Best Practices for Transfer Pricing

During this Q&A, Jim Sarrail, Senior Director of Credit Risk Solutions at Moody’s Analytics, talks with Srini Lalapet, Transfer Pricing Specialist at Dell Technologies. Srini shares his transfer pricing experience, specifically the benefits of using Moody’s Analytics RiskCalcTM software within Dell’s transfer pricing process.


Jim Sarrail: To provide some structure and some specific uses of RiskCalc software at Dell, perhaps you could start with the objectives of your transfer pricing functions in terms of establishing or evaluating arm’s length financial transactions and how this software supports that business objective. Then perhaps you can discuss the regulatory environment and some other considerations.


Srini Lalapet: We have a global footprint, so we have a transfer pricing group that looks at all our entities across the world. We use RiskCalc software and it is critical for all our transfer pricing transactions. Our whole objective is to emulate the market best practice in pricing intercompany loans within our group. We approach intercompany financing the way a third-party bank would when it is providing a loan to a prospective borrower. The first part of the process involves assessing the credit rating or the credit quality of the borrower, and then the next part of the process involves the loan pricing process. If you think about the credit quality of a borrower, think of the important determinants of interest rates. There are other determinants as well, but think about all the factors that are involved in pricing a loan. That is probably the key challenge, and this software provides an efficient mechanism to assess the credit quality of the borrower.

Now you have an established credit rating from a leading credit rating agency. The interesting thing about this software is that, because it has country-specific models, we can use it to come up with the credit ratings for all our group loans wherever they are situated. By doing that, we are able to have a systematic way to compare credit ratings and credit quality of borrowers across the world, after solving the problem of uniformity of assessment of credit quality of borrowers across the world. After solving the problem of uniformity of assessment of credit quality as well the comparability of the credit quality of our borrowers, we look at the various functionalities that RiskCalc software provides.

For instance, we use both the financial statement only and credit cycle adjusted modes to look at Expected Default Frequency (EDF). There might not be much difference between these two modes, the credit cycle does not normally affect the credit rating as much unless there is an extraordinary financial situation like 2008.



"We have a global footprint, so we have a transfer pricing group that looks at all our entities across the world."

Srini Lalapet, Transfer Pricing Specialist, Dell Technologies

The other interesting thing is that this software provides us both qualitative and quantitative assessment metrics. We do occasionally use qualitative data but we do not use that as a standard because using quantitative data is easily defendable in front of tax authorities. Qualitative factors are less defendable so we try to emphasize transparency to the extent possible because these assessments ultimately can buy you tax security and so, if you think about the process, RiskCalc software is fairly critical to take on the assessment. The other way we would think about factors that would affect the standard of rating of the borrower would be the parent’s credit rating. We do look at that and see if for instance there is any kind of inclusive or explicit support provided by the parent and if we should be taking that into consideration in assessing the credit quality of the individual borrower. This is something that we have discussed extensively internally within our group and with external advisors. We set up a policy that we use a parent company’s credit rating because the parent company has full access to a subsidiary’s cash and assets and therefore we generally look to the parent company as a ceiling unless we have special circumstances.


Jim Sarrail: In the absence of our software, what would be your approach to estimate the credit rating?


Srini Lalapet: In my experience, some companies have in-house models. So the way we would build a model is to follow the Moody’s industry position papers on this particular subject. For example, if we were evaluating credit ratings of companies in the technology industry, we would refer to the your industry methodology and try to build an in-house model from scratch. Now that is a time consuming and probably also expensive process in terms of the amount of effort involved. One of the challenges with having an in-house model is that you must explain the in-house model and the more sophisticated it is, the more difficult it is to explain. The other alternative is to get one of the academic models and you need a high level of expertise to build and work with these academic models. In commercial operations such as ours, we do not have the luxury, the time, or the resources to build in-house models and more importantly we are a little worried about building these models and trying to explain them to the tax authorities. Some tax authorities are sophisticated and perhaps they understand and some tax authorities might not. Also, they might feel that perhaps we are doing something with these models that is not entirely transparent. I think RiskCalc software is a good tool because we tell the tax authorities that we are using this tool that is provided to us by Moody’s, therefore this is an unbiased evaluation of our credit ratings.