With the increase in the volume of international exchange and the globalization of multinational companies, problems of transfer pricing control in transactions between affiliates and their alignment with the "arm's length" pricing principle are growing. A proper calculation of transfer pricing affects the amount of taxable income and provides a basis for checking whether in this way profit is "extracted" to other tax jurisdictions. The international transfer pricing guidelines have recommended several methods for calculating transfer prices, depending on the circumstances of the case, including several recommended parameters relevant to the calculation. Also, if the Cost Plus Margin method and Transactional Net Margin (TNM) method can be applied in an equally reliable manner, the recommendation is Cost Plus Margin. Many countries have accepted the international guidelines and incorporated them into their national regulations. However, in the absence of a serious analysis of the effects in practice, the possibility of deviations in the calculation of results was also accepted. If these results are ultimately the amount of tax paid or not paid in a country, then it is understandable why these deviations deserve special attention. The main hypothesis in this paper is that when both methods are applicable, the Cost Plus Margin method should not always be favored, as effects that should be prevented may occur. Due to the circumstances of the case, a proper calculation can be very debatable if certain parameters change a minimum. The paper discusses deviations of the transfer price from the "arm's length" principle when both methods are equally acceptable, and with minimum changes to two parameters: the size of the sample of comparable companies and the observation period. All this has been tested on the example of a multinational company from one of the developed countries in Europe, which has a related entity in one of Europe's developing countries. The tested party is a related entity as a developing country taxpayer. In the implication generally there are many problems and dilemmas that need to be overcome in the relationships between the party calculating the transfer prices, the companies themselves, and the tax administrations. It is very significant from which aspect the calculation is managed. If the aspect of the company is favored, then the best option is the one where the company will pay the lowest tax. However, if the tax administration is already authorized to request a change in the calculation, then it will work in the interest of the state, and the most unfavorable option for the corporation may be obtained by changing some of the parameters used. Both would be formally acceptable, but the tax amount would not be the same. Possible scenarios in the test case showed that internationally recognized recommendations regarding method selection have weaknesses because they do not include all the parameters of the methodology, which significantly affects the end result. In a situation where the observation timeframe is flexible and no minimum sample size of comparable companies is defined, there is practically space to manipulate the final result. In order to avoid ambiguity and various forms of manipulation in the transfer pricing calculation, it is necessary to further specify international guidelines for the development of transfer pricing in the domain of time and size of the sample. Additionally, the taxpayer's obligation to apply transfer price calculations should be introduced for all methods that respond to the circumstances of their case, and not just one of them. This generally can be overcome either by changing the guidelines or by defining rules in greater detail at the level of the depends on multiple parameters, both those defined in theory and those that are not national legislation of each state, whilst improving the professional capacity of the tax administration to control transfer pricing. Purpose - The importance of properly calculating transfer pricing is emphasized, which depends on multiple parameters, both those defined in theory and those that are not, but should be. International guidelines recommend several methods for calculating transfer pricing, and which one should be given priority if two of them are equally acceptable? If they are TNM and Cost Plus, the latter takes precedence. The purpose of the paper is to show that in practice this may not be the case, which is the basic hypothesis on which it is based. Research design/method/approach - The paper first considers the existing international methodology for calculating transfer prices and then the issue of calculation in practice in the case of minimal deviation of two significant components of comparability: sample size and time dimension of parameters. All this will be tested on the example of a multinational company from one of the developed European countries, which has a related party in one of the developing countries in Europe. The test party is a related entity of a taxpayer in a developing country. Several calculation scenarios are given for case of the aforementioned changes, which give different tax amounts. If a developing country accepts (by including in national regulations) an international recommendation on the advantage of the Cost Plus method, it will have accepted that in practice it will in some cases charge a smaller tax. Findings - The hypothesis was confirmed that the Cost Plus method should not always take precedence over the TNM method. A more detailed analysis of the different scenarios for the particular circumstances of the case gives wider opportunities to properly determine the basis for taxation but also to prevent the country's tax losses. An argument is also given for defining national regulations in more detail to avoid unwanted occurrences. Practical implication - It is necessary to amend international guidelines and national tax regulations, as well as to increase the professional capacity of the tax administration in this area. This will increase the control of transfer pricing, as well as the "extraction" of profits from the country to other tax jurisdictions Originality/Value - Theory and practice need to be more connected. Practical examples show that some changes need to be made in this area to more effectively prevent tax evasion. There are many objections to the existing concept of transfer pricing in literature, but no similar practical examples and studies are available to confirm deviations in the calculation as outlined in this paper.
