The Neglected Focus on Managerial Aspect of Transfer Pricing Policy in Multidivisional Companies—Case of Serbia

Abstract

Prices applied to internal transactions between the business segments or divisions of a company in transactions between related entities within a group (transfer pricing) can have a significant impact on a company’s competitive advantage. Transfer pricing policy influences the profits of operating segments, resource allocation and the need for segment reporting. The two main approaches to transfer pricing are the tax and managerial approaches. The aim of this research was to test whether multidivisional companies operating in Serbia give more importance to the tax or the managerial aspect of transfer pricing policy. Another research aim was to determine whether segment reporting is more developed in companies in Serbia that have the legal obligation to prepare consolidated financial statements. Both research hypotheses were confirmed using the questionnaire method on a final sample of 52 large and medium-sized companies (out of 1912 large and medium-sized companies operating in Serbia). First, our findings show that tax compliance is more dominant in transfer pricing than the managerial perspective in the Serbian companies analyzed. Second, we found that mandatory consolidated financial reporting and related segment reporting can influence the managerial approach to transfer pricing in Serbian multidivisional companies and groups. Other factors (production orientation of companies, developed responsibility accounting and managers’ bonuses, for example) also encourage this approach.

1. Introduction

Traditionally, legislators, accountants, managers and researchers pay special attention to transfer pricing. This topic has attracted a lot of academic attention in the past, both in the economic and law-related literature.

So far, the debates have been focused on a wide range of different issues, such as regulation and transfer pricing strictness [1,2] and its political and socio-economic consequences [3,4,5]. Regulatory issues certainly have economic consequences, both at the macroeconomic level and for companies.

The possibility of manipulating transfer prices is on the government’s radar, so some research is aimed at assessing the effectiveness of transfer price regulation [6,7,8,9,10]. As a tool for profit-shifting [11,12,13], transfer pricing affects foreign direct investments(FDI) [14,15].

Profit shifting is closely related to the goal of tax avoidance and is a main motive in creating the transfer pricing policy of multinational companies (MNC). Undoubtedly, internal borrowing is an effective tool for achieving such objectives. These issues are also the subject of extensive research [1,16,17,18,19]. In modern business conditions, transfer prices are one of the most important determinants of the business ecosystem’s sustainability, especially in the supply chains of cross-border companies [20].

Transfer pricing issues also have a reporting dimension, both at the level of individual companies or groups and at the level of tax jurisdictions. Concerns from authorities about unlawful and unethical behavior by MNCs has recently resulted in the adoption of the so-called mandatory country-by-country reporting, which makes the issue of transfer pricing reporting even more important internationally [21,22,23,24].

From the management perspective, “transfer prices are a powerful tool for managing the performance of responsibility centers” [25] (p. 4). Growing management needs for accounting information on revenues, costs, inflows and outflows, as well as results of operating segments, have led to the emergence of responsibility accounting. This is a system of planning and measuring the performance of operating segments or areas of responsibility. The application of responsibility accounting requires decentralization of management activities and delegation of authority, i.e., identification of narrower areas of responsibility, for which reports will be prepared, which leads to the need for segment reporting. Transfer pricing through the applied method of transfer pricing influences the profits of operating segments, resource allocation and the need for segment reporting.

There are “two main approaches to transfer pricing: tax and managerial” [26] (p. 592). The aim of our research was to test whether multidivisional companies operating in Serbia give more importance to the tax or to the management aspect of transfer pricing policy. Considering that transfer prices are an element of income in the selling division, i.e., an element of expenses in the buying division, the chosen method of transfer pricing significantly determines the divisional performances, as well as their management’s decisions on internal or external procurement and sales. Another research aim was to determine whether segment reporting is more developed in companies in Serbia that have legal obligation to prepare consolidated financial statements. Multidivisional companies with a complex organizational structure, as well as production and geographically diversified activities, were observed. The impact of transfer prices on the company’s performances, as well as the dependency of bonuses on the divisional results, was also tested.

Two research hypotheses were set:

H1. 

In the Republic of Serbia, tax motives are more significant than managerial motives when deciding on transfer prices.

H2. 

The development of segment reporting, as one of the main premises for the managerial approach to transfer pricing, is largely conditioned by the legal obligation of companies to prepare consolidated financial statements.

There is a connection between these hypotheses. Namely, it is expectation that in addition to fulfilling legal obligations, management will be interested in transfer pricing policy when segment reporting, and thus, responsibility accounting has been developed. In complex economic entities, such as groups, the parent company, in addition to requiring the application of uniform accounting policies for all its subsidiaries, monitors their operations and performances through reporting segments or establishes reporting segments within the subsidiaries themselves. Therefore, the development of segment reporting is largely conditioned by the legal obligation of consolidated financial reporting, which should result in the mitigation or absence of neglect of the managerial approach to transfer pricing. Our research confirms this assumption.

By defining the main objective and the mentioned hypotheses, we consider the contributions of our study to be twofold.

First, academically, our research should fill a gap in the literature related to the topic we deal with from the perspective of companies operating in Serbia. Prior research conducted by Serbian scholars has mainly examined transfer pricing from the regulation perspective [27,28], transfer pricing policy by MNCs [29,30] or mostly theoretically [31,32]. However, to the best of our knowledge, a very few works on the topic we deal with were published in recent years [33].

Moreover, almost all practice-oriented research in Serbia is based on secondary data, i.e., financial statements, transfer price documentation, public databases, etc. On the other hand, the basis for our findings is primary data collected by the questionnaire method. Therefore, the methodological approach to researching the management aspect of transfer pricing in the Republic of Serbia could be considered as an additional contribution of this work.

Second, since our results show the neglect of the managerial approach to transfer pricing in Serbia, we are convinced that our results can be of interest of both the academic community and managers of divisional firms and/or groups. Managers should be encouraged to consider transfer pricing as a tool not only to meet the requirements of the tax authorities, but also to achieve better performances of responsibility centers, i.e., both ethically and in the accordance with regulation in force. These issues are not only important for already active companies, but also for start-ups in order to optimally prepare for market competition.

The Republic of Serbia was chosen as the research context due to the fact that it is a country in the process of economic transition. About that, for MNCs that are the bearers of foreign direct investments, through acquisitions of existing companies or the founding of new ones, transfer pricing policy is of great importance not only because of the fulfilment of national tax requirements, but also from a management perspective. We believe that the research results could be of interest to the academic and investment public, since the legislation of transfer prices in the Republic of Serbia is similar to the legislation in some other countries in the region. In the Republic of Serbia, there is an obligatory Rulebook on transfer prices that defines methods which should be applied according to the “arm’s length” principle, which determines the price of transactions between related entities. The transfer pricing method is chosen depending on the type of transaction. In some cases, it is possible to use a combination of several methods. Each chosen method must be practically applicable, and the final effect must have a reasonable assessment of the results in accordance with the “arm’s length” principle. In the domestic legislation, related parties are:

• Persons with a direct or indirect participation in capital above 25% (parent companies, sister companies and subsidiary companies);
• Persons who have more than 25% of the votes in the management structures and who have significant influence on the business decisions;
• Family members.

A transfer pricing study is obligatory if (a) the value of annual transactions (excluding loans and interest) with at least one related entity exceeds the amount of EUR 68,000 or (b) the company places or receives loans to/from at least one related entity (except in the case that it has received an interest-free loan). In addition, companies that have transactions with persons and entities from tax havens are also obliged to prepare a transfer pricing study in order to eliminate the possibility for tax evasion.

In this respect, the research results in this paper could be relevant in other countries with similar experience in terms of transfer pricing practice and they could serve as a basis for conducting further research.

The paper is structured as follows: after the introduction, a review of relevant literature is given. After that, the research methodology is presented, as well as the results and discussion. Relevant conclusions of the research are given at the end of the paper.

2. Theoretical Background

2.1. The Role and Aspects of Transfer Pricing

“Transfer pricing is not just an accounting technique, but also a method of resource allocation and avoidance of taxes” [34] (p. 342). Transfer prices are internal prices at which one business segment delivers its products and services to another business segment. Li and Ferrara [35] state that the emergence of transfer pricing is related to hierarchical multidivisional companies, which, with the increase in their size, had to organize production in different divisions. This resulted in the breakdown of knowledge and activities into several specialized business units. Transfer pricing is an important aspect of the relationship between business segments in decentralized organizations, with it specifically determining the profits of vertically integrated divisions [36]. Transfer prices represent an important tool for managing the performance of complex multidivisional enterprises, providing internal signals that direct the allocation of resources and profit within the company [37]. They affect not only the distribution of total profits between divisions and profit centers, but the company’s transfer pricing policy also largely determines purchasing, production and investment decisions. As transfer prices affect the performance of business segments, they consequently affect the way corporate resources are allocated between segments, given that the allocation is mainly based on the achieved performance. “The principles of decentralization suggest that companies should grant managers autonomy to set transfer prices and to decide whether to sell internally or externally.” [38] (p. 513). Therefore, the first important aspect of transfer pricing is the management aspect, as it serves as a tool for managing the performance of decentralized business segments through the use of cost accounting information.

From a management perspective, transfer pricing policy may be motivated by strategic goals of achieving competitive advantage for the company or its segment or affiliation. Namely, transfer pricing policy can be aimed at reducing the costs of the company’s segment or affiliation in order to achieve a competitive advantage. This objective will be more pronounced in MNCs. In the case of MNCs, the goal of international transfer pricing policy is to “minimize taxes, duties and foreign exchange risk, together with increasing the company’s competitiveness” [39] (p. 59). Although transfer pricing is primarily in service to profit shifting, this policy also has an impact on the competitiveness of foreign affiliates [40]. Generally, in research on a sample of Taiwanese MNCs, Lin and Chan [41] found that the focus of transfer pricing policy has moved to enhancing the competitiveness of the companies analyzed.

Transfer pricing policy has a specific role in state-owned companies. Perera et al. [42] examine “the diffusion of transfer pricing as an innovation in a government trading enterprise (GTE) as it moved from protected monopolistic status to commercialization” (p. 140). Their findings indicate the importance of creating a new organizational culture in this type of company, especially through the delegation of transfer pricing policy to certain levels of management as a motive for their interest in the company’s competitiveness in the open market.

Finally, transfer pricing can also be seen as an indicator of a company’s market power. Alles and Datar showed in their study that “the amount by which firms mark up their transfer prices over marginal costs is direct a function of their market power” [43] (p. 452) and is an indicator of the current competitive advantage of a company.

In addition, companies that publish consolidated financial statements have an additional obligation to form profit centers, with the application of segment reporting and segment performance monitoring. Accordingly, operating segments are business activities that may generate revenues and expenses, whose operating results are under control of top management and “for which discrete financial information is available” [44] (par. 5). International Financial Reporting Standard 8—Operating Segments (hereafter IFRS 8)—gives priority to consolidated financial statements in relation to the financial statements of the parent company for the purposes of segment reporting when both sets of financial statements are prepared [44] (par. 4). Therefore, there is an obligation to complete segment reporting as the basis of the managerial approach to transfer pricing, in particular for companies that prepare and disclose consolidated financial statements. In the Republic of Serbia, according to the Accounting Law, large and medium-sized legal entities and listed companies are obliged to prepare and disclose consolidated financial statements, except in certain cases (e.g., when the group is small in size or the parent company is at the same time a subsidiary of another parent company registered in the Republic of Serbia). The International Financial Reporting Standards are to be applied for the preparation of these reports. In this sense, our tests for Hypothesis 2 relate to large and medium-sized companies that are required to prepare consolidated financial statements.

The need for segment reporting stems from the fact that modern companies diversify their business both geographically and by products and services, positioning themselves in the markets of different countries with very disparate assortments. These markets as well as their products and services are very different according to the degree of profitability, risk exposure, growth opportunities, etc. Therefore, business decisions on the allocation of resources to certain products, services and markets must be based not only on the information summarized in the financial statements of the company as a whole but also on information at the level of business segments [45]. IFRS 8 implies the adoption of a management approach when presenting the financial performance of business segments, which means that users of financial statements gain insight into the business of segments through the “eyes of management”. The transfer pricing system is an integral part of the management control process in MNCs regardless of the level of internal trade [46]. In that sense, the transfer pricing method applied affects the profit of the operating segments and the allocation of resources.

However, the transfer pricing approach for management accounting purposes is only the first aspect of using transfer prices. Another important aspect of transfer pricing that has recently become increasingly important is the tax aspect. It can be said that there are “very significant differences between the tax and management aspects of the use of transfer pricing” [25] (p. 10). The tax aspect of transfer pricing is growing in importance due to the internationalization of business and the increase in the number of MNCs, which is accompanied by growing regulatory and compliance requirements. In the case of MNCs, their segments or subsidiaries are located in other countries, so transfer prices in such circumstances have an international character and significant tax implications. A parent company and its subsidiaries use transfer prices for intercompany transactions within a holding corporation, but also as an instrument for tax optimization. With transfer prices that differ from market prices, it is possible to transfer the profits of subsidiaries within the holding company “from high-tax countries to low-tax countries” [47] (p. 41). The tax implications of transfer pricing in MNCs that typically disclose consolidated financial statements are subject to extensive regulation by the OECD Guidelines for Multinational Enterprises [48].

2.2. The Review of Transfer Pricing Models

It is obvious that transfer pricing models differ depending on whether they relate to transactions between segments of a company operating within a national boundary or to international transactions between segments or subsidiaries of MNCs. This article is focused on the management aspect of transfer prices and the ways in which they are determined within national divisionally structured companies. The issue of determining transfer prices is closely related to the distribution of profits between divisions and profit optimization at the company level. Transfer prices affect the level of revenue of the selling division and the level of costs of the buying division, so they significantly affect the performance of business segments. If the company’s policy requires that sales between divisions should be made at a discount, the divisions that buy will pay less for internally delivered products and services than if they buy externally on the market. The basic question that arises is why would a sales division manager make investments and sales when the benefits go to other divisions within the company? In this regard, another important question is what are the internal prices that would simultaneously reconcile the conflicting interests of the divisions while maximizing the company’s profits? In such a situation, the question arises as to the ethical behavior of managers, but also as to whether their actions comply with the rules [49,50]. Bearing in mind the fact that segment managers can act against the interests of the company with their transfer pricing decisions, harmonizing the interests of the segments and the company as a whole is directly conditioned by the adequate choice of transfer pricing policy [51,52]. The chosen model of transfer pricing should not lead to a situation where one segment achieves better results at the expense of other segments of the company, but the goal is to maximize the performance of the company as a whole.

In practice, commonly used transfer pricing approaches are “market-based, cost-based and negotiated” [53]. Although all three approaches are often used by management, Schuster states that “most frequently used are cost-based transfer prices, followed by the market-based transfer prices” [54] (p. 5). Each of the mentioned methods has its advantages and limitations, as well as some justified areas of application that are determined by specific circumstances.

Market-based transfer prices are objective prices by which “goods would be sold on the market if selling division were independent company” [33] (p. 50). Market prices are the objective consequence of market conditions, and their use as transfer prices encourages business segments to interact with third-party buyers and sellers. In such circumstances, the purchasing division will buy internally as long as purchase prices are below or equal to market prices. In practice, “market-based transfer prices may be based on different external prices, such as the price at which the company’s upstream division sells to external customers or the price of a competitor” [55] (p. 1868) in the market. In that sense, although at first glance it is more logical and justified that the transfer price is based on the external price of the sales division, a significant percentage of companies in practice use the competitor’s price, while sometimes both prices are used in combination. Market-based transfer prices clearly favor greater segment autonomy, and their application is also supported by IFRS 8. In conditions of highly competitive markets, market-based transfer prices enable the harmonization of the interests of segments and the entity as a whole, increase management motivation, encourage a greater degree of segment autonomy and result in more adequate evaluations of business segments [56]. The more active and efficient the markets for specific products are, the more suitable market prices are as a basis for transfer pricing.

However, although the use of market-based transfer prices seems to be the superior and optimum transfer pricing method, it often happens in practice that it is not possible to determine market prices for specific products and services transferred from one segment to another. These restrictions are not only related to the market’s inefficiency (which is characteristic of certain industries), but also to the specific characteristics of the products themselves. For example, it is not easy to determine the market prices of semi-finished products that are delivered to another division for further processing. In addition, Raiborn et al. [57] point out other shortcomings of market transfer prices. One of them is that in an effort to maximize profits, divisional managers may act contrary to the company’s goals. For example, purchasing division managers can continuously seek cheaper external sources of supply, avoiding internal transactions, while ignoring the fact that internal transactions between divisions open space for savings in delivery, packaging, advertising or insurance (that exist in external transactions).

In conditions of inefficient and inactive markets for certain products, it is advisable to use a cost-based approach when determining transfer prices, where full or variable costs can be used. Cost-based transfer prices imply a reliance on cost accounting data, and they are based on the production costs of the selling department. This transfer pricing model is characterized by simplicity and objectivity, which is why it is widespread in the practice of companies. Cost-based transfer prices can be established at different levels of cost coverage (full and variable costs) and using actual or planned (standard) costs. Transfer prices based on full costs are more frequently used than those based on marginal costs because the sales division is able to recoup total production costs and are less likely to incur a loss. In addition, full cost-based transfer prices are favored because the internal production of intermediate products is usually organized with a long-term perspective, so the use of variable costs is inadequate as a criterion for long-term decisions. Brickley et al. [58] point out that full costs-based transfer prices rely on objective and hard-to change rules, and they can result in greater value for the company than transfer prices under the discretion of divisional managers. In addition, the same authors emphasize that the use of full cost-based transfer prices is justified in situations where the factory reaches full capacity because full costs in that case become the best approximation of opportunity costs. Banker and Hansen [59] similarly argue that as long as the supplying division is producing at capacity, full costs are an appropriate basis for transfer prices.

However, full production costs containing fixed costs are subject to variations due to changes in capacity utilization and arbitrary allocation of production overheads [60]. Friis [61] states that incentives to use cost-based transfer prices are inherently subject to corruption within the company for two reasons. First, top management controls the performance measurement of divisions, and therefore, it often expropriates segment profits. In addition, “transfer prices are incomplete and prone to manipulation by managers in selling divisions” [61] (p. 138), who are motivated to increase their prices in order to improve divisional performance. Therefore, one of the key issues when using cost-based transfer prices is how to design competition within the company to alleviate these problems. The use of full costs also allows the selling division to redirect all its production inefficiencies and excess capacity costs to the buying division. It means that in such circumstances, the sales division does not have much incentive to increase its efficiency [62]. Consequently, when the producing division has excess capacity, the purchasing division will buy too few units internally. In the case of using full cost-based transfer prices, the selling division does not make a profit, but the company profit is derived from the profit of the purchasing division. Considering that a price that only covers production costs is not acceptable for the sales division, in practice, a modified cost-plus version is often used where the transfer price includes the profit of the sales division as an incentive to intra-company transactions.

Negotiated transfer prices are the result of negotiations between managers of selling and purchasing divisions. Such an agreed upon determination of transfer prices implies a high degree of autonomy of the divisional managers. Negotiated transfer prices should be mutually acceptable and reflect the fair value of the transferred products to the greatest extent possible. The negotiating framework is limited to answering the question of what is the lowest selling price acceptable to the sales division and what is the highest selling price acceptable to the buying division. Negotiated transfer prices are usually used in conditions when it is not possible to use the market or cost transfer pricing model. They are usually lower than current market prices but also higher than the total costs of the selling division. It is important to mention that negotiated transfer prices are often determined by the bargaining power of divisional managers. For example, this happens in situations where the buying division has a lot of options in the field of supply while the sales division does not have many sales options. In this regard, Hilton et al. [63] state that any division should not make higher profits just because its manager has superior negotiating skills or bargaining power compared to other divisional managers. Vaysman demonstrates “that the firm can design managerial-compensation schemes and bargaining infrastructures so that the negotiated transfer-pricing structure allows it to reach the upper bound on available profits” [64] (p. 349). The disadvantages of negotiated transfer prices are also reflected in the fact that their determination at a mutually acceptable level requires a lot of time and effort from divisional managers. This often results in interpersonal conflicts and dysfunctional behavior within the organization, so in such situations, resolving conflicts between divisions may require top management to mediate, which reduces divisional autonomy [65]. However, although negotiated transfer prices are not optimal, the motivational aspect of giving managers autonomy to decide on the prices of their outputs or inputs can bring more benefits than harm to the company as a whole.

3. Materials and Methods

Research methodology in this paper is based on the questionnaire method and thus uses primary data. Most of the previous research on transfer pricing used financial statements data and used only secondary data. The questionnaire was sent to 200 companies who operate in Serbia and whose activities are diversified. There are sampled companies with complex organizational structures, often consisting of many profit and investment centers that exchange goods and services with each other and use transfer pricing for their valuation. We assume that micro and small enterprises are not diversified enough, so we focus our research on medium and large companies. Th final number of companies in the sample whose answers were processed is 52, implying a response rate of 26%. The questionnaire process was conducted in January and February 2021. We believe that our sample is sufficiently representative, since the total number of large and medium-sized entities operating in the Republic of Serbia in 2021 is 1912 [66]. In addition, it is known in the literature [67] that companies, particularly in transition countries, are often not ready to take part in questionnaires about their tax positions and tax strategies, and this was a significant obstacle in obtaining a higher number of respondents. This means that the initial sample was 10.46% and that the final sample represents 2.72% of the total population of large and medium-sized entities registered in the Republic Serbia. The questionnaires were filled out by the persons responsible for setting the transfer prices in the sampled companies. Depending on the distribution of tasks, those are the heads of finance, accounting or controlling. Therefore, a high reliability both of the answers received and the results of our research is to be expected. The structure of the sample is presented in

Table 1. The sample structure.

Main Industry	Number of Companies
Production	22
Non-production	29
No response	1
Total	52
Required to prepare consolidated financial statements	Number of Companies
Yes	17
No	35
Total	52
Required to submit transfer pricing study	Number of Companies
Yes	50
No	2
Total	52

.

As Brune et al. [68] point out, the main industry to which the company belongs may be important determinant of tax-motivated transfer pricing, since production-oriented companies are more likely (primarily due to their large amounts of intangible assets) to shift their profits to tax havens. In this regard, the sample is balanced as the number of production-oriented and other companies is similar.

On the other hand, more than half of the sampled companies are not required to prepare consolidated financial statements. However, the vast majority of the sampled companies is required to submit transfer pricing studies along with their submitted tax balance. Such data indicate that a majority of the sampled companies are subsidiaries, rather than parent entities, of their economic groups.

The questionnaire consists of two broad parts, with the first part presenting basic firm characteristics and the second part presenting the transfer pricing and segment reporting environment of the company. Data were statistically processed in the statistical software SPSS—Statistical Package for the Social Sciences.

In addition to descriptive statistics, numerous statistical tests were used to determine the significance of the obtained results. To test the first hypothesis, we have firstly used tests to determine whether the distributions of the employed variables follow a normal distribution. In this regard, parametric tests for independent samples were used if the assumption of the normally distributed variables was satisfied, while nonparametric tests were used otherwise. To test the second hypothesis, we used statistical tests appropriate for categorical variables. Statistical tests, primarily nonparametric ones, are particularly useful for statistical analysis when the number of observations in a sample is relatively small, as they might not be the appropriate methodology for large-sized samples [69].

4. Results and Discussion

4.1. Development of Transfer Pricing Environment

The first research hypothesis is tested considering the relative significance of tax and managerial aspects when deciding on transfer prices. Before hypothesis testing, it is also important to analyze the transfer pricing environment in which the sampled companies operate. Results of such an analysis are presented in

Table 2. Transfer pricing environment of sampled companies.

There is Master File for Transfer Pricing in the Economic Group to Which the Company Belongs	Number of Companies
Yes	13
Is developing	3
No	33
No response	3
Total	52
There is local file for transfer pricing in the company	Number of Companies
Yes	8
Is developing	6
No	34
No response	4
Total	52
Sector that defines transfer prices of the company ᶺ	Number of Companies
Accounting	28
Controlling	4
Finance	8
Management	18
Other	3
Transfer pricing method used ᶺ	Number of Companies
Comparable uncontrolled price	43
Cost-plus	12
Profit split	0
Resale price	8
Transactional net margin	6

Note: ᶺ denotes questions where more than one answer is permitted.

.

Data from Table 2 show that most of the companies rely only on the macro-economic regulation of transfer pricing, as only 25% of the respondents answered that the master file for transfer pricing exists in their economic groups, while only 15% of the respondents answered positively on the existence of a local file for transfer pricing.

Transfer prices are dominantly defined in accounting sectors of sampled companies. As many as 22 respondents answered that transfer prices are defined only in the accounting sector, while 15 respondents answered that the transfer prices are defined only by management. On the other hand, a comparable uncontrolled price is the most widely used method for determining transfer prices. As many as 28 respondents answered that the comparable uncontrolled price is the only method used in their companies.

Respondents from the companies had to answer two questions on the Likert scale, with the first question being related to the significance of the tax aspect and the second to the significance of the managerial aspect. In total, 51 respondents answered both questions.

Table 3. Significance of tax and managerial aspect when deciding on transfer prices.

Panel A. Descriptive statistics
	Mean	Median	Min.	Max.	St. Dev.
The method for the transfer prices calculation is chosen based on the tax motives	3.20	3.00	1.00	5.00	1.28
The method for the transfer prices calculation is chosen based on the managerial motives	2.78	3.00	1.00	5.00	1.24
Tax/Managerial Score	1.27	1.00	0.50	3.00	0.57
Panel B. Mann–Whitney test results
Mann–Whitney U	943.500
Z-statistic	−3.322
p-value	*** 0.001

Note: the results are statistically significant at the 10%, 5% and 1% (***) levels.

presents descriptive statistics for these questions and the results of the Mann–Whitney test on the difference between the significance of tax and managerial aspects. However, we firstly used the Shapiro–Wilk test (instead of the Kolmogorov–Smirnov test due to the small sample size) to show that the distributions of the obtained answers do not follow a normal distribution (p-values 0.000). In such circumstances, the nonparametric Mann–Whitney test is a better alternative than a parametric Independent Samples t-test.

In order to test the relative significance of tax and managerial aspects, we had to develop a unique measure, called the Tax/Managerial Score. This measure is obtained as a ratio between the answer to the significance of the tax aspect (first question in Table 3) and the answer to the significance of managerial aspect (second question in Table 3). For instance, if the respondent ranks the significance of the tax aspect as a 5 on the Likert scale and the significance of the managerial aspect as a 2, the Tax/Managerial Score is 2.5. In this regard, when the Tax Score is higher than the Managerial Score, this implies that the tax aspect is more significant than the managerial aspect and vice versa. Since the referent value for the Tax/Managerial Score is one, the Mann–Whitney test is conducted comparing the Tax/Managerial Score and its referent value.

Data from Table 2 once again confirm that tax motives are more significant when deciding on transfer prices than the managerial motives. First, the answers for the significance of the tax aspect produced a higher arithmetic mean than the answers for the significance of the managerial aspect. Second, the Mann–Whitney test showed that the Tax/Managerial Score is statistically significantly higher than one, with the results significant at the 1% level. Therefore, the first research hypothesis cannot be rejected.

Furthermore, the significance of the tax aspect when deciding on the transfer pricing is confirmed by the multiple-choice question, as respondents had to choose one or more purpose(s) of the transfer pricing optimization. There were seven purposes available, and the results for this question are presented in

Table 4. The purpose of transfer pricing optimization.

Purpose	Number of Companies
To optimize profitability of segments	14
To optimize taxes within the tax law	35
To optimize cash flow	5
To motivate divisional management	2
To calculate and allocate costs	5
To help the profit center or segment decision making	5
To align targets of segment and the whole company	6

Note: more than one answer is permitted.

. It is interesting to point out that only 44 respondents answered this question and as many as 35 respondents answered that optimization of the taxes within the tax law is one of the purposes. In addition, as many as 21 respondents have chosen only this answer.

Our research confirms the prior finding of Kundelis et al. [70] about the significant importance of the tax-motivated transfer pricing for companies in transition and post-transition countries. In fact, there are many reasons that may explain the higher significance of tax motives than managerial motives when deciding on the transfer prices. Our findings are also in line with those of Sebele-Mpofu et al. [71]. Due to the internationalization and globalization of business, the opportunities to shift profits to tax haven are more available than ever. The complex network of intragroup transactions with the related entities headquartered in tax havens enable important tax savings through tax planning and avoidance strategies [72,73], which, in turn, affects the ROE or the current ratio indicators [74] and other metrics of company performance.

On the other hand, it should be noted that statutory income tax rate in Serbia is only 15% and that vast majority of companies pay the corporate tax at the effective rate significantly lower than the statutory. However, tax morale in Serbia, like in other transition and post-transition countries, is relatively low [75], while national tax authorities do not have enough resources to effectively combat profit shifting. It is rational to assume that such an environment fosters tax avoidance [27]. In addition, there are many MNCs that have established the subsidiaries in Serbia, with MNCs having bargaining power over the national tax authorities. As a result, subsidiaries of MNCs may benefit from tax savings through intragroup transactions.

However, the presented results should be interpreted carefully, as they do not strictly imply that companies use transfer prices to shift their profits to the tax havens. A growing fraction of the literature in the low-tax transition and post-transition countries [30,76], such as Serbia, suggest that MNCs may use such countries as substitutes for traditional tax havens. Therefore, it is possible that MNCs shift their profits from high-tax countries to low-tax transition countries in order to both avoid taxes and negative public reaction for using traditional tax havens.

We have also conducted some additional analysis. First, following prior research, we have examined whether production-oriented companies have higher Tax/Managerial Scores than other companies. Second, we have examined whether companies with implemented responsibility accounting have lower Tax/Managerial Scores. Responsibility accounting is treated as a system of planning and measuring the performance of segments or profit centers and their managers. Third, we have examined whether the companies that pay variable management bonuses have lower Tax/Managerial Scores. We believe that the existence of responsibility accounting and variable management bonuses require proper cost determination and allocation and discourage tax avoidance motives within a company. The results of these analyses are presented in

Table 5. Results of the additional analyses.

Panel A. Production-oriented vs. other companies
Descriptive Statistics
	Mean	Median	Min.	Max.	St. Dev.
Production-oriented companies	1.49	1.13	0.50	3.00	0.76
Other companies	1.10	1.00	0.60	2.00	0.29
Mann–Whitney test results
Mann–Whitney U	225.500
Z-statistic	−1.777
p-value	* 0.076
Panel B. The role of responsibility accounting
Descriptive Statistics
	Mean	Median	Min.	Max.	St. Dev.
Companies with responsibility accounting	1.17	1.00	0.50	2.50	0.44
Other companies	1.36	1.00	1.00	3.00	0.69
Mann–Whitney test results
Mann–Whitney U	196.000
Z-statistic	−1.116
p-value	0.264
Panel C. The role of variable management bonuses
Descriptive Statistics
	Mean	Median	Min.	Max.	St. Dev.
Companies with variable management bonuses	1.20	1.00	0.50	2.00	0.47
Other companies	1.39	1.00	0.67	3.00	0.72
Mann–Whitney test results
Mann–Whitney U	80.500
Z-statistic	−0.717
p-value	0.473

Note: the results are statistically significant at the 10% (*), 5% and 1% levels.

.

The data from Table 5 show that the relative significance of tax and managerial aspects when deciding on transfer prices is significantly different in production-oriented and other companies. In this regard, production-oriented companies have higher both arithmetic mean and median Tax/Managerial Scores than other companies. The difference is significant at the 10% level. Such a finding is in line with the prior research of Huizinga and Laeven [77] and Brune et al. [68], as production-oriented companies have more intragroup transactions (for instance, intragroup raw material trading) and more intangible assets that may underlie the profit shifting to the low-tax jurisdictions.

In line with our assumptions, companies with implemented responsibility accounting systems and companies that pay variable management bonuses have lower Tax/Managerial Scores than other companies, implying that these companies pay more attention to the managerial aspect when deciding on transfer prices. However, according to the Mann–Whitney tests, the differences tested in these situations are not statistically significant. It is also worth noting that 31 companies had implemented responsibility accounting, 16 companies did not, while 5 respondents did not give any answer. On the other hand, 10 companies paid variable management bonuses, 19 companies did not, while 23 respondents did not give a valid answer.

4.2. Development of Segment Reporting

The second research hypothesis is tested considering the obligation of the company to prepare consolidated financial statements and the existence of segment reporting in the company. Before hypothesis testing, it is also important to analyze the segment reporting environment in which sampled companies operate. The results of such analysis are presented in

Table 6. Segment reporting environment of sampled companies.

There Are Developed Profit Centers or Segments to Track Their Performance	Number of Companies
Yes	25
No	26
No response	1
Total	52
Number of profit centers or segments developed in the company	Number of Companies
From two to four	12
Four five to seven	1
More than seven	11
Does not have profit centers or segments	26
No response	2
Total	52
Most important indicators calculated for profit centers or segments ᶺ	Number of Companies
Balance sheet elements	18
EBIT or EBITDA	12
Income statement elements	35
Return on investments	2
Other	2

Note: ᶺ denotes questions where more than one answer is permitted.

.

The data from Table 6 show that the share of companies with and without established segment reporting procedures is nearly similar. On the other hand, the number of developed segments in the company varies a lot, since one group of companies has only a few segments, while another group of companies has a large number of segments. It is also interesting to note that 42 respondents answered the question on the most important indicators calculated for profit centers or segments and as many as 16 respondents answered that income statement elements are the only group of indicators employed.

Regarding statistical tests, we first examined whether the share of companies that developed segment reporting is higher in companies that are obliged to prepare consolidated financial statements. Since we have two categorical (yes/no) variables, we have opted for a Chi-Square Test for Independence. The results of this test are presented in

Table 7. The relation between obligation to prepare consolidated financial statements and the development of segment reporting.

Panel A. Matrix
		Developed Profit Centers or Segments
		Yes	No
Obligation to prepare consolidated financial statements	Yes	14	2
	No	11	24
Panel B. Chi-Square test results
	Value	df	p-value
Pearson Chi-Square	13.814	1	*** 0.000
Continuity Correction	11.662	1	*** 0.001

Note: the results are statistically significant at the 10%, 5% and 1% (***) levels.

. No one cell has the expected count lower than 5, so the assumption for the Chi-Square test is satisfied. In addition, we have also used Yates’ Correction for Continuity since we have only two categories per variable.

Since one respondent did not give the answer on the development of the segment reporting, the Chi-Square test is conducted with 51 observations. In general, the results (both basic results and results including Yates’ correction) show that the share of companies that developed segment reporting is significantly higher in companies that are obliged to prepare consolidated financial statements. The difference is significant at the 1% level. Therefore, it may be concluded that the development of segment reporting is largely conditioned by the legal obligation of companies to prepare consolidated financial statements, so the second research hypothesis cannot be rejected.

In addition, the matrix from Table 7 may explain the results of the statistical test, since 14 of 16 (87.5%) companies that are obliged to prepare consolidated financial statements have developed profit centers or segments. On the other hand, only 11 of 35 (31.4%) companies that are not obliged to prepare consolidated statements have developed profit centers or segments.

To further test the relation between the obligation to prepare consolidated financial statements and the segment reporting, we have examined whether the share of companies that implemented responsibility accounting is higher in companies that are obliged to prepare consolidated financial statements. In fact, the implementation of responsibility accounting is directly related to the segment reporting development. The results of the Chi-Square statistical test for this relation are presented in

Table 8. The relation between obligation to prepare consolidated financial statements and responsibility accounting.

Panel A. Matrix
		Responsibility accounting implemented
		Yes	No
Obligation to prepare consolidated financial statements	Yes	13	2
	No	19	14
Panel B. Chi-Square test results
	Value	df	p-value
Pearson Chi-Square	3.927	1	** 0.048
Continuity Correction	2.727	1	* 0.099

Note: the results are statistically significant at the 10% (*), 5% (**) and 1% levels.

. As no one cell has the expected count lower than 5, the assumption for the Chi-Square test is satisfied.

Since four respondents did not give an answer on the implementation of responsibility accounting, the Chi-Square test is conducted with 48 observations. The test results showed that the share of companies that implemented responsibility accounting is significantly higher (based both on basic results and results including Yates’ correction) in companies that are obliged to prepare consolidated financial statements, thus further confirming the assumption presented in the second hypothesis. The matrix from the Table 8 shows that 13 out of the 15 (86.7%) companies that are obliged to prepare consolidated financial statements have implemented a responsibility accounting system. On the other hand, only 19 of the 33 (57.8%) companies that are not obliged to prepare consolidated financial statements have implemented a responsibility accounting system.

The presented results indicate that segment reporting in the Republic of Serbia is still not sufficiently developed as the implementation of consolidated financial statements is a result of the company’s legal obligations rather than the desire of the company to analyze performance of the segments. This finding is consistent with the prior research on companies in Serbia [78], stating that there is a significant room for improvement in segment reporting.

5. Conclusions

Transfer pricing policy has multidimensional implications, both at the level of public finances and at the level of companies. From the perspective of MNCs and groups, the impact of this policy is seen in many operational and financial performances and decisions (capital budgeting, allocation of profits and resources in general), including the achievement of competitive advantages, as well as on non-financial performances (management motivation, for example). Although transfer pricing policy is generally preferable for tax purposes, a company’s management should not neglect the managerial aspect of transfer pricing. However, the findings of our research show that compliance with the tax rules is more dominant in transfer pricing than the managerial perspective in the Serbian companies analyzed. This finding confirms our Hypothesis 1.

We have further examined this general finding in terms of what factors can influence the reduction in neglecting the managerial approach to transfer pricing. Additional research shows that companies with an implemented responsibility accounting system and companies that pay variable managerial bonuses pay more attention to the managerial approach in transfer pricing decision making. Responsibility accounting is the information basis for segment reporting. Therefore, a developed system of segment reporting is expected to help reduce the neglect of the managerial approach in transfer pricing.

Bearing that in mind, to test Hypothesis 2, we examined whether the obligation to prepare and disclose consolidated financial statements and, thus, whether mandatory segment reporting on a consolidated basis (in accordance with IFRS 8) creates a starting point for a greater attention to the managerial approach to transfer pricing. In line with our assumptions, the majority of the sampled companies that are obliged to prepare and disclose consolidated financial statements have implemented responsibility accounting and have developed segment reporting. In other words, our findings imply that these companies pay more attention to the managerial aspect when deciding on transfer prices. On the other hand, it can be concluded that neglecting a managerial focus in transfer pricing is very pronounced in companies that are not subject to group reporting and thus have not developed segment reporting systems. The aforementioned findings confirm Hypothesis 2.

The importance of our research lies in its implications, i.e., in the contribution we provide to the literature and Serbian managers on transfer pricing in the Republic of Serbia.

First, in the absence of similar research, by collecting data using a questionnaire, we examined the prevalence of the managerial perspective on transfer pricing in Serbian multidivisional companies and groups. Although transfer pricing can be used as a tool by management to achieve the targeted performance of responsibility centers using cost accounting information, this tool is still not widely used in multidivisional companies operating in Serbia. Transfer prices are mainly in focus only for taxes purposes and due to the legal obligation of companies to prepare transfer pricing reports for all transactions between related entities. One of the key reasons for this is the underdevelopment of responsibility accounting and segment reporting as a base for evaluating the performances of divisions and their management.

Second, we point out that there is a need for segment reporting not only on a consolidated basis, but also at the level of individual entities. Our findings show that in case of mandatory consolidated financial reporting, Serbian parent companies establish segment reporting. On the other hand, only one-third of the sampled companies that do not prepare consolidated financial statements have developed reporting segments as the basis for a managerial approach to transfer pricing at the level of an individual company. At the same time, this approach is more prevalent in production-oriented companies, which is in line with other research. In addition to the necessity of the implementation of responsibility accounting, our research shows that the managerial aspect of transfer pricing is also conditioned by the payment of variable managerial bonuses.

Based on the above-mentioned results and contributions, we believe that our research has both theoretical and practical implications.

The theoretical foundations presented in this article are not undisputed. The paper promotes the need to pay more attention to the management approach in creating the company’s transfer pricing policy. A relatively small sample, limited to companies operating in the Republic of Serbia, and gaps in previous research in the analyzed area mean that the results of the research may not be generalizable. Nevertheless, this framework should represent the potential for further research into the managerial approach to transfer pricing and the possibility of its development in multidivisional and multinational companies from Serbia and the surrounding region.

This paper, although to a certain extent theoretical, corresponds to the needs of practice. The results of our research also provide useful implications for owners and managers of multidivisional companies operating in the Republic of Serbia, raising awareness of the need to affirm the managerial approach to transfer pricing. Consequently, an increased focus on the managerial approach in transfer pricing policy should improve the competitiveness of Serbian multidivisional companies and groups, both at the national and international level.

Our research also has several limitations that provide opportunities for future research.

First, in order to consider our findings generally applicable to all large and medium-sized companies operating in Serbia, a larger number of respondents should be included in the research. Our targeted sample of 200 large and medium-sized companies could provide a satisfactory level of generalization since it would reflect 10.46% of the total population of companies of that size in the Republic of Serbia. With a final sample of 52 large and medium-sized entities (2.72%), the acceptability of the generalization of the findings is at a lower level.

Therefore, further research should include a larger number of companies, while taking into account the peculiarities that we highlight below.

Namely, although our sample consists of companies from different industries, we interpreted the results from the perspective of production-oriented companies and other companies. We believe that a more detailed analysis of a specific industry would yield significant results.

Furthermore, a separate analysis of listed companies would be helpful in drawing conclusions about the issues we are examining. As entities of significant public interest, listed companies are expected to have more developed internal control mechanisms, a developed responsibility accounting system and, therefore, a management approach to transfer pricing.

Finally, the limitation of our research is its focus on companies operating only in Serbia. Future research could also include cross-country comparisons, primarily with countries with similar business cultures and legislative frameworks.