*3.2. Methods*

The methodology of this national study is presented in detail in an open publication [45]. The sample of research objects was 2,963 organizations, which provides a probability of 97% and a sampling error of 2% from the total population of 1,185,163 business entities. Sector quotas were carried out on the basis of GDP share, since costs are part of the final cost of the product.

As part of the study, a mass survey was conducted where 5,546 subjects were interviewed: owners, managers, and accountants. The sample corresponded to the planned probability level of 97%. The survey form contained open and closed assessment questions on the Likert scale. Based on a mass survey, the impact of costs on energy supply for enterprises was assessed using the following rating scale: "costs did not change", "increased (up to 10%)", "increased very (10–20%)", "grew very much by 20–30%", "critically increased from above 30%", and "I find it di fficult to answer" (Table 1).


**Table 1.** The share of organizations with a growth in production costs of more than 25%.

Source: [7].

Electricity costs and transaction costs for energy supply monopolies are among the most significant for the economy of the republic. Kazakhstani managers point out that rising energy prices are a serious problem for business development. According to Kazakhstan managers, a significant problem for business development is primarily the rising prices for fuels and lubricants, energy, and raw materials. More than a third of respondents called energy supply costs a serious obstacle to current activities in industry and services. The scale of the answers allowed us to name the urgen<sup>t</sup> problem of managing these costs and consider them as part of the broader problem of ine fficient energy management.

The authors investigated the formation of energy managemen<sup>t</sup> e fficiency as control systems at various levels (see Figure 1).

**Figure 1.** Schematic representation of energy managemen<sup>t</sup> levels. Source: personal elaboration.

The level of energy costs was formed under the influence of factors of all levels. Each energy managemen<sup>t</sup> level has a specific set of influence tools (Table 2).



Source: personal elaboration.

Energy managemen<sup>t</sup> problems become more acute when assessing their effectiveness at various levels.

At present, the state manages the electric power industry through such levers as licensing, setting limit tariffs, regulating the activities of natural monopolies, etc. From the beginning of 2019 in Kazakhstan, along with the electric energy market, there is an electric power market. The tariff is divided into two components:


However, the average levels of energy supply costs that have formed in various regions of the country show the heterogeneity of the implementation of state policy (Table 3).


**Table 3.** The share of electricity costs in the total cost structure (%).

Source: personal elaboration.

The spread of electricity costs is significant across the republic. The average value of the indicator for manufacturing enterprises is 7.2%, with a range of variation of 13.4% (R = Xmax − Xmin = 15.9 − 2.5 = 13.4). In the most energy-intensive industrial region of Pavlodar region, where energy-intensive metallurgical production is located, the average cost of electricity is 15.9% of the cost structure. The lowest indicator −2.5% falls on the Zhambyl region with low volumes of industrial production.

Graphical analysis involves checking the normal distribution, which would be characteristic for the implementation of a unified energy managemen<sup>t</sup> policy in the republic (see Figure 2).

**Figure 2.** Distribution of the average share of energy costs across the republic (%). Source: personal elaboration.

To assess the degree of di fference in the territorial significance of energy costs, standard indicators of variation were calculated.

Coe fficient of variation—as a measure of the relative spread of the values of the population—shows the ratio of the standard deviation to the mean.

$$v = \frac{\sigma}{\overline{\chi}} = \frac{3.136}{7.2} 100\% = 43.56\% \tag{1}$$

Each value of the series di ffers from the average value of 7.2% by an average of 3.136%. Since the coe fficient of variation is within 30% < v < 70%, the variation is moderate. The presence of moderate variation may indicate the presence of a single (state) policy in the field of energy supply management. However, such national energy managemen<sup>t</sup> is not e fficient enough, which creates an asymmetry of indicators. The degree of asymmetry determines the moment coe fficient of asymmetry.

$$\mathbf{As} = \mathbf{M}\_3 \mathbf{/S}^3,\tag{2}$$

where M3 is the central moment of the third order, S is the standard deviation.

M3 = 482.38/16 = 30.15, As = 30.15/3.1363 = 0.977

A positive value indicates a right-handed asymmetry. The right-sided asymmetry of the values of the share of energy costs indicates a shift in costs towards their increase, which may arise due to insu fficient e fficiency of state energy management, which increases costs due to an increase in the administrative burden. In the future, this assumption should be checked, including the possibility of corruption.

The industry level of energy managemen<sup>t</sup> was considered separately for production and services. The industry-wide assessment of the level of influence of energy costs in production is presented in Table 4.

As noted above, electricity costs are in the top 10 costs of industrial enterprises. The share of the item "electricity" in the total cost structure of manufacturing enterprises also has a spread from 6.2% to 10.5% in the processing of agricultural products. However, graphical analysis suggests the proximity to the normal distribution and the relationship between cost increases and the assessment of the strength of influence on the managemen<sup>t</sup> model (Figure 3).

**Figure 3.** Effect of energy costs in production.

Note that the impact of the cost item was assessed according to the responses of managers who indicated that the cost of electricity at their enterprise increased by more than 25% and had a significant impact on the production process. Respondents' answers are a subjective assessment but reflect the e ffectiveness of production managemen<sup>t</sup> models that can regulate the overall cost structure and production e fficiency.


Source: personal elaboration.

**Table 4.** Sectoral assessment of the level of impact of costs on electricity (production).

The relationship between the opinions of managers and the share of costs was checked using Spearman's rank correlation coefficient.

Since among the values of the signs x and y there are several identical ones, i.e., related ranks are formed, in this case, the Spearman coefficient is calculated as:

$$\mathbf{p} = 1 - \frac{\sum \text{Gd}^2 + \text{A} + \text{B}}{\text{n}^3 - \text{n}},\tag{3}$$

where

$$\begin{array}{c} \mathbf{A} = \frac{1}{12} \sum \left( \mathbf{A}\_{\mathbf{j}}^{3} - \mathbf{A}\_{\mathbf{j}} \right) \\ \mathbf{B} = \frac{1}{12} \sum \left( \mathbf{B}\_{\mathbf{k}}^{3} - \mathbf{B}\_{\mathbf{k}} \right) \end{array}$$

j is the number of ligaments for characteristic X.

Aj is the number of identical ranks in the jth connective along X.

k is the number of ligaments for characteristic Y.

Bk is the number of identical ranks in the kth bunch according to U.

$$\mathbf{A} = [(53 - 5)] / 12 = 10$$

$$\mathbf{B} = [(23 - 2)] / 12 = 0.5$$

$$\mathbf{D} = \mathbf{A} + \mathbf{B} = 10 + 0.5 = 10.5$$

$$\mathbf{p} = 1 - \frac{6 \times 112.5 + 10.5}{10^3 - 10} = 0.308$$

 The calculation shows that the connection between the attribute Y and factor X is weak and direct.

The presence of a direct, but weak connection shows that, in production, an increase in electricity costs and an increase in their share in the total cost structure leads to the need to change production managemen<sup>t</sup> models. However, the specifics of production are such that, in the value chain, energy is an input resource and is involved in the process of creating a product at the initial stage. The increase in the cost of electricity in production falls on the total costs and in the future the negative impact is smoothed out either by increasing the price or by adjusting the costs at the subsequent stages of the chain.

## **4. Results**

Processing interviews with Kazakhstani managers shows that they do not consider energy saving at the production stage as a priority task of developing production and increasing its efficiency. The model of Kazakhstan's energy managemen<sup>t</sup> at the enterprise level can be assessed as passive.

An assessment of the level of impact of electricity costs in the services sector is presented in Table 5 and differs significantly from the situation in production.

The authors believe that in the service sector a fundamentally different role is played for electricity costs in the value chain. Due to the specifics of the service (it is impossible to save, separate from the consumer, etc.), electricity consumption is necessary at almost all stages of the chain. Different types of services require a different share of energy costs and fundamentally different energy managemen<sup>t</sup> models are implemented. Graphical analysis also shows differences and contradictions in assessing the strength of the impact of rising energy costs in various types of services (Figure 4).


**Figure 4.** The impact of energy costs in the service sector. Source: personal elaboration.

The relationship was also checked according to the Spearman rank correlation criterion (see Table 6).



Source: personal elaboration.

Since among the values of the signs x and y there are several identical ones, i.e., related ranks are formed, in this case, the Spearman coefficient is calculated as:

$$\begin{aligned} \mathbf{A} &= [(43 - 4)] \text{l/} 12 = 5 \\\\ \mathbf{D} &= \mathbf{A} + \mathbf{B} = 5 \\\\ p &= 1 - \frac{6 \times 226 + 5}{10^3 - 10} = -0.375 \end{aligned}$$

Calculations showed that the relationship between trait Y and factor X is weak and inverse. In a number of types of services, an increase in the share of electricity costs does not lead to a strong influence on the overall business efficiency. IT services have the largest share of energy costs in the overall cost structure, but their managemen<sup>t</sup> is such that they are less concerned about the growth of these costs. At the same time, public catering has a rank of 2 in terms of costs, but in assessing the

impact a rank of 9. In this type of service, energy supply is involved at all stages of the value chain from cooking to creating conditions for consumption.. The cost of electricity in this type of service is not so high in terms of the share of total costs, but catering managemen<sup>t</sup> is energy-sensitive. A similar situation exists in the field of accommodation services.

In general, the energy managemen<sup>t</sup> of Kazakhstan's business services sector is more heterogeneous than in production. Non-volatile control models and energy-sensitive models are used, where electricity takes part in the chains of creating the material and emotional value of the service. Such industries (food, accommodation, and recreation) have not ye<sup>t</sup> developed an e ffective energy managemen<sup>t</sup> model.

Energy managemen<sup>t</sup> at the enterprise level is formed not only under the influence of industry or territorial factors, but is also largely determined by the dimension of the business. During the study, an assessment was received of 83 microbusiness entrepreneurs, 817 small businesses, 348 medium-sized enterprises, and 100 large business companies. Electricity costs have a steady tendency to decline in proportion as the enterprise grows. The share of electricity costs in large business is three times less than at the initial stage of development, i.e., in microbusiness (Figure 5).

**Figure 5.** The portion of electricity costs for a business of various sizes. Source: personal elaboration.

## **5. Discussion**

As the study showed, for small businesses, the cost of electricity is more significant, because enterprises do not have e fficient energy managemen<sup>t</sup> models. Large enterprises build their own energy managemen<sup>t</sup> models taking into account international experience and global influence factors. The state provides for the requirements to introduce energy managemen<sup>t</sup> systems in enterprises consuming more than 1050 tonnes of oil equivalent per year, while ISO 50001 was chosen as the main energy managemen<sup>t</sup> standard, for which the corresponding state standard was adopted and approved.

Small-and medium-sized enterprises need help in mastering energy-saving models. According to the estimates of Kazakhstan Institute for the Development of Electric Power and Energy Saving (JSC), on average, enterprises can reduce energy consumption by 10% by budget organizations and small businesses up to 40%.

In addition to direct production costs, enterprises in Kazakhstan bear the additional administrative burden of providing electricity. In the Doing Business—2019 rating, Kazakhstan took 67th place [46] in terms of connection to the energy supply system. Every fourth entrepreneur noted the strong impact of cost costs in obtaining permits from the state (monopolists and the quasi-public sector) and in the process of state control. Our studies show that, in addition to the costs directly related to the paymen<sup>t</sup> of energy consumption services, business entities have to bear serious financial costs for obtaining additional services of monopolistic enterprises.

In the course of a survey of specialists from companies conducting financial and/or accounting, the following types of additional services of natural monopolies holder (NMH) in the energy sector were found and analyzed:


The administrative burden for energy is not only an extensive list, but also a large variation in prices for similar services. The authors carried out a secondary grouping of the cost of services for connecting to the elements of the electric grid infrastructure (Figure 6).

**Figure 6.** Cost of services for connecting to elements of the power grid infrastructure. Source: personal elaboration.

Data on the cost of services were brought to six conditional levels. The first level at the time of the assessment was 20 thousand tenge (equivalent to 64 dollars at the rate in April 2017). The third level is from 300% to 500% of the first level, while the sixth level includes costs exceeding the initial level by six times or more. According to the article "Obtaining permits from NMH when carrying out work related to changing the electricity metering scheme" for 10% of entrepreneurs, marginal costs were 35 times higher than the services that 80% of enterprises received at the base cost.

We did not set a special goal to investigate the problem of corruption, but Petrenko et al. [45] indicated the significant role of corruption in the economy of Kazakhstan. A special role in this context is given to the energy sector, in which corruption can reach 75%. To a large extent, for this reason, a significant gap in the cost of similar services across territories and entities is noted not only by the low level of energy management, but also contains a corruption component. The study identified the costs of enterprises associated with corruption costs (Table 7).



Source: personal elaboration.

The costs incurred by enterprises under the article "Obtaining permits from NMH when carrying out work related to changing the electricity metering scheme" averaged 277.5 thousand tenge (890 dollars) according to the calculations for the year; 446.5 thousand entities used the service and incurred cumulative costs of 123.9 billion tenge (397 million dollars). According to experts, this service should be completely canceled. Connection (disconnection) of electrical installations to electric networks of energy-transmitting organizations on average in the republic costs 183.3 thousand tenge (588 dollars); it was proposed to reduce it by 50%, which would lead to a reduction in costs of 41 billion tenge (131 million dollars).

According to the maturity model of energy managemen<sup>t</sup> [47], the matrix indicators—"Awareness, knowledge, and skills", "Methodological approach", "Energy characteristics of leadership", "Organizational structure", and "Strategy and alignment"—at the first level are described by the values "fragmented" and "does not exist." The second level of the energy matrix of maturity is characterized by "Appointment of a person responsible for energy" and "Definition of policy and public information campaign".

According to the results of the study, it is obvious that Kazakhstan companies can be considered as being at the first (Initial) or, for larger organizations, possibly at the second (Occasional) level (see Table 8).

The governmen<sup>t</sup> of Kazakhstan also sets mutually exclusive tasks in the current period: frontal reduction of costs while replenishing the budget. It is obvious that for state executive bodies the need to replenish the budget will always be a priority and therefore motivates the fiscal authorities to impose finesand other penalties in every possible case, which places an additional burden, especially on small- and medium-sized businesses. "The Energy Conservation—2020" program, which sets ambitious goals and whose adoption in 2013 was widely reported in the media, was relatively quietly canceled already in 2016, which may indicate both the lack of development of the program itself and its objective inefficiency.

In these conditions, to increase the energy efficiency of national companies, there is a need to review the regulatory framework and enforce the abolition of all unsustainable penalties that impede business development, since they are not only a financial burden, but also a psychological barrier.



Source: [47].

Thus, the conclusions obtained in this article sugges<sup>t</sup> a further study of the resulting phenomenon, which is characterized by:

