̅ is the proportion of employment in sector i of the reference group (the whole country). It measures the absolute distance between the relative importance of a sector (between k and the reference group) and then adds all sectors together to generate an index. The Krugman index will be equal to zero when the area in terms of employment is perfectly similar to the rest of the territory (country); the area has no specificity because it perfectly reflects the structure of the territory. However, if the area is entirely specialized in activities not found elsewhere (employment dominance), the Krugman index will be equal to the number 2, and we will have a perfectly specific area. Following the above formula, the Krugman specialization index is obtained from Table 8 in Annex 1.  The administrative area is relatively specialized 20%  K i < 100%  The administrative area is very specialized  Perfect specialization of the administrative area Specificity indexes:

542
The specificity considers the nature of sectoral activities by comparing the structure of the area with that of the territory as a whole. The specificity of the area in terms of employment is confirmed by (Dissart et al., 2011) which shows the influence of market potential, propensity to consume locally, as well as local attractiveness and access to specific equipments. The specificity helps to compare the production or employment structure of an area in relation to a reference territory.

Indices of sector specificity:
The sector specificity index allows a comparison of the importance of a sector of economic activity in the study area and in the territory as a whole. The specificity index of a territory or an economic sector corresponds to the ratio between the number of employees in a sector in the total number of employees in a given territory and that of the reference territory. It is generally defined by: This index is equal to 200 if sector k has the same importance in area i and in the territory as a whole; it is equal to zero if no activity of sector k is located in area i. When all employment in sector k is located in area i, it equals X/X i . If the result is superior to 100 then the sector is over-represented locally. But if this index is inferior to 100 then the sector is under-represented locally. The maximum value therefore depends on the size of the area, which can be annoying when we study an area that is large in relation to the reference territory.
We define four classes: sector k, is not represented at all when = , a strong representation when ≥ , of average representation for ≤ < and weak representation for the < < Prov. 543

The Krugman Specificity Index
The Krugman index is therefore the sum of the differences in absolute value between the industrial structure of the area and that of the rest of the reference territory defined according to the following formula: Thus its value is between zero and two. The Krugman index can be interpreted as the sum of the divergences, taken in absolute value, of the specificity index of the area in relation to 2, i.e. a situation where the area has the same proportion of employment as the whole territory. This sum is adjusted by the importance of sector k, measured by the relative share of employment in sector k in the rest of the territory. As a result, a synthetic index of the specificity of the area in relation to the rest of the territory is obtained (Krugman, 1991).
From the above, if the area is similar in all sectors to the rest of the territory, we will have for all : ̌= because =̃ or else = = = . As a result, the Krugman index will be equal to zero and it can be said that the area is perfectly similar to the rest of the territory: it has no specificity as it reflects perfectly the sectoral structure of this territory. However, if the zone is entirely specialized in activities that cannot be found elsewhere, the Krugman index will be equal to 2 and the area will be perfectly specific.
This index can also be interpreted as the percentage of sectoral reallocation that should take place in the area in order for it to be perfectly similar to the rest of the territory in terms of productive structure or employment. In fact, this Krugman index is the sum of the differences in percentage between the productive or employment structure of the area and the rest of the territory.
Therefore, in order to resemble the rest of the territory perfectly and to have zero specificity, all the specificity indices must be units (̃= 1) what occurs through a redistribution of productive activities or employments between sectors (Kubrak, 2013).
 The area is very specific =  Perfect specificity of the area

The prevalence index between specialization and specificity:
The Krugman Specialization Index (K i ) as a measure of specialization that is broadly used. We propose the calculation of the global specialization index, which can be considered as a relative specialization that could be compared to another country or a reference country group. This Global Specialization Index expressed as a percentage is calculated according to the following formula: = In case this prevalence index is greater than 1, the preference for bilateral comparison between regions based on the specialization index is the best option, otherwise, or l'IP S G is less than 1, the preference for bilateral comparison by region established on the specificity index would be preferred. For our case study on the index in the Kingdom of Saudi Arabia, the prevalence index between specialty and specificity is = = , % , % = , . In from this index, we will use the bilateral specialization index for our calculations.

Bilateral specialization indexes by administrative area
We can also propose a bilateral specificity index between two areas to search for areas that most resemble the study area or areas that are the furthest away from it in terms of industrial structure or employment. For this aim, the Krugman index is adjusted in such a manner that the comparison of employment by economic sector is no longer made with the rest of the territory, but with another zone j of the territory: Thus we construct a square (symmetrical) table of these bilateral indices which are interpreted in the same way as the Krugman index: if the index is near zero, the two zones will be very similar in their industrial structure, otherwise the index will indicate the percentage of reallocation of activities that would have to be implemented in zone i in order for this zone to have the same employment structure as zone j.

545
It should be noted that we exclude from our evaluation the sector «Other service activities» for its negligible values.

Source: Authors
The numbers colored in Azure blue correspond to a bilateral specificity index between two areas that most similar to the study area in terms of employment, and the numbers write in red correspond to the areas that are most distant from it in terms of employment and industrial structure.

Conclusion:-
Indeed, in a context of international competitiveness, territories are the source of many of the diversity, specialization and concentration factors that determine the success of business creation: infrastructure (transport, energy, and telecommunications), employment areas, the presence of research and training centers, and the quality of the pool of outsourcers and suppliers. In addition, the era of the knowledge-based economy, marked by the predominance of tertiary activities, innovation and dematerialization, and the ability to attract and retain foreign direct investment are the conditions that make the area attractive to businesses, considering factors such as a pleasant quality of life and a positive image as a consequence of local development.
These material and immaterial factors, whose control requires specialization or concentration in order to concretize a strategic alliance between the enterprises and the territory, Therefore, the analysis of local determinants implies identifying the distinctive characteristics of territories through the estimation of concentration and specialization that can contribute to the understanding of the trajectories of enterprises (creation, disappearance, increase and foreign direct investment). In addition, this estimation constitutes a warning factor regarding the trends of the sectors of activity and their impacts on employment, the distribution of the wealth and the growth at the level of each administrative area. The fact remains that the specialization of a territory favors the flow of information, innovation and, more generally, agglomeration economies (Maurel, 1996) related to Marshall-Arrow-Romer (Scitovsky, 1954) and, to which the spatial concentration of enterprises in the same sector of activity established on a network of links that promotes local growth. For (Jacobs, 1969), the diversity of activities is a favorable factor for growth insofar as 546 complementarities of knowledge, technologies or products could emerge. And therefore, inter-sectoral agglomeration economies according to the sectoral diversity of an area must be the issue at stake.
Specialization refers to the mastering by the territory of a number of knowhow related to a sector of activity or a product. It corresponds in a certain manner to industrial districts and, more broadly, to localized production systems.
Thus, we propose a method to calculate the intermediate Krugman specialization index of a country's = (∑ = )/ . This index helps a priori to make a differentiation reading between specialization and concentration (the dominance of one over the other) and to make international comparisons in view of the divergence of classifications of economic activities in the statistical system of every country. We reiterate the Krugman formula: And in regard to the sectors of activity in the Kingdom of Saudi Arabia; N= 13 and for the regions N= 10 and that (∑ = = 364.95%, value obtained from the sums of the values in Table 5  Finally, the concentration reflects a situation of dependence on a small number of economic agents and makes the economy vulnerable to the decisions of large firms which contribute to the production or distribution of goods and services. By contrast, specialization makes the territory dependent on a given sector and extremely vulnerable in case of sectoral crises. All in all, for our case study, we consider a tolerable level or acceptable threshold when K T/R or K T/R K T ≤ 33% considering the margin of error (the variances) and for the case of the Kingdom of Saudi Arabia, no predominance is defined between specialization and concentration. And we consider a situation of dependence or vulnerability to economic cycles when K T/R or K T/R K T ≥ 50%.