There’s a lot of buzzes these days around “transfer pricing value chains” and one of the most common questions we get is how to analyze them.
In this article, we’ll walk you through a five-step process that will help you to identify the appropriate stakeholder relationships, understand the firm-level activities, manage data requirements, and accurately compute transfer pricing values. We’ll be using an example you can download in the Resource section that involves a US-based, US-listed multinational company with global operations selling products and services to unrelated parties across dozens of markets around the world. The company is considering a sale of its product ownership (i.e., intellectual property) to one of its affiliate manufacturing companies in Shanghai. As the new owner, Shanghai would have the right to sell this product line to unrelated third parties anywhere in the world at whatever price they wish.
Before we kick-off, let’s discuss a few things about transfer pricing value chains that you should consider when carrying out your analysis.
The segments can be organized into two dimensions:
- company level and
- product level
Figure 1 shows a simple example in which the same transfer pricing value chain exists at both levels. In this scenario, we know that the same product exists at both levels and that each segment generates sales; however, there will be differences in assets, liabilities, and profit components across the two segments.
To analyze these differences at the company level we’ll use a company-level model with accounting attributes for each segment by Xpeer. We’ll also create an organizational chart for each segment to help us to understand how these assets are divided between parties within the organization. In addition, an analysis of the interrelationship between these segments (or countries) with respect to each accounting attribute can also be useful. For example, one of the key attributes is likely to be a common parent or controlling party; this common parent can be thought of as a value chain hub.
On the product level, we’ll want to create our own company-level model and corresponding organizational chart, but we won’t need a detailed list of assets or liabilities across these different segments. Instead, we’ll focus on what sales we expect to see from each segment separately (i.e., sales in China and sales outside China).
How To Create A Value Chain Analysis:
First, let’s get started with the company-level analysis.
We’ll start by defining our organization of segments as shown in figure 2. This time, we have a more complex set of company-level segments, so we’ll break out the information for each segment separately and create a separate “enterprise model” for each segment.
Organization of Segments
We can start to open our models by creating attributes for each segment. We’ll need to have these attributes appear on the first page of our “enterprise” model so that we can link these attributes back to our segment and/or product models:
Attributes
Once we have a list of attributes from our segments, we can begin to open additional models.
Open Sub-Models
We’ll repeat this process until we have all of our company-level attributes linked to their corresponding segment and/or product level models. Now that we have these segments open, we can start to load the data into our models. We’ll start with the sales data:
Load Data Into Company-Level Model
Now is also a good time to begin thinking about how complex or simple your analyses may be. If you decide to extend your analysis to include product insurance values, for example, it could be a matter of discovering some of the relationships between these different segments. Or if your analysis involves relying on public financial statements of both companies, make sure that the data you’re using is clean enough that these statements are comparable.
With data loaded into our company-level model, we can now go back and link back to our segment and/or product-level models. We should note that in figure 2 there is a reference from segment A to its segment B, but not vice versa. There are two primary reasons for this:
Unidirectional Link
First, segment A is the reporting entity for this company and, therefore, it reports the revenue of segment B as well. So revenue from segment B is included in the revenue of segment A.
Second, we could conceivably create another set of segments that represent a value chain between segments A’ and B’, but within this particular analysis segment A would not report anything to segment A’. In other words, we could imagine a value chain in which sales occur across multiple segments; however, it’s easier to model one direction of the value chain at a time since transaction prices are seldom ever passed back across an existing value chain relationship.
