Where the application of the most appropriate method (or, in relevant circumstances, of more than one method, see paragraph 2.12), produces a range of figures, a substantial deviation among points in that range may indicate that the data used in establishing some of the points may not be as reliable as the data used to establish the other points in the range or that the deviation may result from features of the comparable data that require adjustments. In such cases, further analysis of those points may be necessary to evaluate their suitability for inclusion in any arm’s length range ...

A range of figures may also result when more than one method is applied to evaluate a controlled transaction. For example, two methods that attain similar degrees of comparability may be used to evaluate the arm’s length character of a controlled transaction. Each method may produce an outcome or a range of outcomes that differs from the other because of differences in the nature of the methods and the data, relevant to the application of a particular method, used. Nevertheless, each separate range potentially could be used to define an acceptable range of arm’s length figures. Data from these ranges could be useful for purposes of more accurately defining the arm’s length range, for example when the ranges overlap, or for reconsidering the accuracy of the methods used when the ranges do not overlap. No general rule may be stated with respect to the use of ranges derived from the application of multiple methods because the conclusions to be drawn from their use will depend on the relative reliability of the methods employed to determine the ranges and the quality of the information used in applying the different methods ...

The arm’s length principle does not require the application of more than one method for a given transaction (or set of transactions that are appropriately aggregated following the standard described at paragraph 3.9), and in fact undue reliance on such an approach could create a significant burden for taxpayers. Thus, these Guidelines do not require either the tax examiner or taxpayer to perform analyses under more than one method. While in some cases the selection of a method may not be straightforward and more than one method may be initially considered, generally it will be possible to select one method that is apt to provide the best estimation of an arm’s length price. However, for difficult cases, where no one approach is conclusive, a flexible approach would allow the evidence of various methods to be used in conjunction. In such cases, an attempt should be made to reach a conclusion consistent with the arm’s length principle that is satisfactory from a practical viewpoint to all the parties involved, taking into account the facts and circumstances of the case, the mix of evidence available, and the relative reliability of the various methods under consideration. See paragraphs 3.58–3.59 for a discussion of cases where a range of figures results from the use of more than one method ...

A range of figures may also result when more than one method is applied to evaluate a controlled transaction. For example, two methods that attain similar degrees of comparability may be used to evaluate the arm’s length character of a controlled transaction. Each method may produce an outcome or a range of outcomes that differs from the other because of differences in the nature of the methods and the data, relevant to the application of a particular method, used. Nevertheless, each separate range potentially could be used to define an acceptable range of arm’s length figures. Data from these ranges could be useful for purposes of more accurately defining the arm’s length range, for example when the ranges overlap, or for reconsidering the accuracy of the methods used when the ranges do not overlap. No general rule may be stated with respect to the use of ranges derived from the application of multiple methods because the conclusions to be drawn from their use will depend on the relative reliability of the methods employed to determine the ranges and the quality of the information used in applying the different methods ...